1 Motor accidents happen every day, and when they do, very often people can get seriously injured. Where motorists are found to have driven negligently, courts will not hesitate to impose the full brunt of criminal and civil sanctions on them. However negligent the tortfeasor may be, the law requires the extent of the victim’s right to recover damages from the tortfeasor to be modulated by the doctrine of contributory negligence. In determining the applicability of the doctrine of contributory negligence, the law requires the court to focus on the victim and ascertain whether it was within the victim’s reasonable means to prevent himself from being injured by the tortfeasor. In this connection, a victim’s entitlement to damages could and should be reduced if the danger posed by the accident, against which he failed to take reasonable precaution, was within the reasonable contemplation of the victim. This is because the omission of his contributed to the injury he suffered.

2 The present case involved an accident between a motor vehicle and a pedestrian that took place along a pedestrian crossing controlled by traffic lights. The pedestrian claimed that he was knocked down when the pedestrian signal was in his favour. The driver of the vehicle did not contest this and rightfully conceded on primary liability. However, the driver advances the partial defence of contributory negligence on the ground that the pedestrian could have avoided the accident by checking for approaching traffic before stepping out onto the second half of the crossing situated along a dual-carriageway. This unremarkable, straightforward set of facts generated several interesting and important questions for this court to consider, in particular, whether a pedestrian owes himself a responsibility to take care of his own safety when using a signalised pedestrian crossing and if so, the extent of such a responsibility.

3 The High Court judge (“the Judge”) below found that the pedestrian, Li Jianlin (“the Respondent”) was not contributorily negligent for failing to lookout for errant drivers at a signalised pedestrian crossing. The present appeal is brought by the driver of the vehicle, Asnah Binti Ab Rahman (“the Appellant”), against that decision of the Judge. At the end of the hearing, we reserved judgment to consider the matter more carefully given the public importance it assumes.

4 At the time of the accident, the Respondent was a 21-year-old male National Serviceman. The Appellant was a taxi-driver. On 2 June 2011, the day of the accident, the Respondent booked out of camp at about 9pm. At 10pm, he approached a traffic-light controlled pedestrian crossing located along Bukit Batok West Avenue 5. The crossing was between Blocks 526 and 374 of that road. The Respondent approached the crossing from the pavement located at Block 526 and was making his way across towards Block 374.

5 Bukit Batok West Avenue 5 is a dual-carriageway road, with two lanes of traffic going in the direction of Bukit Batok West Avenue 2 and another two lanes going in the opposite direction of Brickland Road. The Appellant was travelling in the direction of Brickland Road. The road on which the Appellant was travelling curves to the left about 150m before the pedestrian crossing before straightening itself about 60–70m from the pedestrian crossing. ¹ The two carriageways are divided by a 1.4m high metal barricade erected on a raised concrete base. The barricade lowers to 0.9m about 18m away from the pedestrian crossing. The Respondent made it across the first half of the crossing without incident. Just as he took his second or third step into the second half of the crossing, the Appellant’s taxi crashed into him. It was only after the collision that the Appellant realised that the traffic signal was not in her favour.

6 The point of impact appeared to have been close to the midway point between the front headlights of the taxi. The Appellant was travelling at about 55km/h towards the pedestrian crossing. As a result of the collision, the Appellant’s taxi windscreen was smashed. The momentum of the taxi projected the Respondent forward towards the left side of the taxi, landing him by the roadside.

7 At the time of the accident, the weather was fine, the road surface was dry, the traffic flow was light and the visibility was clear.

8 The Respondent suffered severe injuries to his head and hips as a result of the collision. He was hospitalised for close to three months. The concussion he suffered impaired his ability to recall the details of the incident except for the fact that he booked out of camp at about 9pm that night. A National University Hospital psychologist report was tendered in support of this.

9 At trial, only two questions were posed to the Respondent. The first was whether he remembered what clothes he wore on the night of the accident. The second was whether he remembered the colour of the clothes he wore. He answered both questions in the negative.

10 The Appellant concedes on primary liability, admitting that she failed to notice the red traffic signal when her vehicle was approaching the crossing. She was convicted of and sentenced for dangerous driving on 11 October 2013. She, however, takes issue with the Respondent’s conduct, arguing that the Respondent could have avoided the accident if he had checked for approaching traffic. She says that notwithstanding that the pedestrian lights were in his favour, the Respondent should still have in the circumstances, made sure that it was safe to cross the second half of the dual-carriageway.

11 The trial below was bifurcated and the Judge decided only the question of liability. The Judge found in favour of the Respondent, holding that he was not contributorily negligent for the following reasons:

(a) There was no evidence whether the Respondent had looked left or right before crossing.

(b) The Respondent was hit after he had crossed more than half the breadth of the road with the pedestrian signal in his favour. Two implications arose from this situation. First, the Respondent could hardly be blamed for assuming that the vehicular traffic had already stopped, and those that have not would surely have done so. Secondly, since the traffic lights of the pedestrian-crossing had been red against vehicles in both directions of Bukit Batok West Avenue 5 for so long, the Appellant had no excuse for not having enough time to react.

(c) This is not a case in which the Respondent was jaywalking or crossing at a place where he ought to have a heightened sense of caution.

12 The Appellant argues that any prudent person crossing the road should keep a proper lookout to ensure his own safety, regardless of whether he is crossing the road with the aid of traffic lights or a zebra crossing. The act of checking for approaching traffic is a matter of safeguarding one’s own safety as it is reasonably foreseeable that there may be drivers flouting traffic rules, whether negligently or intentionally. The duty to do so is embodied in various provisions of the Highway Code (Cap 276, R 11, S8/75) (“the Highway Code”).

13 The Respondent obviously failed to check for approaching traffic before crossing the second half of the road. The Appellant’s taxi was approaching the crossing at 55 km/h. Had the Respondent looked to his left, he would have realised the Appellant’s taxi was unlikely to be able to stop in time.

14 The Respondent should be held liable for contributory negligence and the damages recoverable should be reduced by 35% to reflect his blameworthiness.

15 The Respondent argues that pedestrians have no duty to continuously look left and right when the signal lights are already in their favour. Furthermore, the Respondent was only knocked down after he had crossed more than half the breadth of the road (ie, the first half of the road and an additional two to three steps of the second half).

16 The Respondent argues that even if there were such a duty, the Appellant’s case should be disallowed on the basis that the Appellant failed to put her case to the Respondent when he was in the witness box, and consequently, the Respondent was not given a chance to respond to the Appellant’s case.

17 The main issue in this appeal is whether the Respondent was contributorily negligent: did the Respondent have a duty to guard himself against the Appellant’s wrongful driving, and if so, did he discharge that duty of his with due care and diligence. If the Respondent is found to have been contributorily negligent, the issue of apportionment arises: to what extent should damages due to him be reduced on account of his contributory negligence?

18  In the introduction to our judgment, we noted that a victim’s right to recover damages from a tortfeasor has to be modulated by the extent to which he could himself have prevented the accident from happening. The doctrine of contributory negligence gives effect to this position. Even though the defendant is found to have been negligent, contributory negligence provides him a partial defence by reducing the quantum of damages payable to a claimant where the claimant failed to take due care for his own safety and thus caused loss to himself. In determining contributory negligence, one looks solely at the conduct of the claimant in the prevailing circumstances of each case (Charlesworth & Percy on Negligence (Sweet & Maxwell, 13th Ed, 2014) (“Charlesworth & Percy on Negligence”) at para 4-03). Contributory negligence connotes a failure by the claimant to take reasonable care for his own personal safety in all the circumstances prevailing at the time of the accident, such that he is blameworthy to the extent that he contributed to his own injury. A person is guilty of contributory negligence if he ought to have objectively foreseen that his failure to act prudently could result in hurting himself and failed to take reasonable measures to guard against that foreseeable harm (Cheong Ghim Fah and another v Murugian s/o Rangasamy [2004] 1 SLR(R) 628 (“Cheong Ghim Fah”) at [83]). The above discussion is neatly borne out by Denning LJ’s (as he then was) dictum in Jones v Livox Quarries Ld [1952] 2 QB 608 (“Jones v Livox”) at 615:

19 While the literal meaning of “contributory negligence” may suggest that the claimant owes some form of duty to the defendant, there is no such requirement in law. A finding of contributory negligence is not premised on a breach of some duty owed to the defendant. What is, however, required of the defendant is that he shows that the claimant owes himself a duty to take care of his own safety in the prevailing circumstances of the case. In this regard, we refer to the following clarifications Lord Simons made in Nance v British Columbia Electric Railway Co Ltd [1951] AC 601 at 611:

20 As a general rule, the standard of care expected of the claimant is measured against a person of ordinary prudence, corresponding in most cases to the standard of care in negligence (Sungravure Pty Ltd v Meani (1964) 110 CLR 24 at 36–37; see also A C Billings & Sons Ltd v Riden [1958] 1 AC 240 at 252; see also generally Glanville Williams, Joint Torts and Contributory Negligence (Stevens & Sons Ltd, 1951) at p 353).

21 In Singapore, the defence is statutorily enacted in the Contributory Negligence and Personal Injuries Act (Cap 54, 2002 Rev Ed) (“Contributory Negligence Act”). The relevant provision in the Contributory Negligence Act is s 3(1) which reads:

22 Section 3(1) of the Contributory Negligence Act was adopted from s 1(1) of the English Law Reform (Contributory Negligence) Act 1945. Both these provisions were enacted to correct the perceived injustice inherent in the common law doctrine of contributory negligence. Prior to legislative intervention, courts were not allowed to apportion damages between the claimant and the defendant. Butterfield v Forrester (1809) 11 East 60 was often cited as a case manifesting such injustice, as it represented the common law rule that where a claimant’s injury was caused in part by his own fault, he was wholly disentitled from claiming damages from the negligent defendant. The passing of the apportionment legislation assisted claimants to recover part of their losses even where they were contributorily negligent. With contributory negligence now being a partial defence, courts are now empowered to apportion the damages recoverable if the injury was partly due to the claimant’s own fault. As would be apparent from the provision, fault can arise not only from a positive act but also from an omission on the part of the claimant. In the context of the present case, the “omission” in question is the Respondent’s alleged failure to keep a proper lookout for the Appellant’s vehicle before he stepped onto the second half of the crossing.

23 As mentioned previously, the collision took place not in the first half but rather in the second half of the crossing. This factual premise presents three main hurdles for the Appellant to overcome before she can successfully invoke contributory negligence to reduce the quantum of damages payable. First, she must persuade the court that a pedestrian owes himself a responsibility to check for oncoming traffic before entering a signalised pedestrian crossing even when the lights are in his favour. Second, she has to establish that the Respondent had in the circumstances of this case a duty to keep a proper lookout before stepping onto the second half of the road. Third, she has to prove that the Respondent had in fact failed to keep a proper lookout before stepping onto the second half of the road. We now proceed to examine each of these three questions in turn.

24 Central to the concept of contributory negligence is the notion that the claimant ought to have taken reasonable precautions against the risk of harm. But a claimant is not expected to take precautions against all risks – he needs only to guard himself against forms of injury that might reasonably have been foreseen and avoided (Jones v Livox at 615). As Lord du Parcq puts it, a prudent man will guard against the possible negligence of others when experience shows such negligence to be common (Grant v Sun Shipping Co Ltd [1948] AC 549 at 567). In the context of road traffic accidents, this principle may be taken to mean that the user of the road is not bound to guard against every conceivable eventuality, but only against such eventualities that a reasonable man ought to foresee as being within the ordinary range of human experiences (Fardon v Harcourt-Rivington (1932) 146 LT 391 at 392).

25 The issue before us is whether the pedestrian’s right of way within a signalised crossing entitles him to also assume that all motorists would obey the law and drive in a reasonable manner. To answer this question, we first consider whether the risk of vehicles running red lights (whatever may be the cause for the lapse, eg, fatigue, day-dreaming or some other form of distraction) is a remote possibility that would not have occurred to the mind of a reasonable pedestrian, or whether the possibility of such danger emerging is a reasonably apparent one such that the pedestrian ought to safeguard himself.

26 Traffic signals and pedestrian crossings are the mainstay of the road transport infrastructure in any modern society. The impetus for its invention in the last century came about as part of a slew of road safety measures implemented by governments in response to the rapid growth of vehicular traffic. ² Given the obvious risks posed by fast moving vehicular traffic, traffic signals and pedestrian crossings together formed a compromise of sorts between pedestrians’ and motorists’ respective interests in the utilisation of road spaces. Drivers are not only expected but required to stop when traffic signals turn amber or red and to give way to pedestrians, whilst pedestrians are not allowed to utilise road spaces outside of designated pedestrian crossings. These measures, amongst other traffic control regulations, infrastructure and campaigns serve the purpose of better ensuring that pedestrians and drivers would occupy and utilise road spaces in an organised, efficient and safe manner.

27 In Singapore, traffic rules are strictly enforced and violations often come with hefty penalties. A motorist caught for failing to conform to traffic signals can expect to receive 12 demerit points under the Driver Improvement Points System. This is half of the maximum number of demerit points one is allowed to accumulate in 24 months before one’s driving license becomes liable for suspension. ³ Those who drive under the influence of alcohol, fall asleep behind the wheel, and drive above designated speed limits often pose significant danger not only to pedestrians but also to fellow motorists. Correspondingly, the law imposes severe punishments to deter motorists from driving irresponsibly.

28 Today, traffic signals and pedestrian crossings are extensively deployed across Singapore. They can be found in almost every corner of our complex and comprehensive road transport network. As the vehicle population in Singapore continues to grow, the land transport authorities have been making consistent efforts to deploy better pedestrian crossings, implement road safety measures and step up its enforcement of traffic rules to better safeguard pedestrians’ safety. Together with the strong deterrent effect that our stringent traffic rules and regulations impose on all road users, these measures have helped reduce the number of accidents on our roads over the years. Indeed, perhaps one indicator of the success we have achieved thus far is the 13% reduction in the number of fatal accidents since 2009 (Singapore Parliamentary Debates, Official Report (8 September 2014) vol 92 (Masagos Zulkifli B M M, Senior Minister of State for Home Affairs)).

29 We also acknowledge that the vast majority of Singapore motorists can be said to be by and large law abiding. It is a common sight to see drivers stop at junctions and pedestrian crossings when the lights turn against them. Motorists frequently slow down and give way to pedestrians at designated uncontrolled pedestrian crossings. And even in the face of jaywalkers, many motorists do slow down to avoid accidents. These are all good road safety habits that ought to be encouraged and fostered.

30 But the success we have achieved thus far in promoting road safety is not sufficient cause for pedestrians to be lulled into a sense of complacency when they utilise pedestrian crossings. It is difficult to ignore the latent dangers that lie within our roads. Fast moving vehicles are potentially dangerous and capable of delivering fatal blows to the unguarded pedestrian. Despite the government’s well-placed efforts to improve road safety conditions, one must also recognise that humans are not infallible and do suffer from occasional lapses. Accidents inevitably occur. The manner in which accidents manifest themselves as well as their root causes are multi-faceted, and in the present case, the type of accidents we are concerned with stems from motorists running red lights. Numerous causes underlie such accidents. They could result from lapses of attention on the part of motorists due to fatigue or distractions, eg, from crying or quarrelling children in the rear seats; motorists seeking to beat an amber light that is turning red but fail to do so in good time; motorists driving under the influence of alcohol; motorists inadvertently falling asleep behind the wheel; and even motorists who deliberately run the red light, along a quiet road at night, thinking that no pedestrian or vehicle would come in his way. The problem is compounded in recent years by the proliferation of smartphones, posing an additional source of distraction to our drivers. This is borne out by the following remarks made by Mr Hri Kumar Nair during the parliamentary debates at the second reading of the Road Traffic (Amendment) Bill (Singapore Parliamentary Debates, Official Report (8 September 2014) vol 92):

31 The different forms of lapses highlighted above can conceivably result in motorists running red lights. Our understanding that motorists do occasionally run red lights is grounded not only on common experience but is also borne out by statistics. Between 2004 and 2008, about 2,100 accidents occurred annually at road intersections of which about 330 or 16% of them occurred because the motorist disregarded a red light signal. During the same period, the Traffic Police issued about 19,700 summonses annually to motorists for flouting a red light signal (Singapore Parliamentary Debates, Official Report (14 September 2009) vol 86 at cols 1423–1424 (Wong Kan Seng, Deputy Prime Minister and Minister for Home Affairs)).

32 In 2013, 5.3% of the 150 fatal accidents involved vehicles running the red light. In other words, about eight people were killed that year by errant drivers running the red light (Traffic Police News Release, Annual Road Traffic Situation 2014 http://driving-in-singapore.spf.gov.sg/services/driving_in_singapore/documents/Annual_Road_Traffic_Stats.pdf> (last accessed: 9 March 2016)). Whilst the number of fatal accidents in those circumstances dropped to four in 2014, the reduction in the number of fatalities may not be entirely representative of the frequency drivers have been running red lights or how egregiously it was done. In fact, statistics appear to suggest otherwise. In 2013, 18,796 motorists were caught for the offence. In 2014, that figure jumped to 25,833. ⁴ And even though these figures may already seem high, it becomes apparent that they reveal only the tip of the iceberg when one puts them in proper perspective – they do not actually represent all incidents of motorists running the red lights across the island for the simple reason that only a small fraction of all traffic junctions in Singapore are monitored by red light cameras. As was reported in The Straits Times on 16 July 2015, amongst the numerous traffic junctions deployed across the city, only 240 of them are monitored by red-light cameras. ⁵

33 We make two more points about the above statistics. First, it is true that the statistics above are not sufficiently granular to differentiate between motorists trying to squeeze past amber lights but failing to do so in time resulting in them beating red lights within the first one or two seconds after the lights have turned red against them and those running red lights after a number of seconds have elapsed. That does not necessarily mean we should ignore the statistics altogether in that they are totally irrelevant. The court should make the best of whatever relevant evidence is available in each given case, even if the evidence lacks a certain measure of specificity in relation to the issue that is to be adjudicated, as is often the situation in many cases. And while we further acknowledge that intuitively speaking the former scenario is probably more common than the latter, these statistics nevertheless suggest that the latter scenario is within the bounds of real possibilities in the absence of evidence suggesting the contrary. In our judgment, it is conceivable that a proportion of the errant motorists captured by the statistics were those that run the red lights after the light had turned red against them for more than a few seconds.

34 Second, we are cognisant of the fact that amongst the many summonses issued to motorists who disobey the red lights, only a small fraction resulted in accidents. But in our judgment, the low frequency of accidents eventuating from motorists paying no heed to traffic signals should not detract from the fact that there is a high incidence of motorists disobeying red lights. The mismatch between the two statistics is explicable with reference to a number of plausible scenarios, such as there being no pedestrians or other road users coming in the pathway of the errant motorist or other road users having taken care of their own safety to avoid getting hit by the errant motorist. The point to be gathered from these statistics is not so much that accidents have in fact happened, but that road users should be prepared to encounter – albeit occasionally – a motorist who disobeys the red light, for whatever reason. That accidents of this sort are not to be regarded as far-fetched and that they can happen is illustrated by a letter written by a member of the public to The Straits Times recently on 30 August 2015, in which he shared his experience of almost being hit by a driver charging past a signalised pedestrian crossing during which the lights were in the pedestrian’s favour (http://www.straitstimes.com/forum/letters-on-the-web/install-red-light-cameras-at-danger-junctions>):

35 The letter suggests that the author, along with other pedestrians around him, might have taken some care to avoid being hit by the errant driver described. Had the pedestrians, including the author, relied exclusively on the pedestrian signal, the result would have been far worse than a mere forum post to The Straits Times. This highlights the seriousness of such incidents, and correspondingly, pedestrians should not lightly brush them aside. We point out two accidents that took place along signalised pedestrian crossings by reason of motorists running red lights that were recently reported in the newspapers to underscore the significance of this risk:

(a) Article published in The Straits Times on 6 November 2015 titled “Hit-and-run lorry driver jailed and banned” (<http://www.straitstimes.com/singapore/courts-crime/hit-and-run-lorry-driver-jailed-and-banned#xtor=CS1-10>).

(b) Article published in The Straits Times on 24 November 2015 titled “Lorry driver jailed over fatal crash and commercial sex with girl, 17” (http://www.straitstimes.com/singapore/courts-crime/lorry-driver-jailed-over-fatal-crash-and-commercial-sex-with-girl-17>).

36 We acknowledge that pedestrians have a statutory right of way over motorists when they are moving across a pedestrian crossing. The right that a pedestrian has over motorists is encapsulated in various provisions of the Road Traffic (Pedestrian Crossing) Rules (Cap 276, R 24, 1990 Rev Ed) which we reproduce:

37 That said, it is trite that the fact that a victim’s rights have been violated does not preclude the tortfeasor from raising the defence of contributory negligence. Otherwise there would be no room for the defence of contributory negligence to operate in many cases where the main tortfeasor has breached a duty that has the aim of protecting the claimant from being harmed. It has been observed that “[p]edestrian’s crossings were instituted … with the object of providing at reasonable intervals along the highway areas where pedestrians may cross in safety, and without having to pay undue attention to motor vehicles” [emphasis added] (“Negligence at Pedestrian Crossings” (1938–1939) Mod L Rev 233 at p 239). In our judgment, this suggests that having to pay some attention at pedestrian crossings would not be inconsistent with the objective behind its deployment. We say “some attention” is required having regard to the fact that pedestrian crossings are not in and of themselves fool-proof safety measures – they mitigate road dangers rather than eliminate them altogether.

38 In similar terms, the Ontario Traffic Manual describes pedestrian crossings as markings which “serve to reduce the potential for conflicts with motor vehicles” by delineating pathways for pedestrians to cross the road http://www.cedarsignsinc.com/usercontent/documents/Book%2015%20-%20Pedestrian%20Crossing%20Facilities.pdf> at p 21 [emphasis added]). We emphasise the word “reduce” because it is apparent from this that pedestrian crossings are not meant to lull pedestrians into a false sense of security thinking that no mishap could possibly occur simply by virtue of them using a signalised pedestrian crossing, but rather that they are installed to enhance the safety of pedestrians using the road. The purpose of pedestrian crossings must thus be understood in the proper context.

39 Tort law exacts a degree of care commensurate with the danger the situation presents (Read v J Lyons & Co Ltd [1947] AC 156 at 173). Danger comprises two components: the court must consider both the probability of the risk eventuating and the gravity of the potential injury itself. It has therefore been widely recognised that, from an economic perspective, the degree of care demanded of a person by an occasion is dependent upon a consideration of three key factors: the likelihood of injury; the gravity of the injury and the cost of taking precautions to guard against such risks. Liability is generally imposed when the cost of taking precautions is less than the likelihood of injury multiplied by the gravity of the injury (United States v Carroll Towing Co (1947) 159 F 2d 169 C A 2, per Learned Hand J; Bolton v Stone [1951] AC 805 at 867, per Lord Reid; Sutherland v Hatton [2002] P I Q R. 241 at [32] per Hale LJ; see also Guido Calabresi, The Costs of Accidents: A Legal and Economic Analysis (Yale University Press, 1970)). This economic analysis of the law is widely recognised as a rule that sets the appropriate standard of care for the avoidance of accidents (Simon Deakin, Angus Johnston & Basil Markesinis, Markesinis and Deakin’s Tort Law (Oxford University Press, 7th ed, 2013) at 199), and is used by leading textbooks to analyse whether a particular conduct falls below the requisite standard of care (see for instance, Charlesworth & Percy on Negligence at paras 7-01 to 7-39; R P Balkin, Law of Torts (LexisNexis Butterworths, 2009, 4th ed) at pp 258-264; Francis Trindade, Peter Cane and Mark Lunney, The law of torts in Australia, (Oxford University Press, 4th ed, 2007) at pp 427-436). While it offers a good starting point for our analysis, we should also caveat that these three factors may not necessarily be exhaustive in every situation (in this regard, see generally Allen M Linden and Bruce Feldthusen, Canadian Tort Law (LexisNexis Butterworths, 8th ed, 2006) at p 131). Ultimately, the court balances these three factors with reference to the facts of each given case to determine the appropriate degree of care to be imposed.

40 Viewed from this perspective, a remote probability of harm may not excuse conduct or omissions if serious injury is a possibility (Beckett v Newalls Insulation Co Ltd and another [1953] 1 All ER 250 at 255 per Singleton LJ; see also Paris v Stepney Borough Council [1951] AC 367). On the other side of the equation, the ease with which the risk can be avoided is also a relevant consideration. As Denning LJ stated in Latimer v A E C Ltd [1952] 2 QB 701 at 711, “[i]n every case of foreseeable risk, it is a matter of balancing the risk against the measures necessary to eliminate it”. Thus, where small risks can easily be avoided, precautions may nevertheless be necessary. Similar remarks were made by the Privy Council in the seminal decision of Overseas Tankship (UK) Ltd v The Miller Steamship Co Pty Ltd and another (The Wagon Mound (No 2)) [1967] 1 AC 617 at 643–644: a reasonable man would not ignore a small risk “if action to eliminate it presented no difficulty, involved no disadvantage and required no expense.”

41 Against this backdrop of legal principles, it is true that the risk of pedestrians encountering road accidents is greatly attenuated by the usage of signalised pedestrian crossings. But as the statistics and our common experience demonstrate, pedestrians should not take the greater measure of safety provided by pedestrian crossings for granted, and should continue to exercise some care for their own safety. Even though the chances of encountering errant drivers at a pedestrian crossing are comparatively smaller than those at roads uncontrolled by traffic lights, pedestrians should not turn a blind eye to them altogether given the potentially severe injuries road accidents are capable of inflicting upon its victims. At risk of stating the obvious, road collisions often inflict serious, and at times fatal, injuries to its victims. When life and death or severe personal injury is at stake, a prudent pedestrian should be mindful that motorists do not invariably exercise due care (whatever may be the cause for the momentary lapses on their part). The dangers we have shown above necessitates the pedestrian to exercise at least some caution in approaching crossings (even when the traffic lights are in his favour). We turn now to examine the appropriate standard of care to be expected of pedestrians that would be commensurate with the risk of harm and the ease at which the duty can be fulfilled.

42 In keeping with the factorial approach set out at [39] above, the possibility of severe injuries being inflicted on pedestrians must be balanced against the admittedly small probability of it eventuating and the ease with which preventive measures may be taken to guard against such risks. In balancing these three factors, the extent to which preventive measures ought to be taken by a pedestrian at a signalised pedestrian crossing with the traffic lights in his favour is, in our judgment, the central inquiry in this case. In this connection, the first port of call for this inquiry should be the Highway Code. We will first discuss the significance of the Highway Code before examining the relevant standards prescribed therein.

43 The importance that the Highway Code plays in regulating the responsibilities of all road users cannot be overstated. The Highway Code is a piece of subsidiary legislation approved by Parliament after multiple governmental and non-governmental agencies were consulted. In presenting the Highway Code to Parliament for approval in 1968, the Minister of State of the Prime Minister’s Department had this to say (Singapore Parliamentary Debates, Official Report (24 January 1968) vol 26 at cols 1132–1133, (Mr Tuan Haji Yaacob Bin Mohamed, The Minister of State for the Prime Minister’s Department)):

44 When the next iteration of the Highway Code was approved by Parliament in 1975, the Minister of State for Communications, Mr Chai Chong Yii (“Minister Chai”) echoed similar observations (Singapore Parliamentary Debates, Official Report (3 March 1975) vol 34 at col 172):

45 As Minister Chai said, the Highway Code is not merely a digest of traffic laws. The government intended the general public to take the Highway Code seriously. Subsequent to the passing of the 1975 edition of the Highway Code, a nationwide “Road Safety For You” Campaign was launched in June 1977. Its aim was to remind both motorists and pedestrians of the need to observe the Highway Code (Singapore Parliamentary Debates, Official Report (8 February 1977) vol 36, at col 50). As Member of Parliament Mr J F Conceicao puts it, the Highway Code is a “very important instrument of civic education” (Singapore Parliamentary Debates, Official Report (3 March 1975) vol 34 at col 174):

46 It is evident from all of this that the Highway Code has the primary objective of reducing the incidence of road accidents. This objective of the Highway Code is enshrined in r 2, which reads: “[t]he primary cause of road accidents … is the failure of road users of all classes to behave properly in traffic” [emphasis added]. To achieve this, the Highway Code apprises all road users of standards that they ought to observe when they use the roads. Motorists, being in a position to inflict harm, are undoubtedly the primary target audience of the Highway Code. But the Highway Code also expects pedestrians to exercise care. This is evident as rr 7–28 of the Highway Code are all dedicated to pedestrian safety.

47 In this regard, V K Rajah JC’s observation in Cheong Ghim Fah is also worth reproducing in its entirety:

48 The case of Cheong Ghim Fah is useful insofar as it declares the importance of adhering to the Highway Code. A jogger in that case was knocked down by a motorcyclist whilst he was jogging along the extreme left side of the road with his back to the traffic. Rajah JC highlighted the importance of all road users – pedestrians included – in paying “real heed” to the requirements of the Highway Code. On the facts, he found that the jogger was in breach of r 7 of the Highway Code for failing to use the adjourning pavement and for this the jogger was held to be contributorily negligent.

49 In the context of the present case, the relevant provision in the Highway Code is r 22, which reads:

50 Rule 22 prescribes the standard of care expected of pedestrians when using signalised pedestrian crossings, and in our judgment, we should as far as possible give effect to this rule. Even when the lights are in the pedestrian’s favour, r 22 expressly advises pedestrians to wait for the traffic in front of them to have come to a standstill before they begin their crossing. Implicit in this rule is the recognition that the possibility of motorists running red lights is real rather than remote, and that pedestrians must be alive to the real possibilities of accidents happening even with the pedestrian signals operating in their favour.

51 Rule 22 is not to be applied in a rigid or mechanistic fashion. While r 22 advises pedestrians to wait until vehicles have come to “a standstill”, approaching vehicles may not have always come to a complete standstill when a pedestrian is ready to enter the crossing and may be in the midst of slowing down for the crossing. By the time the vehicular traffic has come to a complete standstill, there may well be insufficient time left for pedestrians to complete their journey across the road before the pedestrian signal reverts in favour of vehicular traffic. Accordingly, it would not be practical for pedestrians to adhere to the literal words of r 22. Statutes are to be interpreted in a manner that promotes the purpose or object underlying its enactment (Public Prosecutor v Low Kok Heng [2007] 4 SLR(R) 183 at [39]-[41]). The purposive interpretation is paramount and takes precedence over any other common law principles of statutory interpretation including the plain meaning rule (Dorsey James Michael v World Sport Group Pte Ltd [2013] 3 SLR 354 at [18]). The spirit of r 22 requires pedestrians to take reasonable care for their own safety against the possibility of vehicles running the red light even at signalised pedestrian crossings. Viewed from this perspective, the purpose behind r 22 would equally be achieved when pedestrians have assured themselves that approaching vehicles are slowing down towards the pedestrian crossing before they leave the footway for the crossing. Where vehicles are seen to be slowing down towards the crossing, it signals to the pedestrian that the motorist has noticed the red traffic light and is complying with it, and as such the pedestrian would be considered to have satisfied himself that it would be safe to cross the road.

52 The duty that r 22 requires of the pedestrian is not meant to be onerous. The pedestrian’s duty to satisfy himself that approaching vehicles have stopped or are stopping can easily be discharged by taking a quick glance before the pedestrian enter the crossing. This would usually take no more than a split second.

53 Can it be argued that the pedestrian’s duty extinguishes after a number of seconds have elapsed since the traffic signal has turned red against vehicular traffic? It may be argued that the risk of motorists being wholly oblivious to the traffic lights is such a remote possibility that it is not within the bounds of reasonable risks that pedestrians should guard themselves against, and to that extent, pedestrians need only guard against the supposedly more common risk of motorists beating the red light within the first few seconds of it turning red (as a result of motorists trying to squeeze past the turning light but failing to do so). In our judgment, this argument is inconsistent with the duty prescribed by r 22. Nothing in the express wording of r 22 restricts the operation of the rule exclusively to the first few seconds after the pedestrian light has turned in favour of the pedestrians. The provision suggests that pedestrians entering into a crossing should always keep a lookout for errant motorists regardless of how long the lights have turned red against the motorist and in favour of the pedestrian. He must make sure that vehicles in front of him have stopped or at least have slowed down. Parliament would have drafted the rule more restrictively if they had intended otherwise. For instance, they could have qualified the rule by adding the phrase “when the green man appears” at the start of the sentence. It is understandable why Parliament did not seek to do that as it would undermine the basic policy behind r 22, which is to reduce the incidence of road accidents by expecting all road users to always exercise care for their own safety to the extent that it is feasible (see [46] and [51] above). The effect of r 22, in its present form, is that vehicles running the red light in the first few seconds and those running it after some time has elapsed are both instances which the pedestrian should guard against.

54 Thus, even if motorists running red lights in the first few seconds of it turning red is a more common occurrence than those who do so afterwards, it does not ipso facto mean that the latter occurrence is an unforeseeable risk. Quite the contrary, r 22 tells us that the risk of motorists running red lights after several seconds have elapsed from the moment the traffic signal turned red is not only foreseeable but also reasonable for the pedestrian to guard himself against. There is good sense as to why pedestrians should be mindful of both these occurrences. In this regard, we reiterate our observations at [30] and [33] above, that motorists deliberately beating red lights by reason of them failing to squeeze past the amber lights in time cannot be the only type of motorist behaviour that results in motorists running red lights. Motorists, professionally trained as they may be, are human beings after all. They are not infallible. A pedestrian can conceivably encounter instances where motorists unintentionally or carelessly run red lights due to the more common forms of lapses identified at [30] above. To recapitulate, some of them are as follows:

(a) motorists suffering from fatigue falling asleep behind the wheel;

(b) motorists’ attention being distracted by crying or quarrelling children in the rear seats;

(c) motorists driving under the influence of alcohol; and

(d) motorists driving while being distracted by mobile devices.

In fact, there could even be instances where motorists deliberately run the red light, along quiet roads late at night, thinking that no pedestrian would come in his way. All these instances could happen not just within the first few seconds of the traffic signal turning red but also thereafter. In these circumstances, it makes eminent sense that r 22 requires the pedestrian to remain attentive and to satisfy himself that vehicles in front of him have come to a stop or are slowing down as he commences his crossing.

55 There is a further problem with the restrictive interpretation of r 22, ie, that a pedestrian no longer has a duty to keep a lookout after the red light has turned on against the vehicular traffic for at least a number of seconds. Pedestrians do not invariably attempt to cross the road immediately upon the green man appearing. Common experience tells us that many of them take advantage of a flickering green man (ie, the green man has been in favour of the pedestrian for some time). It cannot be correct to suggest that pedestrians who cross the road after some time has elapsed after the green man has lit up should not guard against errant motorists even though r 22 requires them to do so.

56 There is a further point to r 22. The rule does not distinguish between one-way roads (ie, roads that facilitate only one-way traffic), single-carriageways (ie, roads that have a single lane of traffic going in opposite directions) and dual-carriageways (ie, roads with two lanes of traffic going in each opposite direction). It would therefore apply to single and dual-carriageways with the same force as it does to one-way roads. This means that, just as a pedestrian attempting to cross a one-way road should keep a lookout for vehicular traffic, a pedestrian attempting to cross a single or dual-carriageway should similarly keep a proper lookout for vehicular traffic on each half of the road to ensure that vehicles have either stopped or at the very least are slowing down to stop at the pedestrian crossing. In the context of a dual-carriageway with a centre-divider between the two halves of the road, if the pedestrian is unable to obtain an unobstructed view of the second half of the road in order to discharge this duty of his, he should keep a proper lookout again near or at the centre-divider. We examine this point in greater detail below (at [92]).

57 In short, r 22 requires a pedestrian to always check to ensure that approaching vehicles have stopped or are coming to a stop before he enters a signalised pedestrian crossing, even when the pedestrian light is in his favour. Clearly implicit in this rule is the recognition that motorists do (whatever may be the cause) occasionally fail to observe traffic light signals, and as such pedestrians have a duty to guard against such lapses on the part of motorists.

58 A number of points follow from the duty laid down by r 22:

(a) If the pedestrian chooses to enter the crossing despite having noticed a vehicle approaching at a speed and from a distance which suggests that a collision could happen, he should take care of his own safety before stepping onto the path of that particular vehicle. This aspect of the duty is to be understood and applied commonsensically rather than in a rigid or technical fashion. It does not require the pedestrian to unabatedly divert his attention to the approaching vehicle during his journey through the crossing. The pedestrian can easily discharge the duty by casting an eye on the vehicle during his journey through the crossing until he has satisfied himself that a collision is unlikely to take place.

(b) The logical corollary to what is stated in the preceding sub-paragraph is that the pedestrian is not required to guard himself against vehicles approaching from a considerable distance because it would be a pointless exercise; the pedestrian would have completed the crossing by the time the vehicle reaches the crossing.

(c) If the pedestrian has reasonably assessed the vehicular traffic to have stopped or to be slowing down before he enters the crossing, in the general scheme of things, there is no continuing duty on him to guard against those motorists running the red light. Rule 22 does not require the pedestrian, having commenced his journey through the crossing, to continuously look out for vehicular traffic which has already stopped to ensure that such traffic would not prematurely move against the traffic signal. If a collision should happen as a result of errant motorists suddenly accelerating towards the pedestrian crossing after having already stopped or slowed down, the pedestrian cannot be blamed as such risks are unforeseeable and it would be unreasonable to expect the pedestrian to guard against such risks.

59 By requiring the pedestrian to take some care, we are not at all suggesting that motorists should be any less vigilant. The driver, having been trained to handle a dangerous machine, like a vehicle, undeniably bears a higher standard of care to prevent the occurrence of an accident. All we are saying is that a pedestrian should bear a small portion of the shared responsibility to guard against the risks of accidents if he is well-placed to do so with minimal cost and effort on his part. The duty that is required of the pedestrian under r 22 of the Highway Code is not onerous – it literally takes no more than a split second for the pedestrian to satisfy himself that it is safe to cross the road. The difference in the degree of responsibility each actor should bear is reflected and recognised in the significantly larger share of the damages that the driver should ordinarily bear in the event of an accident in such circumstances (see discussion at [119] below). In this connection, we refer to Member of Parliament Mr Lim Biow Chuan’s remarks made during the parliamentary debates at the second reading of the Road Traffic (Amendment) Bill 2014 (Singapore Parliamentary Debates, Official Report (8 September 2014) vol 92):

60 Deputy Prime Minister and Minister for Home Affairs, Mr Wong Kan Seng, made similar remarks in response to concerns raised regarding an increase in the number of vehicle and fatal accident cases (Singapore Parliamentary Debates, Official Report (28 February 2007) vol 82 at col 2015-2016):

61 In arriving at this conclusion, we also draw guidance from the public advisories issued by governmental agencies, who similarly express the view that road safety is a responsibility shared by all road users, pedestrians included. The Traffic Police issued a media release on 25 May 2014 as follows (http://www.spinet.gov.sg/mic/2014/05/20140525_others_1175_summonses.html (last accessed: 7 July 2015)):

62 Similarly, the Singapore Road Safety Council advises road users to “always be alert for inattentive drivers even at signalized crossings” (http://srsc.org.sg/advisories/children-senior-citizens-road-safety-accidents/ (last accessed: 7 July 2015)). Again, neither of these two public advisories draws a distinction between the situation of motorists running red lights as oppose to them beating red lights (see [32] and [53] above). The advice suggests that pedestrians should always keep a lookout before crossing the road regardless of how long the traffic signal has turned against vehicular traffic.

63 Our analysis thus far pertains to a pedestrian crossing controlled by traffic signals. Would a different analysis apply to one that is controlled by a police officer instead? Rule 22 prescribes the pedestrian’s duty when using “light controlled crossings”; it is however silent on the pedestrian’s duty insofar as crossings controlled by police officers are concerned. One can possibly argue that the outcome should be no different in that the pedestrian should still keep a proper lookout. An inattentive motorist who is oblivious to traffic signals by reason of, for example, fatigue or the influence of alcohol, may similarly not have noticed a police officer standing ahead signalling for him to stop. On the other hand, however, it could also be said that a different rule should apply to pedestrian crossings controlled by a police officer rather than by traffic signal insofar as a police officer is able to satisfy himself that vehicular traffic has in fact stopped before he signals to the pedestrians that it is safe to cross the road whereas a traffic signal would not be capable of communicating the same to pedestrians. This harks back to our basic premise that pedestrians cannot rely exclusively on traffic signals to tell them whether vehicles have in fact stopped for their crossing. But we express no conclusive view on this scenario since it does not concern us on the present facts.

64 We have thus far established that a pedestrian, whilst enjoying a statutory right of way over motorists, cannot disregard attendant traffic perils arising from human lapses. The motorists’ obligation to comply with the red traffic signals does not necessarily mean that the pedestrian can simply rely on it being observed to negate his own responsibility to avoid road perils. We now turn to discuss two local cases to illustrate the proposition that the failure on the part of the pedestrian to check for approaching traffic can lead to a reduction of damages a pedestrian is entitled to receive from the negligent driver.

65 The first case is that of Khoo Bee Keong v Ang Chun Hong and another [2005] SGHC 128 (“Khoo Bee Keong”), where a bus failed to stop at a traffic-light controlled junction and negligently collided into a pedestrian just when he had started to cross the road. Andrew Phang Boon Leong JC (as he then was) held that the pedestrian was guilty of contributory negligence for failing to check the lane on which the bus was travelling. On the facts, the pedestrian did not do enough despite having checked the other two lanes of the road. The court found it imperative that the pedestrian should ensure the vehicles had come to a stop or would be coming to a stop before crossing the road, even though the traffic lights were green in his favour and red against the bus. He only noticed the bus when he had already stepped onto the road, and by then it was too late. Notably, the court found that the pedestrian was not entitled to assume just because cars on the other two lanes had stopped and the green man had come on, the bus which he noticed was travelling at some speed about 8m away would also come to a stop (at [10] and [92]). Given that the risk that the bus might not come to a stop in time was foreseeable, the court found that the pedestrian was guilty of contributory negligence for not ensuring it would be safe to cross, and accordingly reduced the damages awarded to him by 20%.

66 For present purposes, it is significant that Phang JC made no definitive finding that the plaintiff stepped onto the crossing immediately upon the green man appearing (see [92] of Khoo Bee Keong). While there was some suggestion on the evidence that the pedestrian entered the crossing after the green man had appeared, it was not entirely clear exactly how long the green man had turned on before the pedestrian stepped into the path of the errant bus-driver. What we do know is that the vehicles at the other lanes had stopped. We also note that the court did not confine the pedestrian’s duty to guard against errant drivers to the first two seconds of the green man appearing and not afterwards.

67 We turn to the second case of Yip Kok Meng Calvin (a minor) v Lek Yong Han (Yip Ai Puay, third party) [1993] 1 SLR(R) 147 (“Yip Kok Meng”), in which Judith Prakash JC (as she then was) similarly found contributory negligence against a pedestrian for failing to keep a proper lookout. An infant plaintiff, aged 9, was knocked down by a motor car driven by the defendant near the junction of Geylang East Central and Geylang East Avenue 2. The plaintiff was with his family when the accident happened, but he walked ahead of them and had reached the pedestrian crossing first. The defendant intended to make a left turn (after the pedestrian crossing) into Geylang East Avenue 2. Even though the traffic light was green, the defendant stopped his vehicle at the junction waiting for the plaintiff to cross as the pedestrian light was also green, but decided to move along when the plaintiff turned backwards along the pedestrian pavement. Without looking out for the defendant’s car, the plaintiff dashed out onto the road and was hit. The court found at [10] that the plaintiff’s own negligence had contributed to the accident and assessed his contribution at 30% (at [11]). While Yip Kok Meng was not a case of a driver running a red light but of a driver failing to give precedence to pedestrians when he made a turn at the traffic junction, it illustrates an analogous situation where a pedestrian must still exercise due care for his own safety.

68 While neither of these two cases referred to r 22 of the Highway Code, they are nevertheless useful insofar as they illustrate the shared responsibility that a pedestrian has in protecting his own safety along signalised pedestrian crossings even when the traffic lights are in his favour.

69 Before we depart from this issue, we propose to deal briefly with an English case relied on by the Respondent. The Respondent cites the English Court of Appeal decision of Bailey v Geddes [1938] 1 KB 156 as authority for the proposition that a claimant within a pedestrian crossing can never be subject to contributory negligence.

70 The plaintiff in Bailey v Geddes claimed damages from the defendant for colliding into him while he was making his way across a pedestrian crossing. Notably, he had almost reached the other end of the crossing when the accident occurred. All three Lord Justices in that case ruled in favour of the plaintiff, rejecting the defence of contributory negligence. Greer LJ held that the Pedestrian Crossing Places (Traffic) Provisional Regulations 1935 (“the Regulations”) which required drivers to give way to pedestrians at pedestrian crossings were so framed to make it impossible for contributory negligence to arise (at 161). Slesser LJ and Scott LJ made similar remarks. The Respondent relies on those holdings for the proposition that a driver who wrongfully enters a pedestrian crossing has no right to assert the defence of contributory negligence.

71 We disagree with the Respondent’s interpretation of Bailey v Geddes. Bailey v Geddes was a case decided before contributory negligence was statutorily amended to become a partial defence in 1945, and therefore has to be understood in that light. Prior to that, courts had no power to apportion the losses between the plaintiff and the defendant in accordance with their relative blameworthiness. When both the pedestrian and the driver were negligent, whoever’s conduct that was the “effective or predominant cause” of the damage would therefore be ascribed full responsibility for the damage (Davies v Swan Motor Co (Swansea) Ltd [1949] 2 KB 291 (“Davies v Swan Motor”) at 322 per Denning LJ). In our view, the Lord Justices’ comments in Bailey v Geddes must be understood in that context. When they said it was impossible for contributory negligence to arise, they must have meant it was inconceivable for the pedestrian to have been the “effective or predominant cause” of the accident rather than it was impossible for any fault to be attributed to the pedestrian.

72 For completeness, we should make a brief reference to the evolution of the rule in Bailey v Geddes in England. Three English Court of Appeal decisions delivered shortly after Bailey v Geddes restricted its holding to operate only in situations where the pedestrian is already in the midst of crossing the road as opposed to a situation where the pedestrian is run down immediately after he has left the pavement for the crossing (Chisholm v London Passenger Transport Board [1939] 1 KB 426 (“Chisholm”); Wilkinson v Chetham-Strode [1940] 2 KB 310 (“Wilkinson”); and Sparks v Edward Ash Ltd [1943] 1 All ER 1 (“Sparks”)). The following holding made by Scott LJ (who was one of the Lord Justices that decided Bailey v Geddes) in Chisholm aptly summarises the narrow construction given to Bailey v Geddes in subsequent decisions (at 439):

73 In our judgment, the holding in Bailey v Geddes cannot be considered in isolation; it must be understood in the light of the subsequent decisions involving Chisholm, Wilkinson and Sparks. When one appreciates Bailey v Geddes together with the trilogy of decisions, it becomes apparent that the English Court of Appeal did not go as far as to lay down a general principle of law that a pedestrian can in no circumstances be liable for contributory negligence when using a pedestrian crossing. The proposition that Bailey v Geddes instead stood for, as distilled by the Lord Justices in the subsequent decisions, was that a pedestrian who is already in the midst of moving through the crossing has no duty to constantly look out for approaching traffic.

74 The trilogy of cases involving Chisholm, Wilkinson and Sparks may be understood as laying down a rule that once a pedestrian has set foot on the crossing, his duty to keep a proper lookout ceases entirely – he can never be blamed if he is run down in the midst of his crossing. In our judgment, an important qualification must be grafted upon the rule in its current form – the duty to check would cease only after the pedestrian has satisfied himself that it is safe to cross the road before leaving the footway. The qualification is necessary because the Highway Code – which did not appear to have arisen for consideration in the three English cases – takes pre-eminence in the present case. Rule 22 of the Highway Code requires a pedestrian to satisfy himself that vehicles in front of him have either stopped or at the very least are coming to a stop (see [50]-[51] and [58] above). This means that if the pedestrian chooses to enter the crossing despite having noticed a vehicle travelling at a speed and from a distance which suggests that a collision might occur unless the vehicle slows down, he should still take reasonable care for his own safety to keep an eye on that vehicle as he embarks on the crossing. In this regard, we reiterate our analysis at [58] above.

75 In sum, we hold that a pedestrian is not entitled to be guided exclusively by the green pedestrian signal. He ought to keep a proper lookout for approaching traffic before commencing to cross even at a signalised pedestrian crossing with the lights in his favour. He should, as prescribed by r 22, ensure that the vehicular traffic in front of him has stopped or at least is slowing down to a stop. If he fails to discharge this duty, the damages which he is otherwise entitled to may be reduced by reason of contributory negligence on his part.

76 Given our decision to answer the first question in the affirmative, the second question arises for determination. The Respondent was already halfway through the pedestrian crossing when the collision took place. To be specific, he was knocked down in the second half of the crossing just two to three steps past the centre-divider. We have said earlier that there is no obligation on the pedestrian to constantly check for traffic when he is already in the midst of completing the pedestrian crossing (save for the instance where he has already noticed a vehicle coming his way at a certain speed which a reasonably safety-conscious pedestrian would regard as unsafe to ignore) (see [58] above). The Appellant, however, argues that certain configurations of the road in this case would have prompted a reasonably safety-conscious pedestrian to check for approaching traffic at or near the centre-divider before he enters the second half of the road. In essence, she urges us to construe the crossing in question as comprising two separate crossings – such that a duty to check arises again when the pedestrian reaches the centre-divider. In principle, we agree with the Appellant’s basic proposition that the configuration of a road is relevant to determining the duty of care and the assessment of contributory negligence (see Cheng (by his tutor Qian) v Geussens (2014) 66 MVR 268 at [26]). In the light of the Appellant’s arguments and the proposition that a pedestrian has no general duty to constantly check for approaching traffic when he is in the midst of moving through a crossing, an important though not necessarily decisive, consideration that arises in this case is whether the crossing in this case should be regarded as a single, unbroken crossing. This is because of r 20 of the Highway Code to which we now turn.

77 Bukit Batok West Avenue 5 has a centre-divider running through the length of the road. The centre-divider breaks up at the marked pedestrian crossing, leaving a gap for pedestrians to pass through. Rule 20 of the Highway Code reads:

78 Mr Anthony Wee (“Mr Wee”), counsel for the Appellant, submits that the centre-divider is akin to a central refuge. Accordingly, on the authority of r 20, the pedestrian should exercise due care before crossing the second half of the road.

79 In order to determine whether the present case comes within r 20, we need to consider the meaning of the terms “crossing” and “central refuge”. In this regard, two cases are pertinent.

80  The first case is Wilkinson (cited at [72] above). The plaintiff was a pedestrian who left the western pavement of a road and started to go over a controlled pedestrian crossing when the traffic lights were in his favour and against the defendant’s car. Before he reached a refuge in the centre of the crossing, the lights changed and the defendant received a signal that she was allowed to proceed over the crossing between the refuge and the eastern pavement of the road. The plaintiff paused at the refuge before stepping onto the roadway in front of the defendant’s car which knocked him down. In an action for damages, the plaintiff argued that the defendant breached the statutory duty imposed on her by r 5 of the Regulations which required her to allow him free and uninterrupted passage over the crossing. The defendant raised contributory negligence, even though she admitted that she did notice the plaintiff on the refuge. The plaintiff responded by arguing that the defence was not open to the defendant by reason of Bailey v Geddes.

81 Bailey v Geddes was distinguished by the court and the defendant’s plea of contributory negligence succeeded. Slesser LJ (who was also one of the Lord Justices that decided Bailey v Geddes ), in delivering the leading judgment of the court, initially observed that there may be some difficulty in distinguishing Bailey v Geddes as in both cases the accident occurred when the pedestrian was moving through a crossing. Thus, he remarked, “[i]f the crossing be regarded as one continuous crossing within the regulations and the presence of the refuge is to be ignored, the case is clearly within the principle of Bailey v Geddes” (at 321). But he eventually arrived at the conclusion that the presence of the refuge divided what would have been a single crossing into two separate crossings. The refuge was not part of the crossing as a pedestrian resting atop the refuge was in no reasonable peril of encountering vehicular traffic (at 323). In reaching that conclusion, Slesser LJ first considered the definition of “crossing” under the Regulations, which in his understanding meant an area where passengers might expect to encounter crossing peril from traffic other than that of foot passengers. Construing the term in that manner advanced the Regulations’ objective of protecting pedestrians as it required motorists to give way to pedestrians at designated crossings. As a foot passenger could not possibly be in any reasonable peril of encountering vehicular traffic at the refuge, he found that the refuge could not have fallen within the meaning of “crossing” intended by the legislature. He thus concluded that an island or refuge in the middle of a road was not part of a crossing, and consequently, the crossing was in actuality two separate crossings.

82 Scott LJ shared the same view. He observed that islands in the middle of the road are often erected at busy roadways for pedestrians to pause for safety before they resume their passage onto the second half of the crossing. Viewed from this perspective, islands or refuges understandably divided what would appear as a single crossing into two separate crossings (at 326-329):

83 The second case is that of Chun Sung Yong v Au Sze Hung Christopher & anor [1991] 1 HKC 556 (“Chun Sung Yong”), a decision by the High Court of Hong Kong. The plaintiff in that case was a pedestrian walking across Nathan Road at the junction of Kimberley Road by way of a pedestrian crossing when he was knocked down by a car driven by the second defendant who was the first defendant’s agent. Nathan Road was a dual-carriageway road, with two lanes in either direction and divided by railings erected on a raised concrete base. The pedestrian traffic light was in the plaintiff’s favour when the accident occurred. When the plaintiff crossed the centre pedestrian refuge through the opening in the railings into the northern-bound carriageway of Nathan Road (still part of the pedestrian crossing), the pedestrian light remained green. He was hit when he had already walked past the fast lane into the slow lane of the northern-bound carriageway.

84 The plaintiff maintained that he only looked at the traffic light and decided that he need not pay attention to his surroundings when crossing the road. The court allowed the claim, but reduced the damages payable by 25% for contributory negligence on the plaintiff’s part – 15% for failing to keep a proper lookout and 10% for not stopping at the central reservation. Whilst it was found that the defendant must have deliberately flouted the traffic law, ploughing recklessly through the pedestrian crossing, the court found that the plaintiff contributed to the accident by failing to do two things. First, he failed to pause at the centre refuge for his own safety; and secondly, he failed to keep a proper lookout throughout the journey. The accident could well have been avoided had he done either of these two things or, better still, both (at 562-563):

85 For present purposes, we are only concerned with the first of the two holdings made by the court touching on r 33(5) of the Hong Kong Road Traffic (Traffic Control) Regulations (Cap 374G) (which is almost in pari materia with r 20 of our Highway Code). Regulation 33(5) reads as follows:

86 The court answered the question of whether a safety island falls within the meaning of “refuge” under Regulation 33(5) in the affirmative. It gave the following reasons for regarding the crossing as constituting two separate crossings (at 563):

87 The court in Chun Sung Yong defined the term “refuge” to mean a place for the pedestrian to pause so as to be sheltered from traffic. On the facts before the court, the central reservation in Chun Sung Yong fell within that definition as its deployment specifically demarcated a spot within the crossing for pedestrians to pause before resuming their journey. The same sort of safety island persuaded the Lord Justices in Wilkinson to require the pedestrian to exercise due care before leaving the island for the road. In our judgment, we agree that safety islands have the effect of breaking up what would otherwise have been a single crossing into two separate crossings. The presence of a safety island – a place normally in the centre of a dual-carriageway road where a pedestrian could take refuge from the dangers of traffic on the road – interrupts the pathway not just visually but also functionally. Typically found along busy and/or wide roads, they are installed to inform and assist pedestrians to cross the road in two stages, prompting them to check again for approaching traffic before leaving the “zones of safety”. They also provide a place for the pedestrian to pause should the lights blink or turn red while he was crossing.

88 The construction of what constitutes a “central refuge” found in Wilkinson and Chan Chun Hong accords with our understanding of the statutory objective behind r 20 of the Highway Code. A crossing with a central refuge somewhere in the middle would invariably involve a broad dual-carriageway road with at least two or more lanes in each half of the road and such roads often experience heavy and/or fast-flowing vehicular traffic. For roads of this sort, central refuges are deployed to promote pedestrian safety by allowing pedestrians to cross them in two stages. In our judgment, r 20 is meant to remind pedestrians that when they are moving through pedestrian crossings found in roads of this sort, for their own safety, they should treat each half of the entire pedestrian crossing as a separate crossing and comply with the safety precaution prescribed by r 22 when they cross each half. Implicit in r 20 is the recognition that dual-carriageways of this sort, by their sheer width and by the sheer volume of vehicular traffic they carry, would often pose additional dangers as opposed to single carriageways. They should therefore exercise care, as required by rr 20 and 22, before they enter the second half of the crossing.

89 In the present case, the crossing in question was a straight passageway which remained unbroken and uninterrupted from one side of the road to the other. While the facts of Wilkinson are somewhat different from those of our case in that here there is no safety island in the middle of the crossing in question, the principle that can be extrapolated from Wilkinson is that it is possible for a dual-carriageway to be treated as two separate crossings in certain circumstances. Centre-dividers constituting part of a pedestrian crossing come in various forms and dimensions and can be found across a wide variety of road configurations. In some instances they come in the form of a narrow strip of concrete slab. In other instances, they could be of a dimension that is sufficient to accommodate a few people to pause there for a brief respite before crossing the second half of the crossing. In principle, and for reasons discussed at [88] above, r 20 should be interpreted liberally in a manner that would advance the object of that rule, which is the promotion of the pedestrian’s safety. Rule 20 does not prescribe the required dimension of a centre-divider before it can constitute a “central refuge”. It is a question of fact in each case if the centre-divider constitutes a “central refuge”. In this case, though we have not been provided the dimensions of the centre-divider, it appears to us to be broad enough in terms of the length of the gap to offer refuge to a few people in order that they may pause there and obtain shelter with safety. However, since the exact dimensions of the centre-divider have not been admitted as evidence and no evidence or submissions were tendered in relation to the specific purpose of the centre-divider in our case, we do not premise our decision that the Respondent was guilty of contributory negligence entirely on the basis that the centre divider here comes within the meaning of “central refuge” as it appears in r 20. We now elaborate why there are other sufficient reasons to support a finding that the Respondent was contributorily negligent in the circumstances.

90 If r 20 is not applicable, the crossing in question would be regarded as a single crossing, in which case r 22 of the Highway Code would require the pedestrian to assure himself that the vehicular traffic in front of him has stopped or is stopping before he enters the crossing. The present facts concern a dual-carriageway and there is nothing to indicate that dual-carriageways are to be exempted from this rule (see [56] above). Looking at the factual scenario before us through the lenses provided by r 22, the pedestrian must assure himself that the vehicles coming from both sides of the dual-carriageway have either come to a stop or are slowing down with a view to stopping before he may leave the footway for the crossing. To do so, the pedestrian would have to assess the vehicular traffic flow on both halves of the road. We illustrate the pedestrian’s duty by reference to two examples of road features within which a pedestrian crossing may be situated:

(a) A straight dual-carriageway (with two lanes on each half of the road) without a fence erected at its centre.

(b) A dual-carriageway (with two lanes on each half of the road) without a fence erected at its centre, but the road curves away from the pedestrian some 60–70m away.

91 In the first example, a pedestrian can see for himself that vehicles on each half of the road has come to stop or are coming to a stop.

(a) He can safely step onto the crossing having satisfied himself that vehicles on both halves of the road have noticed the traffic signal and are complying with it.

(b) But if, for instance, he notices a vehicle on the second half of the road showing no signs of slowing down and nevertheless decides to enter the crossing, r 22 would require him, before entering the second half of the crossing, to glance to his left to guard against the possibility of being run over by a potentially errant motorist (see [56] and [58(a)] above).

92 In the second scenario, a pedestrian who is preparing to step into the crossing may check to see whether vehicles on each half of the road have complied with the traffic signal. Let us assume that he is satisfied that vehicles on the first half of the road has stopped or are coming to a stop.

(a) Let us also assume for a moment that, having done so, he notices no vehicle coming on the second half of the road. Because the road curves away from the pedestrian, he can check for vehicles coming from his right on the first half of the road but not for vehicles coming along the second half of the road that have yet to finish navigating the bend as these vehicles will be outside his field of vision. For this reason, there is no way for the pedestrian to satisfy himself that there are no vehicles approaching the crossing, and if there are vehicles, whether they are slowing down to a halt at the traffic signal unless he checks again as he approaches the centre-divider.

(b) In an alternative scenario where the pedestrian notices a vehicle emerging out of the bend, but shows no signs of slowing down for the crossing, and if he reasonably ascertains that he might walk into the path of that vehicle by the time he enters the second half of the crossing, then in furtherance of r 22 he should take care of his safety before he enters the second half of the crossing (see [56], [58(a)] and [91(b)] above).

The two hypothetical scenarios postulated above are of course not exhaustive and are examined for illustrative purposes. There could be other factual permutations that, depending on the prevailing circumstances in each case, could accordingly adjust the standard of care required of the pedestrian.

93 With these illustrations in mind, we turn now to the facts of this case. In our judgment, the following features of the road would have impeded a pedestrian’s assessment of vehicular traffic on the second half of the road if the pedestrian made his assessment from the footway before he commenced crossing:

(a) there is a 1.4m tall metal barricade (erected on a raised concrete base) along Bukit Batok West Avenue 5 that separates each half of the road from the other;

(b) the accident happened at about 10pm at night; and

(c) the road on which the Appellant was travelling on, Bukit Batok West Avenue 5, curves to the left from about 150m before the pedestrian crossing and then straightens itself from about 60-70m to the pedestrian crossing.

94 The following photograph of the accident scene (taken from the perspective of a vehicle approaching the pedestrian crossing) exhibited in the expert report illustrates how the presence of the metal barricade would have impeded the visibility of a pedestrian before he reached the midway point of the crossing:

95 From the photograph, taken in broad daylight, it is readily apparent that a pedestrian standing on the pavement just before the first half of the crossing would barely be able to see vehicles travelling on the second half of the road. This visual obstruction would have certainly been more pronounced at night.

96 This other photograph of the scene (taken from the perspective of a pedestrian walking towards the centre of the crossing) exhibited in the same expert report shows the curvature of the road which began some 60–70m from the pedestrian crossing:

97 The three features highlighted, namely, the metal barricade, the bend of the road and the fact that the accident happened at night, would have undoubtedly affected the Respondent’s assessment of the vehicular traffic flow on the second half of the road. We also take cognisance of these salient aspects of the three features:

(a) While there were gaps in between the vertical metal bars that form the metal barricade, the Respondent would nevertheless have been prevented from acquiring an accurate assessment of the speed at which the vehicles were travelling on the second half of the road and whether those vehicles on that side of the road had stopped or slowed down.

(b) We also place emphasis on the fact that the accident occurred at about 10pm at night which means that even with functioning street lights, the Respondent’s vision of the activities across the metal barricade would not have been the same as if it were in normal daylight.

(c) Moreover, the fact that the road started to curve away from the Respondent’s field of vision some 60–70m away from the pedestrian crossing compounded the problem. The curve in the road would have probably put the Appellant’s vehicle out of the Respondent’s field of vision when the latter was on the pavement about to cross. We make this assumption on the basis that an ordinary person, the Respondent included, would have taken at least five seconds to complete the first half of the crossing, which in turn means the Appellant’s vehicle (assuming it was travelling constantly at 55 km/h) was at least 76m away from the crossing just before the Appellant entered the crossing.

98 These three factors would have affected the Respondent’s assessment of how fast the Appellant’s vehicle was travelling along Bukit Batok West Avenue 5. Given the difficulties we have highlighted in making a proper assessment of the vehicular traffic on the second half of the road from where the Respondent was standing before he entered the crossing, we find it unlikely that the Respondent would have looked to his left and noticed the approach of the Appellant’s vehicle. In any event, even if he had done so, any assessment of the speed at which the Appellant’s vehicle was travelling would have been of limited reliability. In these circumstances, he should have ensured that vehicles on the second half of the road had also come to a stop or were at the very least slowing down before he commenced crossing (assuming we treat this crossing as a single crossing). But as we have said, even if the Respondent had tried to satisfy himself that the vehicles were at least slowing down if not having already stopped, his assessment of vehicular traffic on the second half of the dual-carriageway would have been unreliable. In these circumstances, even assuming r 20 does not apply, he should have, when nearing the centre divider, taken care to ensure that vehicles on the second half of the road had in fact halted or were slowing down with a view of stopping.

99 We are cognisant that our above analysis presupposes that the Respondent would not have seen the Appellant’s approach when he was about to enter the crossing. The main basis for saying this is the curvature of the road, our assumption that the Appellant was driving constantly at 55 km/h and that the Respondent took at least five seconds to finish the first half of the crossing (see [97(c)] above). On these factual assumptions, we surmised that the Respondent would not have noticed the Appellant’s approach when he was just about to enter the crossing and therefore should have checked again near the centre-divider. But what if in the unlikely chance that these assumptions are not true? For instance, the Respondent may have in fact taken fewer than five seconds to reach the second half of the crossing. On this alternate set of facts, the hypothesis would be that just before the Respondent entered the crossing, the Appellant’s vehicle would have emerged from the bend. Let us also assume for the sake of argument that, notwithstanding the metal barricade and the reduced visibility at night, the Respondent, having looked to his left, noticed the Appellant’s approach but nevertheless decided to enter the crossing. Even then, r 22 would still require the Respondent to guard against the possibility of being run over by a potentially errant motorist (see [92(b)] above). In this regard, his duty under r 22 would be to satisfy himself that the Appellant’s vehicle would be coming to a stop at the crossing. Because the metal barricade and the reduced visibility at night would have impaired his assessment of the speed at which the Appellant’s vehicle was travelling at, he would have been in no position to establish that fact unless he made a separate assessment when he got nearer to the centre-divider.

100 The Judge below found that the Respondent was entitled to assume that vehicular traffic would have stopped or would be coming to a stop. This was also the argument which the Respondent sought to run at the hearing of this appeal, namely, that the law in this area should proceed based on the duration the traffic lights has been in favour of the pedestrian. The Respondent argued that the pedestrian should only be expected to check within the first few moments of the traffic lights turning green in his favour to guard himself against the more common risk of motorists trying to beat the red lights but failing to do so. Vehicles are expected to stop at red lights, and having made it through half the crossing, it is not unreasonable for the pedestrian to be lulled into some sense of security that vehicles would by then have stopped or would come to a stop.

101 While we acknowledge that the risk of motorists beating red lights is certainly abated by the passage of time, that is only so with regard to motorists who deliberately try to squeeze through the amber lights but fail to do so. We reiterate our earlier point that this is not the only instance where motorists run red lights; such risks could equally eventuate as a result of motorists suffering from lapses of attention (eg, being distracted by crying or quarrelling children in the rear seats); motorists falling asleep behind the wheel and motorists driving under the influence of alcohol (see [30]–[32] above). These are all real risks which could happen not just within the first few seconds of the traffic signal turning red but could occur anytime, even after a number of seconds have elapsed (see [53]–[54] above). Rule 22, as we have explained, gives statutory foundation to this premise. A reasonably prudent pedestrian would thus pay heed to and guard against motorists running red light regardless of how long the lights have turned in his favour. We would reiterate again that under r 22 a pedestrian should make sure that the vehicles in front of him have come to a stop (even in a signalised pedestrian crossing which is in his favour) before he commences to make the crossing, and this will include vehicles travelling on the second half of the dual carriageway. In other words, the duty which the pedestrian should exercise here is not unlike that imposed under r 20 because of the special road features surrounding the pedestrian crossing in question (see [97] above). In view of the foregoing, we find that the Respondent should have at or near the centre-divider of the pedestrian crossing checked whether it was safe to cross the second part of the crossing, notwithstanding the fact that the crossing was an unbroken pathway.

102 The Respondent referred us to three English cases, namely, Tremayne v Hill [1987] RTR 131, Frank v Cox (1967) 111 Sol Jo 670 and White v Saxton (1996) 11 PMIL 10 as authorities for his proposition that contributory negligence may not be invoked against a pedestrian where the pedestrian signal is green. In relation to the latter two cases, the Respondent tendered two case summaries found in Paul J Taylor, Bingham and Berryman’s Personal Injury and Motor Claims Cases (LexisNexis, 13th Ed, 2010) as follows:

103 The facts of Tremayne v Hill may be briefly stated as follows. The pedestrian in that case was crossing at night diagonally over two roads which converged at a busy junction governed by traffic lights. He did not make use of a light-controlled pedestrian crossing nearby. Just before the pedestrian had completed the crossing he was struck and injured by a car driven by the defendant that had crossed into the junction against a red traffic light. The question in that case was whether the pedestrian was contributorily negligent. The English Court of Appeal answered this in the negative, reasoning as follows (at 134J–135B):

104 In our judgment, none of these three cases assist the Respondent. All of them are easily distinguishable from the facts of our case. In all these cases, the pedestrian was knocked down after he had crossed a significant portion of a crossing. None of these cases concerned a pedestrian being knocked down right after he had stepped onto the second half of a dual-carriageway with each half of the road separated by a metal barricade. In Tremayne v Hill, the collision took place “just before [the pedestrian] completed his crossing of the road”. In Frank v Cox, the pedestrian “had almost reached the centre refuge” when he was knocked down, ie, the pedestrian had not even completed the first half of the dual-carriageway. In White v Saxton, it is evident that the pedestrian was knocked down some time after she had entered the crossing and there was no indication that the facts involved a dual-carriageway.

105 There is a more fundamental reason why these cases ultimately do not assist the Respondent. Even if the Respondent were correct to interpret the outcome arrived at in these cases in his favour, r 18 of the UK Highway Code (2015 edition) would likely have impacted the court’s reasoning had due regard been given to this rule. Rule 18 of the UK Highway Code reads as follows:

106 Rule 18 does not appear to have been considered in any of these cases. Just like r 22 of our Highway Code, r 18 clearly required pedestrians to ensure vehicular traffic in front of him has come to a stop before entering the crossing. Any holding to the contrary would be inconsistent with this rule. The rules in the UK Highway Code were specifically enacted with the object of reducing road casualties and in this regard it carries the same objective as our Highway Code. The introduction to the UK Highway Code states as follows:

107 In these circumstances, we caution against placing undue reliance on these three English cases.

108 Our primary concern with using these three English cases as relevant precedents have been captured in the two grounds stated above (at [104]-[106]). For completeness, we should add that the full law report of White v Saxton was not made available to us. Without a full appreciation of the facts before the court that heard White v Saxton, it is difficult for us to derive significant precedential value from that case. Rajah JC sounded similar cautions in Cheong Ghim Fah as follows:

109 In this connection, Chan Sek Keong CJ’s caution against placing undue reliance on unreported precedents in Luong Thi Trang Hoang Kathleen v Public Prosecutor [2010] 1 SLR 707 at [21] are also apposite:

110 Given that motor accident cases by their very nature tend to be highly fact-sensitive, we would refrain from placing undue reliance on unreported authorities that fall within this category of cases.

111 For these reasons, we answer the second question in the affirmative. This leads us to conclude that the Respondent should have, in the factual situation here, kept a proper lookout before entering the second half of the pedestrian crossing.

112 Our decision to answer the first two questions in the affirmative necessitates an answer to the final question – did the Respondent in fact check for approaching traffic at the centre-divider?

113 The burden is on the Appellant to prove that the Respondent was contributorily negligent. No direct evidence exists to substantiate her case. The Respondent understandably could not shed light on this matter as his recollection of the incident was impaired by the concussion he suffered as a result of the accident. The Appellant too would not have known for a fact whether the Respondent did or did not check for traffic before crossing the second half as she only caught sight of the Respondent just moments before running into him.

114 However, we do have some objective facts from which reasonable inferences can be drawn. It is not in dispute that the road was dry on that day, visibility was clear and the weather was fine. The Respondent was hit in his second or third step past the centre-divider. An ordinary person, the Respondent included, would have taken no more than 1 to 1.5 seconds to cover that distance by foot. It was not disputed that the vehicle was travelling at about 55 km/h. Based on that speed, the Appellant’s vehicle was probably about 15–23m away from the Respondent at the time the Respondent was walking past the centre-divider. We arrive at this figure having regard to the fact that the minimum braking distance of a car travelling at the speed of 50 km/h is about 23m (r 11 of the Highway Code); at 55 km/h, the braking distance would probably have been a few metres more than 23m at the very most. Any reasonable person upon seeing a vehicle hurtling towards him at that speed and within such close proximity with no signs of slowing down would hesitate to step into the path of the vehicle. If the Respondent had checked, it would have been apparent that the vehicle was not going to stop in good time, and he would not have continued his journey across the second half of the crossing. Based on the premises we have examined, it was evident that the Respondent did not check for approaching vehicular traffic. Even if he did, he clearly failed to make a reasonable assessment of the risks posed by the approaching vehicle.

115 Counsel for the Respondent argues that the Appellant’s case that the Respondent had failed to check for oncoming vehicles was not put to the latter when he was on the witness stand, violating the rule in Browne v Dunn (1893) 6 R 67 and it should therefore be struck down. We disagree. The rule in Browne v Dunn is not a rigid, technical rule; it is not meant to be applied mechanically as to require every single point to be put to the witness. In deciding what questions to put to the witness, one must bear in mind the rationale of the rule, which is to give him an opportunity to respond to allegations made and to explain himself (Hong Leong Singapore Finance Ltd v United Overseas Bank Ltd [2007] 1 SLR(R) 292 at [42]). On the facts before us, the Appellant’s failure to put her case to the Respondent does not violate the mischief targeted by the rule having regard to the following circumstances:

(a) The only factual issue arising for determination was whether the Respondent failed to keep a proper lookout. The Appellant’s case that the Respondent had failed to keep a proper lookout was made abundantly clear not only in her pleadings ⁶ but also her affidavit of evidence in chief (“AEIC”). ⁷

(b) The Respondent had already given his answer to the Appellant’s case in his AEIC, which was that he could not remember a single thing about the incident due to post-concussion symptoms. The Appellant is well-entitled, as she did, to accept that as the truth in the light of the medical reports attesting the same by choosing not to challenge the Respondent’s evidence-in-chief in cross-examination.

(c) Even if the Appellant had questioned the Respondent on whether he failed to keep a proper lookout, the Respondent’s answer would in all likelihood have been “I cannot remember”. It would thus have served no meaningful purpose for the Appellant to put her case to the Respondent.

116 Before leaving this point, we make one observation. The Respondent further seeks to persuade us that imposing a general obligation on all pedestrians to check at the midway point of a crossing would be too onerous on selected classes of claimants such as the wheel-chair bound or a very young child. His contention touches on the issue of whether the objective standard of care expected of the claimant can be modulated in circumstances where the claimant is labouring under certain infirmities or disabilities. We are aware that this area of the law remains unsettled and is evolving. It has generated some controversy, as is evident from the Western Australia Court of Appeal decision in Town of Port Hedland v Hodder (No 2) (2012) 294 ALR 315 which saw judicial disagreement as to the proper characterisation of the appropriate standard of care courts should adopt in assessing contributory negligence (see also Margaret Fordham, Contributory Negligence and the Disabled Claimant [2013] SJLS 192 for a perceptive analysis of this issue). However, as this issue does not arise for our determination on the facts of the present case, we would leave it to be considered at a more appropriate juncture in the future.

117 In these premises, we find that the Respondent should have checked for approaching vehicular traffic before seeking to cross the second half of the road, and given the circumstantial evidence available we find that he did not do so.

118 We now come to determine the appropriate reduction in the Respondent’s damages for his contributory negligence. Section 3(1) of the Contributory Negligence Act requires the damages to “be reduced to such extent as the court thinks just and equitable having regard to the claimant’s share in the responsibility for the damage”. Consistent with the court’s wide discretion conferred by statute in determining what is “just and equitable” on the facts of each case, it has been said that a finding of apportionment is a finding upon a question, “not of principle or of positive findings of fact or law, but of proportion, of balance and relative emphasis and of weighing different considerations” (British Fame (Owners) v MacGregor (Owners) [1943] 1 AC 197 at 201 per Lord Wright). The apportionment exercise is to be applied in a “rough and ready” manner, for the factors that the court is required to consider are “incapable of precise measurement” and are often “incommensurable” (Jackson v Murray and another [2015] UKSC 5 (“Jackson v Murray”) at [27]–[28] per Lord Reed and at [46] per Lord Hodge). Amongst the many considerations that may fall to guide the exercise of the court’s discretion, two aspects stand out in shaping the court’s determination of what is a “just and equitable” apportionment: first, the relative causative potency of the parties’ conduct; and second, the relative moral blameworthiness of the parties’ conduct (Davies v Swan Motor at 326 per Denning LJ; Baker v Willoughby [1970] AC 467 at 490 per Lord Reid; Eagle v Chambers (No 1) [2003] All ER (D) 114 (“Eagle v Chambers”) at [10] per Hale LJ).

119 Drivers, unlike pedestrians, are in a position to cause harm. In cases of motor accidents, courts tend to hold motorists as the more culpable party having regard to the “destructive disparity” between a driver and a pedestrian (Eagle v Chambers at [16]; see also Pennington v Norris (1956) 96 CLR 10 at 16–17; Anikin v Sierra and another (2004) 211 ALR 621 at [51]). But this is not a hard and fast rule as apportionment is, at its core, a highly fact-sensitive exercise (Teubner v Humble (1963) 108 CLR 491 at 504). One instance justifying a departure from the general rule is when a pedestrian suddenly moves into the path of an oncoming vehicle. Pedestrians falling within this category may be adjudged to have exhibited conduct which has a higher degree of causative potency than the conduct of the driver (see Eagle v Chambers and Jackson v Murray).

120 No such difficulty in attributing causative potency and blameworthiness to the respective parties arises in this case. On the facts, it is clear to us that the Appellant was mostly to blame for the accident. While we found that the Respondent could have avoided the accident, it does not detract from the fact that the harm emanated from the irresponsible manner in which the Appellant drove her vehicle (whatever might be the cause of her inattentiveness at the relevant moment) in ploughing what was essentially a dangerous weapon through a pedestrian crossing in complete disregard of the traffic signal. We further consider that the Respondent had taken some precaution by crossing the road in compliance with the traffic signal. To reflect the greater causative potency and blameworthiness of the Appellant’s atrocious driving, we reduce the Respondent’s damages by a modest 15%.

121 At the heart of today’s appeal lies the question of the appropriate duty of care which pedestrians ought to exercise when using signalised pedestrian crossings. It is a legal issue that is in essence heavily driven by public policy considerations. How frequently do motorists run red lights after the first one or two seconds of the traffic lights turning red against vehicular traffic as opposed to them doing so several seconds afterwards, and are both forms of risks reasonable for the pedestrian to guard against? What are the objects behind the rules prescribed in the Highway Code with reference to the configuration of traffic signals and road features? How best should we as a society allocate the risk of road accidents and the burden of preventing them from happening? It is trite that courts are not well-placed to deal with public policy issues, and given what is prescribed in the Highway Code, we can only discern such policies from those rules. It is important to note that the Highway Code was promulgated after several road traffic agencies were consulted and these stakeholders are in a much better position to answer these questions (see [43] above). As far as we can see, while vehicular traffic must observe traffic signals, including those at pedestrian crossing, our society has recognised through the making of rr 22 and 20 of the Highway Code that there will be, from time to time, lapses on the part of motorists (intentional or not), regardless of how long the traffic signal has turned red against them, and for that reason pedestrians should take charge of their own safety and should be conscious of errant motorists before they enter pedestrian crossings, especially in road conditions involving a dual-carriageway like the present. This message comes out loud and clear from those two rules. Underlying this legislative framework is the policy objective of reducing the incidence of road accidents, and in this regard, accidents are certainly much less likely to occur if pedestrians are mindful of their own safety. The Highway Code was first promulgated in 1931, ⁸ a time where consumer mobile technology was unknown to mankind. Today, with the advent of various modern technological devices which could distract not only the motorists but also the pedestrians on the road, there is added urgency in promoting care on the part of all road users to achieve the objective of reducing the incidence of road accidents.

122 On the facts of the present appeal, we are satisfied that the Respondent failed to properly discharge this duty of his. The pedestrian crossing concerned was a dual-carriageway, and we surmised that given the prevailing circumstances, he was unable to have satisfied himself whether there were vehicles on the second half of the dual-carriageway, and if so, whether they had in fact complied with the traffic signal. For this reason, r 22 would have required him to remain attentive before stepping into the second half of the crossing. Had he done so, he would not have simply stepped into the path of a vehicle which was coming towards him at close proximity while showing no signs of slowing down.

123 For the above reasons stated, we allow the appeal with half costs. We are not disturbing the costs order made below. We must, however, emphasise that whilst we have found the Respondent to be also negligent, we have not overlooked the fact that the accident was almost entirely occasioned by the atrocious lapses of the Appellant and the fact that the Respondent’s life was placed in great peril as a result. Pedestrians must be given precedence when using pedestrian crossings, and drivers who fail to do so can expect not only civil but also serious criminal sanctions to be meted out. But whilst the law can do its best to deter and compensate, it cannot bring back a life lost nor can it salvage irreparable wounds. As the Chinese saying goes, “the road is like a tiger’s mouth”. This saying remains valid even in the context of modern day signalised crossings because its deployment can only reduce road dangers but not eliminate them altogether. In arriving at our decision, we bear in mind that many accidents such as the one before us can well be avoided if pedestrians could simply remain attentive when they cross the road and refrain from getting distracted by mobile devices. The simple act of keeping a proper lookout can potentially translate to an accident averted and a life saved.

124 I have had the advantage of reading in draft the judgment of Chao Hick Tin JA, with which Quentin Loh J concurs. I refer to this as the majority Judgment. The majority Judgment holds that the respondent was contributorily negligent in failing to keep a proper lookout for oncoming traffic as he approached the road’s centre. The majority Judgement also holds specifically that the respondent was under a duty to check for such traffic even though he was crossing an unbroken pedestrian crossing with no centre refuge that was controlled by a traffic light and even though the light had been in his favour for a considerable period of time. The majority Judgement further holds that as a matter of fact the respondent did not check as aforesaid (see [114]–[117] above). The majority has accordingly reduced the damages payable to the respondent by 15% (see [120] above). With great respect, I am unable to agree that the respondent should be found to be contributorily negligent and that the damages awarded to him should be reduced as a result. My reasons follow.

125 The central question raised in this appeal is whether a pedestrian, who has crossed a significant portion of a controlled pedestrian crossing should be held to be contributorily negligent in circumstances where:

(a) the pedestrian crossing lights were in the pedestrian’s favour throughout;

(b) the lights had been in the pedestrian’s favour for a considerable period of time by the time of the collision;

(c) the pedestrian was knocked down by a motorist who for some inexplicable reason wholly failed to see the traffic lights and so not only did not stop but did not even slow down; and

(d) the motorist has admitted that she was driving dangerously and has been convicted of an offence for so doing.

126 The salient facts surrounding the circumstances in which the collision occurred have been recounted in large part in the majority Judgment (see [4]–[8] above). It bears reiteration that the road in question, Bukit Batok West Avenue 5 is a two-way road with dual lanes on each side. The pedestrian crossing is an uninterrupted passage that goes right across both sides of the dual carriageway. There is a break in the road divider between the two sides of the dual carriageway where the pedestrian crossing is situated. A sketch plan of the site taken from one of the expert reports is reproduced below.

127 The respondent was knocked down after he had completed crossing two lanes or half of the whole dual carriageway and had taken a further two or three steps into the other half of the dual carriageway. It was therefore indisputable that the pedestrian crossing lights had been in his favour for a considerable period of time and correspondingly, the traffic lights must have been red against vehicular traffic travelling in both directions for a similar period of time. It is clear that the appellant was driving in a heedless and dangerous manner without any regard for the safety of pedestrians. She was estimated to have been driving at a speed of 55 km/h at the point of the collision despite the red light having been against her for a substantial period of time. Her own expert, based on the analysis of the accident scene, estimates that she was driving at a speed of between 42 km/h and 68 km/h. It seems indisputable that she was oblivious to the red light against her. Her own evidence was that she only realised that the traffic signal was against her after she collided with the respondent. The road was well-lit and the weather was fine at the time of the accident.

128 After being hit by the taxi at its left bonnet, the respondent crashed into its left windscreen before being flung 23.3m from where he was hit. As a result of the appellant’s grave disregard for other road users, the respondent suffered catastrophic, extensive and life-affecting injuries.

129 The culpability and the conduct of the appellant, egregious though they are, should not distract one from the central inquiry when considering contributory negligence. The focus in the latter context is the conduct of the plaintiff, in this case, the respondent (see [18] above). On this, the majority holds against him in broad terms for the following principal reasons:

(a) First, the majority considers that the critical consideration is whether from the perspective of a pedestrian, such as the respondent, the risk of being knocked down by a motorist running a red light is a sufficiently foreseeable one such that the respondent has a duty to guard against such a risk materialising (see [25] above). The majority considers that it is, having regard to various statistics on the number of motorists who beat red lights, two newspaper reports and a forum letter, which I will turn to in the course of my judgment. While the majority acknowledges that the statistics are not sufficiently granular to differentiate between motorists who beat red lights within the first one or two seconds and those running red lights after a considerable time has elapsed, that does not mean that the statistics should be wholly disregarded; indeed, the court must strive to make the best of whatever relevant statistics that are available. Moreover, it is conceivable that at least a proportion of errant motorists captured by the statistics would fall within the latter scenario (see [33] above).

(b) Second, while pedestrians have the right of way at pedestrian crossings pursuant to rr 4, 6 and 7 of the Road Traffic (Pedestrian Crossing) Rules (Cap 276, R 24, 1990 Rev Ed) (“Pedestrian Crossing Rules”), that does not preclude a tortfeasor from raising contributory negligence as a defence (see [36]–[37] above).

(c) Third, even though it has been suggested in “Negligence at Pedestrian Crossings” (1938–1939) Mod L Rev 233 (“Negligence at Pedestrian Crossing”) at 239 that pedestrian crossings are meant to be safe havens in which pedestrians can cross roads without having to pay “undue attention” to vehicular traffic, pedestrians must at least still pay some attention when using pedestrian crossings because they are not “fool-proof” safety measures and indeed merely serve to mitigate rather than eliminate dangers. Indeed, pedestrian crossings merely serve to reduce, and not eliminate the potential for conflicts with motor vehicles (see [37]–[38] above).

(d) Fourth, it is implicit in r 22 of the Highway Code (Cap 276, R 11, S8/75), which exhorts pedestrians at light controlled crossings to wait on footways until traffic has come to a standstill, that the possibility of motorists running red lights is a real one, even in a situation where the traffic signal has turned red against them for a substantial time. This is because the wording of r 22 draws no distinction between a situation where the traffic light has turned red against traffic for a considerable time and that where it has turned red for just a short time; and if Parliament intended this rule to just apply in the former situation, it would have added such words as “when the green man appears” at the beginning of this rule (see [49]–[57] above).

(e) Fifth, the pedestrian crossing here should be treated as two separate crossings such that the Respondent came under a duty to check once more at the road’s centre before embarking on the second half of the crossing. In this context, the majority says that r 20 of the Highway Code, which states that pedestrian crossings divided by centre refuges are to be treated as two separate crossings, should be read together with r 22, and that its position is supported by two cases namely Wilkinson v Chetham-Strode [1940] 2 KB 310 (“Wilkinson”) and Chun Sung Yong v Au Sze Hung Christopher & anor [1991] 1 HKC 556 (“Chun Sung Yong”), both of which held pedestrians liable in contributory negligence for failing to check for oncoming traffic once more at centre refuges. Even though the precise dimensions of the road divider are not in evidence, the majority was prepared to accept that the road divider was akin to a centre refuge because it appears wide enough for a few people to stop there safely to check before embarking on the second half of the pedestrian crossing (see [76]–[89] above).

(f) Sixth, the presence of three features of the road in question in this case at the time of the accident mandates a finding that the respondent should have exercised particular care or attention before he started to cross the second half of the pedestrian crossing and that therefore he should have stopped at the road divider and checked again before resuming the act of crossing. These features are:

(i) the road curvature about 150m before the pedestrian crossing from the direction of the traffic in the second half of the road such that the respondent would not have been able to see oncoming traffic in the second half of the road until he had reached road divider;

(ii) the presence of a 1.4m high metal fence on the road divider which would obstruct the vision of pedestrians (in respect of oncoming traffic in the second half of the road) as well as of motorists (in respect of pedestrians crossing over into the second half of the road from the gap in the road divider); and

(iii) the poor visibility at night of approaching vehicles, in particular the appellant’s taxi (see [93]–[99] above).

(g) Seventh, the cases referred to us by the respondent, namely, Tremayne v Hill [1987] RTR 131, Frank v Cox (1967) 111 Sol Jo 670 and White v Saxton (1996) 11 PMIL 10 are thought to be distinguishable from the facts of the present appeal because those were cases where the pedestrians in question had already crossed a significant distance whereas here the accident occurred shortly after the respondent entered the second half of the dual carriageway which also had a fence in the middle. More fundamentally, it does not seem that the courts in these cases considered r 18 of the UK Highway Code (the present edition of which was published in 2015), which could have impacted the result in those cases (see [104]–[107] above). Moreover, the full law report of White v Saxton was not available to us, and for this reason, “undue reliance” should not be placed on that case (see [108]–[110] above).

(h) Eighth, it was “evident” that the Respondent did not check for approaching vehicular traffic at the road’s centre because if he had done so, he would not have continued completing the second half of the crossing. As to Mr Liew’s objection that the appellant’s counsel should not be permitted raise this point in argument because it was not put to the appellant while he was on the witness stand and that therefore the rule in Browne v Dunn (1893) 6 R 67 is offended, the majority disagree and hold that the rule was not violated (see [112]–[117] above).

130 The majority Judgment also suggests, albeit somewhat tentatively at [59] that the respondent here was akin to a least-cost avoider, since all that he had to do to avoid the harm he suffered was to keep a proper lookout for oncoming traffic and on this basis, it is thought just that he should be ascribed with some responsibility for his injuries.

131 Before I turn to my reasons for the view that I have taken, it would be useful to begin by sketching out some of the general principles undergirding the law of contributory negligence. In the recent decision of the Court of Appeal in Cheng William v Allister Lim & Thrumurgan and another and another appeal [2015] 3 SLR 201 (“Cheng William”), writing for the court, I explained the nature of the defence of contributory negligence in the following terms at [13]:

132 The defence of contributory negligence finds statutory expression in Singapore in s 3 of the Contributory Negligence and Personal Injuries Act (Cap 54, 2002 Rev Ed) (“Contributory Negligence Act”), which reads:

133 As is apparent from s 3(1), both the fault of the claimant and the fault of the defendant are relevant. The term “fault” is in turn defined under s 2 of the Contributory Negligence Act as follows:

134 It is clear from this that the definition of “fault” is a broad one, and that it is not just an “act” that is caught; an omission would equally fall within the ambit of “fault” (Cheng William at [15]; Glanville Williams, Joint Torts and Contributory Negligence (Stevens & Sons Ltd, 1951) at p 354; K M Stanton, The Modern Law of Tort (Sweet & Maxwell, 1994) at p 104). The defence of contributory negligence must also be specifically pleaded under O 18 r 8 of the Rules of Court (Cap 322, R 5, 2014 Rev Ed). As was mentioned in Cheng William (at [16]), the underlying purpose of this requirement, like virtually all pleading requirements, is to ensure that the other side is not taken by surprise.

135 The requirement that the defence of contributory negligence be specifically pleaded is amply met in this case. Indeed, right from the outset, the appellant had made it clear in her pleadings that her assertion of contributory negligence on the part of the respondent centred on his alleged failure to keep a lookout for oncoming traffic before he started to cross the second half of the road. This very issue was ventilated in the appellant’s opening statement and closing submissions before the High Court Judge (“the Judge”). I also accept in principle, that a failure to keep a proper lookout for oncoming traffic can amount to contributory negligence as it is an omission and therefore falls within the ambit of “fault” under s 2 of the Contributory Negligence Act.

136 To this point, I see no divergence between the majority Judgment and my own views. I note the majority’s view that in principle the defence of contributory negligence is not precluded simply because pedestrians have a right of way (see [36]–[37] and [129(b)] above). In my judgment, whether the defence operates must very much depend on the circumstances of each case; and it thus necessitates due consideration of whether there was indeed fault on the part of the pedestrian.

137 It is also important to note that in considering contributory negligence, the court’s attention is directed not at the conduct of the defendant tortfeasor but rather at the conduct of the victim. It is the victim’s failure to take preventive or protective measures that is the subject of this inquiry. In Nance v British Columbia Electric Railway Co Ltd [1951] AC 601, Lord Simon put it in these terms at 611:

138 It is suggested on this basis that the court’s attention should therefore be directed solely at the victim’s conduct (see eg, Charlesworth & Percy on Negligence (Sweet & Maxwell, 13th Ed, 2014) at para 4-03).

139 While I agree that the crux of the case where contributory negligence is in issue is the conduct of the victim, it would be artificial and, in my judgment, wrong to view this in isolation. It is of course correct that every road user has a responsibility to take measures to prevent accidents or limit injuries that arise when accidents do occur. But in my judgment, the nature, extent and degree of that responsibility calls for a fact-sensitive inquiry that must have regard separately to the foreseeability of the risk that has materialised as well as the viability of measures that could be taken to guard against that risk.

140 It seems to me that herein lies the nub of the difference between the majority’s view and my own. In my judgment, it is incumbent upon us, when we undertake the inquiry into whether a victim’s failure to take certain measures should result in a diminution of the tortfeasor’s liability, to have due regard to the nature of the risk that has materialised and whether that is a risk in respect of which the victim had a continuing duty to take steps to guard against. In my judgment, the point is encapsulated in the fact that the victim’s responsibility is to take reasonable care and what is reasonable depends as much on the likelihood or foreseeability of the risk eventuating as it does on the availability of preventive measures to the plaintiff. To put it another way, the touchstone is whether the claimant acted reasonably, with the amount of self-care that a normal person would have exercised in the circumstances of a given case (see eg, Cheong Ghin Fah and another v Murugian s/o Rangasamy [2004] 1 SLR(R) 628 at [83]; Roda Sam (previously known as Rawdah al-Sam) v Pascale Atkins [2005] EWCA Civ 1452; Smith v Finch [2009] EWHC 53 (QB) at [43]; A C Billings & Sons Ltd v Riden [1958] 1 AC 240; Lewis v Denye [1939] 1 KB 540; Simon Deakin, Angus Johnston & Basil Markesinis, Markesinis and Deakin’s Tort Law (Oxford University Press, 7th Ed, 2013) at pp 754–755; John Murphy & Christian Witting, Street on Torts (Oxford University Press, 13th Ed, 2012) at p 111; Michael A Jones (gen ed), Clerk & Lindsell on Torts (Sweet & Maxwell, 21st Ed, 2014) at paras 3-68 and 3-70). Because such an enquiry is necessarily a fact-sensitive one, the court must have regard to the entirety of the circumstances (per Tasker Watkins J in Owens v Brimmell [1977] 1 QB 859 at 867).

141 The point has relevance in that it qualifies the absolute nature of the proposition suggested elsewhere that the focus is on the victim’s conduct solely (see [138] above). In my judgment, this is true in the sense that the only question to be asked is whether the victim should have done something or forborne from doing something that may have avoided the accident or limited his injuries; but I also consider that this question cannot be considered in isolation from the nature of the defendant’s conduct. The more egregious the defendant’s conduct, the less likely it will be. At least in part, what makes it egregious is the fact that it is uncommonly bad. To focus on the measures the victim could have taken to prevent the accident, to the exclusion of how egregious the conduct of the defendant tortfeasor was in the circumstances would, in my view entail missing a step in the reasoning: namely, was it reasonable to expect the victim to guard against the particular type of conduct that caused the damage. It is only when that question is answered in the affirmative, that one then goes on to consider whether protective or preventive steps were available and should reasonably have been taken.

142 In the present appeal, the critical facts for the purposes of the inquiry into contributory negligence are these:

(a) The appellant was oblivious to the traffic lights. Her evidence is that she was completely unaware that the lights were against her. Indeed, the fact that she did not even slow down before the collision bears this out.

(b) The respondent evidently waited until the lights were in his favour before he commenced crossing. From his perspective, he had crossed more than half the crossing as he entered the second half of the dual carriageway.

(c) Throughout the crossing, the light remained in the respondent’s favour. From his perspective, he was entitled to cross throughout this time.

(d) The respondent was not crossing at an intersection and so had no reason to be conscious of or to guard against the possibility of unexpected traffic that was turning from another direction into the path of the crossing.

(e) The only source of traffic that the respondent was and needed to be concerned with was that flowing in either direction of the road that he was crossing.

(f) The lights controlling the pedestrian crossing operated simultaneously and contemporaneously on both sides of the road. Hence, the respondent was entitled to proceed on the basis that for so long as the lights remained in his favour, they were against the vehicular traffic that ran in both directions on either side of the dual carriageway.

(g) The respondent had completed crossing more than half the dual carriageway by the time of the collision. This must have taken several seconds. Although there was no evidence on this, it seems implausible to me, assuming the respondent had covered approximately 8m by the time of the accident that the lights had been in his favour for anything significantly less than eight to ten seconds. He would therefore have been aware that the lights stopping vehicular traffic had been in place for a correspondingly long period.

143 In these circumstances, the risk that materialised and which we are obliged to assess is whether it was reasonable for the respondent to be expected to guard against a driver, not so much “beating” a traffic light but being wholly indifferent to it and driving as if there was no traffic light and no controlled pedestrian crossing there at all.

144 In my judgment, this is a critical and weighty consideration that features heavily in the determination of whether the respondent should be held contributorily negligent. The collision in this case occurred because as far as the appellant was concerned there was no traffic light or controlled pedestrian crossing in place at all. The question then is whether that is a risk in respect of which it is reasonable to expect that the respondent should guard against, several seconds into crossing a traffic light controlled junction with the lights in his favour throughout this time.

145 In this context, I find myself in agreement with what the Judge held are two important implications that arise from this undisputed fact (Li Jianlin v Asnah bte Ab Rahman [2014] SGHC 198 at [5]):

146 Even though pedestrian crossings are not, as the majority puts it, “fool-proof” safety measures and cannot serve to completely eliminate risks posed to pedestrians by vehicular traffic (see [37]–[38] and [129(c)] above), I consider it important to bear in mind that ultimately, virtually every activity carries with it some degree of risk. The law of negligence entails a balancing of interests that is designed to avoid harm and this balance is struck by imposing a duty to take reasonable care to avoid causing harm whether to others or to oneself. An unduly wide conception of the risks against which one must guard oneself can have a paralysing effect. To avoid this, it is against risks that are reasonable to anticipate that one is obliged to guard oneself against.

147 In my judgment, the duty to guard against the risk of being knocked down by a motorist at a pedestrian crossing that is controlled by traffic lights is greatly attenuated by the passage of time after the lights have turned in favour of the pedestrian and against vehicular traffic. I base this on two premises, which I consider arise in such circumstances:

(a) A pedestrian, in my judgment, may harbour a legitimate expectation when crossing at a controlled pedestrian crossing that other road users (in this case vehicular traffic) will see the traffic light. It follows from this that I do not consider that a pedestrian is required to guard against the specific risk that a motorist will be wholly oblivious to the existence of a traffic light. To hold otherwise would mean, in effect, that a pedestrian must always act as if he were crossing a road that is uncontrolled by traffic lights and that is untenable in my view. Pedestrian crossings are meant to be safe havens in which pedestrians can cross roads without having to pay undue attention to vehicular traffic (Negligence at Pedestrian Crossings at 239).

(b) Because the pedestrian is entitled in my view to expect that other road users in particular motorists, are aware of and perceive the presence of traffic lights, with the passage of time, he is entitled to assume that these road users will have stopped. Any other conclusion would mean that a pedestrian has a continuing duty to guard against a motorist who notices a traffic light that is red against him but then intentionally chooses to disregard it and I find that, again, untenable. Indeed, it is a matter of importance that the majority too accepts at least the latter proposition (see [58(c)] above).

148 In my judgment, cases which suggest that claimants must be alive to the possibility of carelessness on the part of potential tortfeasors are not inconsistent with what I have stated thus far. The more common such carelessness is, the greater will be the duty to guard against it. Indeed, as Lord du Parcq in Grant v Sun Shipping Co [1948] AC 549 (“Grant”) at 567 so pithily puts it:

149 This was echoed by Denning LJ (as he then was) in Jones v Livox Quarries Ltd [1952] 2 QB 608 (“Jones”) at 615 in the following terms:

150 In my judgment, these extracts make the same point in slightly different ways. As Denning LJ put it in Jones, the prudent pedestrian must take into account the possibility of others being “careless” and this is correct because paraphrasing Lord du Parcq’s words in Grant, such carelessness may be thought to be somewhat commonplace. But the conduct of the appellant in this case went well beyond carelessness. Indeed, it went even beyond recklessness and was downright dangerous. The majority Judgment as I have noted, appears to accept that the position might have been different if the appellant had been acting intentionally. Hence, if the appellant had seen the lights and intentionally decided to beat it and then knocked the respondent, there would be no reduction of the latter’s damages on account of contributory negligence; but yet, it is said that because the present appellant was apparently oblivious to the lights, a different outcome should follow. With respect, given that in the context of contributory negligence, our focus is on the risks that the plaintiff ought in all the circumstances to have safeguarded himself against, it is difficult to see why the result would be different if the tortfeasor was acting dangerously as in this case but not intentionally as in the alternative hypothesis set out above. It should be noted that at 55 km/h (being the appellant’s admitted speed) and 68 km/h (being the high end of her own expert’s estimate) the traffic lights would have been visible to the appellant for between 8 and 10 seconds prior to the collision. To remain oblivious to the lights for that length of time goes far beyond anything that could fairly be described as a lapse. In my judgment, it is material that the appellant’s conduct was so egregious that it was dangerous to other road users. I find it unsatisfactory that a road user who is acting entirely within and in accordance with the law should be subjected to a duty to guard against the dangerous (and not merely careless) conduct of others and this, in my judgement, goes beyond what was contemplated in either Grant or in Jones.

151 Perhaps, the foregoing analysis might be impacted by empirical evidence suggesting the frequency of such conduct. I turn here to the various statistics relied on in the majority Judgment, which, in some senses, demonstrate that the foreseeability of a driver beating a red light is a real possibility and not just a fanciful one. These statistics emanate from the Traffic Police or from the Parliamentary debates and were not cited by either party. For instance, the then Minister for Home Affairs Mr Wong Kan Seng said in Parliament that between 2004 and 2008, about 2,100 accidents occurred yearly at road intersections and of these, about 330 or 16% happened because the motorist in those cases disregarded a red light signal. During that same period, the traffic police issued about 19,700 summonses annually for motorists who flouted a red light signal (see Singapore Parliamentary Debates Official Report (14 September 2009) vol 86 at cols 1423–1424) (see [31] above).

152 The majority judgment also makes reference at [32] to other reports suggesting that the number of traffic violations involving beating traffic signals has been on the rise. These statistics do indicate that the possibility of a driver running a red light is a real one that cannot be lightly brushed aside. But, in my judgment, they do not provide a sufficient basis for concluding that the possibility of a pedestrian being knocked down in the circumstances of this case is a real or reasonably foreseeable one. I return here to the observations I have made at [142(a)], [143] and [144] above where I make the point that the conduct in question of the appellant in this case is not that she “beat” the lights but that she wholly failed to apprehend that there were any lights at all and so drove past a red light eight or so seconds after it had turned against her. With great respect, and as the majority acknowledges (see [33] above), the statistics tells us nothing about the frequency of this type of collision because they are simply not sufficiently granular and wholly fail to differentiate between the various types of situations where a motorist unlawfully drives through a red light. Indeed, what the statistics, which are mentioned in the previous paragraphs, indicate is that out of an annual average of 19,700 summonses issued against motorists who flouted red lights between 2004 and 2008, just 330 or 1.6% resulted in accidents. These accidents would include a wide range of types of collisions of which those involving cars and pedestrians would involve a fraction. And of these, the ones involving collisions with pedestrians crossing with the lights already in their favour for some time must involve a miniscule fraction of that. While it is true that we do not have the empirical data to prove this, we should be reminded first that the burden of proof is on the appellant; and second, that the utter lack of precedents involving facts that are remotely similar to those before us suggests that the incidence of this type of fact pattern is much rarer than it is suggested to be.

153 The common experience suggests that a motorist might try and fail to beat the light before it turns red. This may result in the motorist crossing the road a second (or two or three) after the lights change. That is a risk that is sufficiently common that it seems entirely reasonable to impose a duty upon a pedestrian to guard against. But that is of a wholly different nature than the risk in this case, which strikes me, as a much rarer occurrence for the reasons that I have just set out. It is also a different type of risk altogether and as I have noted at [147] above, if a pedestrian were held to have a duty to guard himself against this type of risk, it must mean that a pedestrian would generally be obliged to act as if he were crossing at an uncontrolled location even when it is a controlled crossing and as I have already noted, I find that untenable. While I agree with the majority that the court must strive to make full sense of these statistics as far as possible and that it is conceivable that they would include situations where errant motorists beat traffic lights even though they have been against them for a considerable period of time, the statistics, in my judgment, are of limited utility because if anything, they suggest that incidents of this sort are indeed exceedingly rare. They certainly do not support the conclusion that the risk that is presented in the present case is sufficiently common such that the respondent here came under a duty to guard against it. The statistics therefore do not persuade me to come to a different view than I have.

154 As to the forum letter and the newspaper articles (see [34]–[35] above), I make the following brief points. First, the events painted there are at best anecdotal, and perhaps even isolated incidents. Second, both the forum letter and the newspaper article referred to in [35(a)] do not make clear the precise circumstances in which those incidents happened. Both the writers of the forum letter and that newspaper article, for instance, did not make clear whether the traffic lights had been against vehicles for a considerable period of time, or whether they involved instances where the motorists in question had attempted to rush past traffic lights that had just turned red. Third, the accident reported in the newspaper article referred to in [35(b)] happened in entirely different circumstances from those in the present case because the errant driver drove with faulty brakes. Hence, with great respect, I do not consider that these sources can add anything useful to the determination of the issues that are before us.

155 As to the majority’s holding that r 22 of the Highway Code implicitly recognises that the possibility of a motorist running the red light is a real one, I do not accept that this provision advances the analysis in the present case. Rule 22 on its express terms exhorts pedestrians to wait on footways until traffic, in front of which they intend to cross, has come to a standstill. This rule, as applied to the present case, enjoins pedestrians such as the respondent to check to ensure vehicles have come to a standstill before entering a pedestrian crossing (since there is just one footway from which the respondent started crossing from, and as will be explained later, there was no footway in the road’s centre). Indeed, it appears to me to be a rule that is targeted at a situation where the traffic light has just turned red (as opposed to one where it has turned red for a considerable period of time). This would be an entirely sensible rule for such a situation because, as I have already observed, the risk of a car rushing past a red light that has just turned red is a sufficiently common occurrence that pedestrians should be expected to guard against. But beyond that, r 22 can hardly be said to have any application and cannot, in my judgment, be invoked as a basis for holding that the possibility of a pedestrian being knocked down a considerable time after the traffic lights had turned red, as was the case here, is real one.

156 I note the majority’s observation that Parliament could well have included such words as “when the green man appears” at the beginning of r 22 if the legislative intention had been to limit the application of this rule to those situations where the traffic light has turned red against traffic for just a short while. But I do not find this persuasive for a few reasons. First, the words used speak of a footway which is commonly found at the edge of a road. Aside from this, there are no parliamentary materials as far as I am aware which supports the majority’s point. Finally, it is equally plausible that Parliament did not include words such as “when the green man appears” precisely because r 22 was to apply in those situations where there are centre dividers with footways, in which case a pedestrian should wait and check once more before making the second half of the pedestrian crossing. That, however, is not the case before us.

157 To overcome this, the majority relies on r 20 to suggest that the present crossing should be treated as two separate crossings. While I agree with the majority that in principle, pedestrian crossings that are in fact divided by centre refuges or islands can be treated as two separate crossings such that pedestrians should be expected to check before embarking on crossing the second half of the pedestrian crossing, I do not think that there was anything akin to a centre refuge here. It is clear that centre refuges or islands must be sufficiently wide so as to afford pedestrians a safe place to stop and look before continuing to cross. I illustrate this with the following photograph from the Land Transport Authority’s (“the LTA”) website (see http://www.lta.gov.sg/apps/news/page.aspx?c=2&id=59d4005d-c2e3-4c17-9ee6-06daab5cfc44> (last accessed: 8 January 2016)):

158 In a similar vein, I have noted that the LTA, in its Standard Detail of Road Elements Handbook (revised in 2002), requires centre refuges to be at least 1.2https://www.lta.gov.sg/content/dam/ltaweb/corp/Industry/files/CLTA020218.pdf> (last accessed: 8 January 2016):

159 There is nothing like this in the present case. I therefore do not find it possible to hold that the present crossing is anything other than a single crossing. The majority bases its holding primarily on its conclusion that it seemed possible that a few people could stop at the centre of the crossing safely to check before embarking on the second half of the pedestrian crossing. Having reviewed the photographic evidence, I am unable to agree. Indeed, the width of the road divider was so small that it would barely cover a single person’s width. To illustrate this, I reproduce a photograph from one of the expert reports:

160 I also provide a photograph of the crossing from the pedestrian’s vantage point:

161 The majority contends in the alternative that r 22 is nonetheless implicated because the respondent ought not to have started crossing until he was satisfied that the traffic he intended to cross in front of had come to stop. But, with respect, this overlooks the fact that if the respondent started crossing eight or so seconds before the collision, then using a speed of either 55 km/h (admitted by the appellant) or 68 km/h (the upper limit suggested by the appellant’s expert), the appellant’s taxi would have been between 122.2m and 151.1m away. To suggest that a pedestrian cannot start crossing when a vehicle is noticed to be that distance away, would make life unbearable for pedestrians. Hence, I cannot accept that either rr 20 or 22 of the Highway Code have any relevance to this case. Moreover, r 22 does not apply because there is no footway in the middle of the road and because as mentioned, the width of the road divider is too narrow and can barely cover a person’s width.

162 I also consider that my view is supported by the cases which the respondent, through his counsel, Mr Liew, has placed reliance on. Broadly speaking, these are cases where the courts have refused to find contributory negligence in circumstances where the plaintiff pedestrians have already crossed a significant portion of the pedestrian crossing, and therefore had the lights in their favour for a considerable period of time at the time of the collision.

163 I do not agree with majority’s basis for distinguishing these cases for the following reasons:

(a)  First, it cannot be said that the distance covered by the respondent before he was knocked down was not significant. Indeed, the respondent had already completed crossing more than half the dual carriageway by the time of the collision and although there is no evidence on this, as I have already explained, seems implausible to me, assuming the respondent had covered approximately 8m by the time of the accident that the lights had been in his favour for anything significantly less than eight to ten seconds (see [142(g)] above). This distance also appears to be longer that covered by the pedestrian in Frank v Cox, who was approaching the centre refuge when he was knocked down (see [164]–[165] below for the facts).

(b) Second, even though these cases do not appear to involve fences, I find this distinction to be immaterial for the purposes of the present appeal because the metal fence in question consisted of bars with spaces and gaps between them such that either party’s view is unlikely to have been substantially obstructed, if at all.

(c) Third, I also make one further point in relation to the supposition in the majority Judgment that r 18 of the UK Highway Code could have impacted the result reached in those cases. It is hard to see just how this provision would have affected the result because it does not even apply in a case such as the present to begin with. This is because that rule deals with the situation where a pedestrian is about to step on to a pedestrian crossing, in which it would be entirely sensible for him to check before crossing for the reasons that I have already canvassed. It has no application to a pedestrian who starts to cross when it is reasonable to do so and is then struck down, which happens also to be the case here. It would therefore not be surprising if that provision was not considered, let alone not mentioned in those decisions.

164 I turn now to discuss the cases Mr Liew relies on. The first of these is Frank v Cox. There, the claimant, a 73-year-old pedestrian, wished to cross King William Street at the intersection of Arthur Street and Monument Street in the city of London. There were traffic lights at the intersection and the plaintiff waited until they had turned red against vehicular traffic for some time to allow for cars trying to beat the changing light before he began to cross the road. He had almost reached the centre refuge in the road when the lights changed. The defendant’s car turned right from Arthur Street and knocked the claimant down, causing him severe injuries. The claimant sued for damages and the trial judge held that the defendant was solely to blame. The defendant appealed to the Court of Appeal and made two arguments on appeal. The first of these need not concern us as it had to do with a specific issue of whether the award was excessive as the claimant was of an advanced age and had a life expectancy of only six years.

165 The second argument is of direct relevance to this appeal and it concerned the issue of whether contributory negligence on the part of the claimant was made out by reason of the his failure to keep a look-out for the defendant’s car. This second argument was rejected by Sachs LJ (with Danckwerts and Sellers LJJ concurring), who held that there was no contributory negligence. The claimant had already crossed a significant portion of the road and indeed was approaching the centre refuge in that case when he was knocked down. Sachs LJ said:

(a) There was a paramount duty on motorists turning at junctions controlled by lights to give precedence to pedestrians who, having alighted when the lights were in their favour, were still crossing when the lights changed.

(b) The defendant’s driving was not only “foolish”, it “verged on the monstrous”. Although pedestrians had a duty to guard against careless acts of driving, they were under no duty to safeguard themselves against monstrous driving.

166 In my judgment, Frank v Cox is relevant in this case on both points noted by Sachs LJ. As to the first, the position in the present case is even stronger because not only had the respondent commenced crossing with the lights in his favour, it remained so up to the time of the collision. As to the second point, the appellant’s conduct was no less “monstrous” than that of the appellant in Frank and the view expressed by Sachs LJ accords with my own view expressed at [143]–[147] and [152]–[161] above. I also find it significant that the position in Frank is precisely in line with r 7 of the Pedestrian Crossing Rules which states:

167 I find it impossible to reconcile r 7 with the implicit view of the majority that if a police officer had signalled traffic to stop and permitted a pedestrian to cross and if a drunk or even a reckless driver failed to notice the police officer and knocked down the pedestrian, the latter’s damages could be reduced. How would this make sense to the pedestrian, who had simply followed the directions of the police officer? Is it suggested then that a different rule should apply to junctions controlled by a police officer as opposed to those controlled by a set of lights though r 7 gives not the slightest hint of any such intent?

168 The next case is Tremayne v Hill, where the claimant pedestrian was crossing diagonally at night over two roads that converged at a busy junction controlled by traffic lights. Just before the claimant had completed the crossing he was struck down and injured by a car driven by the defendant that had crossed, not at a fast speed, into the junction against a red traffic light. The claimant was aware when he made the crossing that the traffic lights from the direction in which the defendant’s car had been travelling were red. At trial, the defendant raised a defence of contributory negligence that rested, among other things, on the argument that the claimant failed to keep a proper lookout. This was rejected by the trial judge, Macpherson J, who concluded that the accident had been caused wholly by the defendant’s negligence. The English Court of Appeal dismissed the appeal brought by the defendant and agreed with Macpherson J’s holding. Sir Roger Ormrod, who delivered the lead judgment of the English Court of Appeal, held that the claimant was justified in assuming that traffic would have stopped as the traffic light had been red for some time and that he had no reason to suppose that vehicles in those circumstances would ignore the traffic light. He said as follows (at 134J–135B).

169 With respect, even if the facts may be slightly different, this part of the court’s reasoning is exactly on point.

170 The last case that I find useful is White v Saxton. Even though we have before us only a summary of that from Paul J Taylor, Bingham and Berryman’s Personal Injury and Motor Claims Cases (LexisNexis, 13th Ed, 2010) which was tendered by Mr Liew (see [102] above for the excerpt of that case), the facts, holding and essential import of that case are sufficiently clear in my view and it therefore should not be disregarded. The majority cites the principle that the court should be slow to apply a precedent where little if anything is known of the reasons of the court. That however is not the case here. We may not have the full report but we do have a digest of the report which includes the essential reasoning of the court. The facts in the case are that the claimant pedestrian, who was 16 years old at the material time, attempted to cross a single carriageway (with two lanes, one for each opposing direction) using the designated pedestrian crossing. She pressed the button and waited for the traffic light to change to red against the vehicular traffic and for the green man to appear. This duly transpired and a car travelling from her left stopped at the pedestrian crossing. The claimant started to cross the road, and while doing so, noticed some headlights coming from her right. She was struck by this vehicle and suffered injury as a result. The defendant driver conceded liability and at trial, the only issue that the trial judge had to deal with was the question of contributory negligence. It was held that the claimant was not contributorily negligent because when she commenced her crossing there was either no traffic to her right, or if there was a vehicle approaching she was correct to assume that the vehicle would see the traffic lights and would come to a halt. The court therefore held that it was reasonable for the claimant to assume that it was safe to cross when the lights permitted her to do so in the circumstances. In my judgment, this reasoning accords entirely with the observations I have already made and referred to above.

171 Further, our courts have found pedestrians liable in contributory negligence in spite of using pedestrian crossings cases only because there was clear fault on the part of those pedestrians. This is borne out by the following four local cases. The first is Yip Kok Meng Calvin (a minor) v Lek Yong Han (Yip Ai Puay, third party) [1993] 1 SLR(R) 147, where contributory negligence was found and the plaintiff had his damages reduced by 30% for dashing across the pedestrian crossing at a road junction without looking when the lights had just changed in his favour (at [5]). Even though the plaintiff was only nine years old at the time of the accident, the court found that he was mature enough to cross roads on his own and that he had used that pedestrian crossing to reach home from school on numerous occasions; he also knew the importance of keeping a lookout before crossing because his parents had impressed that upon him (at [5] and [10]). Similarly, the court in Khoo Bee Keong v Ang Chun Hong and another [2005] SGHC 128, reduced the damages payable to the plaintiff by 20% because he had not checked for oncoming traffic in the first lane (the lane nearest to him when he started to cross) and was content to check only for oncoming traffic in the next two lanes. Here too, he had started crossing when the lights had just changed in his favour (at [9] and [12]–[14]). And the courts in both Kim Anseok and another (personal representative of the estate of Kim Miseon, deceased) v Shi Sool Hee [2010] SGHC 124 and Ng Weng Cheong v Soh Oh Loo and another [1993] 1 SLR(R) 532 found contributory negligence on the part of the plaintiff pedestrians who had crossed at pedestrian crossings when the lights were against them.

172 As against this, I consider that the few other authorities that have been referred to us can be distinguished. I turn first to the decision of the Hong Kong High Court in Chun Sung Yong (see [83]–[84] above for the facts). There, a pedestrian walked across Nathan Road at the junction of Kimberly Road through a pedestrian crossing when he was knocked down by a car. Nathan Road was a dual-carriageway, with two lanes on each side for traffic in opposing directions and a central refuge in the middle. The pedestrian had crossed about 9.6m or more than half of the width of the dual carriageway with the crossing lights in his favour when he was knocked down in the third lane and was found to have been contributorily negligent. Although the judge accepted the plaintiff’s evidence that the green man was showing in his favour when he commenced crossing the second half of the dual carriageway and that the defendant driver had acted in reckless disregard for the safety of the pedestrian plaintiff, he also held that the plaintiff had failed to stop at the road’s centre refuge, and was therefore in breach of Regulation 33(5) of Hong Kong’s Road Traffic (Traffic Control) Regulations (Cap 374G) which provides that “[f]or the purpose of this regulation the parts of a road on each side of a pedestrian refuge or central reservation shall each be treated as a separate carriageway”. This is worded similarly to Singapore’s equivalent in r 20 of our Highway Code. The judge found contributory negligence on this basis to the tune of 10%. The judge also found contributory negligence to the tune of 15% on the basis that the plaintiff ought to have kept a lookout for oncoming traffic at the centre refuge. In the present appeal however, there was no such centre refuge and as mentioned earlier, it cannot be said by any measure that there was anything akin to a centre refuge such that it would be reasonable to expect a pedestrian to stop before embarking on the second half of the pedestrian crossing.

173 The next case is Wilkinson, the facts of which have already been highlighted in the majority Judgment (see [80] above). Although the pedestrian there was knocked down some steps after the road’s centre and contributory negligence was found, that was because the pedestrian carried on walking past the road’s centre refuge when the pedestrian crossing lights had already changed against him. That, in my judgment takes the case out of the ambit of the principles that I have discussed. In my judgment, the finding of contributory negligence was entirely justifiable because the pedestrian had gone ahead and started the second half of the crossing when the lights were against him. This is precisely the context in which it makes sense to say the two halves of the dual carriageway should be treated as two separate roads and in that light, it was as if the pedestrian had started to cross at a pedestrian crossing when the lights were against him. In fact, he could and should have stopped and waited at the central refuge for the crossing lights to turn in his favour again. Those are not the facts before us.

174 I now discuss briefly, the other features of the road at the time of the accident which the majority considers necessitated greater care on the part of the respondent before he continued into the second half of the pedestrian crossing. To recapitulate, these features are: (1) the road curvature some 150m away before the pedestrian crossing from the direction that the appellant was coming such that the pedestrian would not be able to see such traffic in the second half of the dual carriageway when he commenced crossing the road; (2) the presence of a 1.4m high metal fence on the road divider which, it was said, would obstruct the vision of both pedestrians (in respect of oncoming traffic) and of motorists (in respect of pedestrians stepping off into the second half of the road from the gap in the road divider); and (3) the poor visibility of approaching vehicles, because the accident happened at night at around 10pm (see [93] above).

175 With respect, I do not consider that these features bear weight in the present appeal. In the first place, if there was no duty on the respondent to guard against the risk of motorists who would be oblivious to the presence of traffic lights, then this cannot change by reason of these considerations. This remains my central point of departure from the majority in this case for the reasons set out at [147]–[150] above. But there are also more specific objections:

(a) First, although the collision occurred at night, it is a fact that the road was well-lit and the weather was fine at the time of the accident. Hence, any compromise of visibility would have been marginal and there was no evidence to suggest otherwise. Notably the collision occurred because the appellant never saw the traffic light against her and this had nothing to do with poor night visibility.

(b) Second, as for the curvature of the road, it is first unclear to me that a pedestrian starting to cross the road would have realised there was such a curvature by reason of which he needed to take special care. Moreover, 150m is a considerable distance. If a car was approaching the junction at a speed of 50 km/h it would take about 11 seconds to cover that distance assuming there was no reduction in speed at all. This is ample time to slow down provided the driver has seen the traffic light. And if the supposition is that the driver, as is the case here, has not seen the light, then it is irrelevant whether there was a curvature or not; and moreover, I reiterate, this is not a risk that a pedestrian should be responsible to guard against.

(c) Lastly, as for the metal fence on the road divider, essentially the same points can be made. Moreover, the photographs taken at the scene of the accident (see [94] and [96] above) are of a poor resolution, and therefore cannot be said to capture accurately either party’s view. It is not clear to me that either party’s view would have been substantially obstructed, if at all, because the metal fence in question consisted of bars with large spaces and gaps between them.

176 I turn now to discuss whether it was “evident” that the respondent did not check for vehicular traffic at the road’s centre, and whether the appellant’s counsel should be permitted to raise this point to begin with in the light of the rule in Browne v Dunn. Here, I proceed on the assumption that the respondent had a duty to check, which as I have said, I do not accept.

177 I am prepared to accept that the rule in Browne v Dunn may not apply here because putting anything to the respondent would have been pointless given the injurious effects of the accident. But I am unable to see how it can safely be said that it was “evident” that the respondent did not check for vehicular traffic at the road’s centre. The majority bases this finding on the inference that if the respondent had checked, it would have been “apparent” that the taxi driven by the appellant was not going to stop in good time, otherwise he would not have gone on to complete the second half of the crossing (see [114] above).

178 With respect, I consider that this position is impermissibly speculative. It should first be noted that the burden of proof on this falls on the appellant who asserts that her liability should be reduced on account of the respondent’s negligence.

179 The appellant herself cannot give any evidence as to whether the respondent did or did not check for oncoming traffic because she was oblivious to the lights and to the respondent until after she had run him down. As for the respondent, he cannot say anything because of the injuries he sustained as a result of the collision.

180 The majority’s view is that such negligence on the respondent’s part is made out by his failure to check. It is not suggested that the respondent had darted out of nowhere; on the contrary this was a national serviceman crossing the road at the end of the day and with the light in his favour for some time. The majority holds in the present circumstances that it may safely conclude that he did not check because if he had checked, the accident would not have happened. With respect, this seems to me to be objectionable, firstly because it seems to create a virtual rule that any pedestrian who is struck down at a pedestrian crossing will almost inevitably have his damages reduced on account of contributory negligence on the basis that the fact he was run down must mean that he did not check. This runs contrary to my view that contributory negligence is always a fact-sensitive exercise. Second, I cannot accept the majority’s view because there are at least two other possible explanations for the fact that the collision did take place, which do not necessitate the conclusion that the respondent did not check:

(a) The first is that the respondent perhaps did check by looking towards the direction of the appellant’s taxi shortly before he reached the half-way point, but failed to notice it. On this hypothesis, he would have glanced to his left two or three seconds before the collision. On the two hypotheses as to the appellant’s speed, the appellant would have been between 30.6m (two seconds at 55 km/h) and 56.7m (three seconds at 68 km/h). It seems reasonable to infer that he might have checked and yet not noticed the taxi. To hold the respondent contributorily negligent in these circumstances is without basis unless it is also suggested that his duty was not merely to check but to check with such diligence and care as would ensure that he would have seen the taxi.

(b) Second, he may have seen the headlights of the appellant’s taxi at those distances and fairly misjudged the speed at which the respondent was travelling. Such misjudgement would not have been out of place given that the appellant had travelled for eight to ten seconds without seeing the traffic light at all. Any reasonable person would have expected a taxi 30m to 57m away to be slowing down as it approached the traffic lights which were and had for some time been showing red.

181 In either of these scenarios, I do not see how it can be said that there would nonetheless be contributory negligence. Even if there was a duty to check by looking up, this cannot extend to responsibility for taking steps to avoid anything but the most obvious of risks. Given that the appellant’s taxi could have been anything from 30m to 57m away at the time the respondent might have checked, it cannot safely be concluded from the fact that the collision did take place that he did not check or even that he must be held culpable for failing to appreciate the risk since it is not clear that it would have been such a risk that was sufficiently obvious.

182 It bears reiteration that the burden of proving that the respondent did not in fact check lies squarely on the appellant, and in the light of the other equally plausible accounts of what might have happened, I take the view that the appellant has in any event failed to discharge her burden in demonstrating that the respondent did not check for vehicular traffic as he approached the centre of the crossing.

183 Finally, I have reservations over any suggestion that tort law places a duty on people to prevent risks from materialising in every given scenario and the point alluded to in the majority Judgment that the respondent here is akin to a least cost-avoider and could have acted to prevent loss or injury with minimal cost and effort because all that he had to do was to keep a look out for oncoming traffic before attempting to cross the second half of the pedestrian crossing (see [59] above).

184 First, any suggestion that tort law expects people to prevent risks from materialising in every given situation would place an undue burden on parties; moreover this is a wholly unrealistic standard because virtually every activity carries with it some degree of risk. There is no proposition in law that parties are expected to prevent risks from materialising in every situation. As I have discussed earlier, to focus on the measures that a victim could have taken to prevent the accident, to the exclusion of how egregious the conduct of the defendant tortfeasor was, entails missing a step in the analysis, namely: whether it was reasonable to expect the victim to guard against the particular type of conduct that has caused the damage to begin with. It is only when that question is answered in the affirmative, that one then goes on to consider whether protective or preventive steps were available and should reasonably have been taken.

185 In any event, it is not clear to me how the respondent can be said to be the least-cost avoider when compared to the appellant. All the appellant had to do was to look out for traffic lights – perhaps the most basic of duties cast on a driver. It is the driver who has the potential to cause catastrophic injury and to suggest that as between the driver and the pedestrian, the pedestrian is the least-cost avoider seems to me to be troubling, to say the least.

186 The majority’s decision today goes further than any other decision from any jurisdiction anywhere in the world that has been brought to our attention, in finding the pedestrian contributorily negligent in circumstances where:

(a) the pedestrian acted safely when he commenced crossing;

(b) there was no island or other break in the crossing;

(c) the vehicle which collided into him was oblivious to the traffic light for between eight and ten seconds and may have been as much as 150m away when he started crossing;

(d) the pedestrian was acting throughout in accordance with the law;

(e) the collision occurred eight or so (or more) seconds after the lights had changed; and

(f) there is no evidence at all as to whether the respondent did or did not check before he reached the half-way point of the crossing in part because the respondent’s inability to recall the details of the accident due to the injuries caused by the appellant’s appalling driving.

187 I am, with great respect, unable to agree with the majority that this is a development we should sanction. The result of the ruling today is that pedestrians will no longer be able to take comfort in the fact that they are crossing at a point controlled by a police officer or by traffic lights. They will have to safeguard themselves in precisely the same manner in such circumstances as if they were jaywalking. I cannot go along with this and would hold that the respondent should not be held to have been contributorily negligent at all. Accordingly, I would dismiss the appeal with the usual consequential cost orders.

Sundaresh Menon    Chief Justice