Gordon Martin Pty Ltd v State Rail Authority of New South Wales & Anor

The collision 19 At trial, the parties gave contradictory versions of the events that occurred immediately prior to the collision. Both parties acknowledged that the liability issues arising on the claim and cross-claims fell to be determined to a significant extent upon which version of events was accepted by the trial judge.

Evidence of the witnesses 20 Mr Stackman, the driver of the truck, contended that he had stopped at a stop line marked on the road prior to entering upon the level crossing. He said he looked left and right and satisfied himself that no trains were approaching before proceeding to enter the crossing. He said that he did not hear the train horn sound until he was on the track, but when he did hear it, he attempted to accelerate as fast as he could, but was not able to completely clear the track. The driver's side window of his truck was partially open and the truck's radio was off. Mr Stackman said that there was nothing wrong with his hearing and that there would have been no difficulty in hearing the horn, if it was sounded, as he was coming down Carroll Road. At trial, Mr Stackman said that in addition to the pepper tree, there was another tree on the right hand side of the road closer to the railway track, which obscured his vision. He also said there was some long grass which also obscured his vision. His Honour rejected Mr Stackman's evidence in all critical respects. 21 The train driver, Mr Matthews, said that he first saw the truck travelling along the Kamilaroi Highway "probably just coming out of the cutting", which was about 450-500 m from the level crossing. At that time, the truck was just turning off the highway onto Carroll Road. Mr Matthews gave varying distances and times as to when he sounded the horn, and when he commenced to brake, and the speeds at which he was then travelling. It was apparent that Mr Matthews' assessment of such matters was inaccurate. However, the train's speed at various critical points and the time at which Mr Matthews sounded the horn and when he applied the brakes were established by information derived from the train's data logger. Mr Matthews said, and the appellant relied on this, that the appellant's truck was already on the railway track when he applied the brakes. Mr Matthews also said that usually, when approaching Breeza, he would maintain the speed of the train through the level crossing and commence to slow down at some stage after the crossing. 22 The trial judge accepted Mr Matthews as a reliable witness other than in respect of his assessment of distance and time, a task at which Mr Matthews acknowledged he was poor. Given that qualification and the fact that matters such as when Mr Matthews sounded the horn and applied the brakes were recorded on the train's data logger and thus could be objectively determined, his Honour accepted: "(a) That [Mr Matthews] saw the truck well prior to it reaching the crossing after it turned off the highway.

(b) He decided that he would give the truck driver plenty of warning and sounded the horn as recorded by the data logger (the evidence of witnesses on the train and others nearby also established that the horn sounded for a longer period than a normal signal on approach).

(c) The truck did not stop. It continued through the crossing.

(d) [Mr Matthews] first applied the brakes when the truck entered the crossing." (at [144]) 23 Mr Matthews' evidence was essentially supported by the other lay evidence. Beverley Lawlor lived about 200 m southwest of the crossing. When the accident occurred, she was two doors away from her own home, visiting her daughter-in-law, Trina Gorton. They were on the front verandah of Ms Gorton's house, where Ms Lawlor said she had a clear view down Carroll Road to the crossing. She could only see east of the line as far as the signal boxes, which were located 50-60 m from the level crossing. 24 Ms Lawlor said that usually when trains approached the crossing, she could see them when they sounded their horns. On this occasion, she could hear a train sounding its horn for longer than usual and that caused her to look to see what was happening. She could not see the train. As she described it, the train horn "blew down the line further" than usual and the length of the sounding was "unusual". However, she said that when she "looked down" the road, she saw "Martin's truck come around the corner and … going towards the railway crossing". She explained that she saw the appellant's truck just as it was turning into Carroll Road and that it proceeded straight across the railway track without first stopping at the stop sign. She said that she heard the truck change gears as it went around the corner but she did not hear the truck's Jacob brake being applied (a sound with which she was familiar) as it proceeded down Carroll Road. She expressed the opinion that he was going "too fast to pull up". She said that she did not mean the truck was going "fast fast", but that it was "going faster than it should have been to pull up". She said she could still hear the train horn when the truck entered the crossing. 25 Ms Lawlor's daughter-in-law, Ms Gorton, had been concentrating on watching the children get off the school bus, and her observations in relation to the truck and train had not been as acute as Ms Lawlor's. She said that saw the truck on the level crossing just before she heard the sound of the collision. Ms Lawlor's evidence confirmed that the level crossing was observable from her home. 26 A passenger, Mr Ryall, said that he heard the train sounding its horn for about 11-12 seconds prior to the crash. As his Honour noted, this was remarkably accurate, corresponding with the data from the train's data logger. When the train first sounded its horn, Mr Ryall looked to see what was happening and saw a truck heading for the crossing and disappear in front of the train. 27 Mr Southam was the Passenger Service Supervisor on the train. He was checking his manifest in the rear guard's compartment at the back right hand side of the train, when he heard the horn blow for an "excessively long" period, "eight or nine, 10 seconds". He looked out the window and saw the back end of a semi-trailer. He closed the window and braced himself in preparation for a collision. He said that he anticipated a crash because the driver was still blowing the horn. He denied that the truck was on the track at this point, but said that it was sufficiently close for him to form the view that the truck was going to end up in front of the train. 28 His Honour found these witnesses to be reliable, except where he noted that a particular witness' assessment of distance or time was not accurate. His Honour rejected Mr Stackman's evidence that he had stopped and looked in both directions before proceeding to cross the level crossing. Other evidence also established that there was no stop line at the time of the accident, and that it had been installed at the crossing at a later time. His Honour did not make any finding as to when Mr Stackman had first heard the train horn, but the effect of his acceptance of the evidence of Ms Lawlor and Ms Gorton was that the horn could be heard for at least 300 m from the level crossing, given that they heard it from about that distance. Mr Stackman estimated that the train was 300 m away when he first heard the horn, but said it could have been 100 m, or closer. 29 The appellant, in argument, initially suggested that Mrs Lawlor's evidence was to the effect that the truck was still on the highway when she first heard the horn sound. It conceded, ultimately, that her evidence and that of Mr Matthews was to the same effect, namely, that the truck was turning off the highway onto Carroll Road when Mr Matthews sounded the horn. From that point there was a distance of about 60 m to the railway tracks, which would have taken the truck approximately 15 seconds to reach, assuming that the truck was travelling at about 15 km per hour. However, as described later, these times, speeds and distances were best estimates only. What was known, however, was that the horn first sounded about 11.25-12 seconds before the collision at a distance of 285-310 m before the level crossing.

Evidence derived from the train's data logger 30 When the train came out of the cutting on the approach to the level crossing, the train's data logger recorded that it was travelling at 88-90 km per hour. The experts retained by the parties agreed that two seconds before the collision, the train was travelling at between 90-92 km per hour. The experts also agreed upon the following sequence in the actions of the train driver leading up to the point of the collision:

The train driver commenced sounding the train's horn 11.25-12 seconds before the collision;

At that time, the train was 285-310 m from the level crossing and the truck was rounding, or had just rounded, the corner onto Carroll Road;

The horn sounded continuously for 9.25-10.75 seconds;

When the truck reached the stop sign, the train was travelling at about 93 km per hour and was approximately 129 m or 5 seconds from the point where the collision occurred (at 93 km per hour, the train travelled approximately 25 m per second);

The train driver activated the emergency brake when the train was 99-102 m and 4 seconds from the point of collision (but subject to the qualification recorded by his Honour in [46] of his judgment, that the brakes were activated 0.2 seconds before the brake cylinder pressure began to rise (set out below)); and

The horn was switched off 1.25-2 seconds before the collision when the train was 30-50 m from the level crossing. 31 His Honour also recorded the agreement reached by the experts as to the following matters (some of which replicates the information already mentioned above): "[44] The parties set out in Exhibit D their agreement as to information derived from the data logger on train events and times. As discussed below, the agreed data was relied upon as providing an objective analysis for determining the most likely sequence of events. The following matters in relation to the movement of the train throttle, the sounding of the train horn, the application of the braking system on the train and the point of collision were agreed.

Time range Event 15:29:35 to 15:29:35,75 Throttle moves from one-third to idle 15:29:37 to 15:29:37,75 Horn begins to sound 15:29:45 Brake cylinder pressure begins to rise 15:29:47 to 15:29:47,75 Horn ceases to sound 15:29:49 Close to collision occurring

[45] As noted above, the parties also agreed that the horn sounded at 11.25 to 12 seconds prior to the collision, and that the brake cylinder pressure began to rise 4 seconds prior to the time of collision. The evidence of the train driver established that on sounding the horn, strobe lights on the roof of the train would activate. Dr Wigglesworth stated that the chief 'attention getters' were the roof-mounted strobe lights activated in conjunction with the horn.

[46] The agreement between the parties also recorded the following distances relevant to events:-

Range of distances Event 30 - 50 metres Horn ceases to sound 99 - 102 metres Brake cylinder pressure beings to rise, with actuation occurring 0.2 seconds prior to this 285 - 310 metres Horn begins to sound 337 - 363 metres Throttle moves from 1/3 to idle

[47] … the train driver's reaction time in applying the brakes would have been in the range of 1.0 to 1.5 seconds."

"[48] … (a) At 20 seconds before the collision, the train was travelling between 92 to 93 kilometres per hour. (b) At 14 seconds before the collision, the train was travelling between 93 to 94 kilometres per hour. (c) At four seconds before the collision, the train was travelling between 92 to 93 kilometres per hour. (d) At two seconds before the collision, the train was travelling at 90 to 92 kilometres per hour. (e) At the moment of impact, the train was travelling in the range of 83 to 88 kilometres per hour." 32 The rear of the train came to rest at 216.5 m from the northern road alignment of the level crossing and the front of the train was 266.5 m from the level crossing. 33 The significance of some of the above matters requires explanation. The effect of moving the throttle to idle was to maintain the speed of the train, which at that point was 93 km per hour. The brakes started to come on 4 seconds, or about 100 m, before the collision (having been activated 0.2 seconds earlier). As the brake was activated 0.2 seconds before the brake pressure began to rise, the train was approximately 99-102 m from the crossing when Mr Matthews applied the brake. The horn ceased sounding 2 seconds, or 50 m, before the collision because Mr Matthews tried to get out of his seat so as to protect himself from the force of the collision, which at that point was a virtual certainty. There was no criticism, either by the appellant or by the trial judge, of Mr Matthews taking this action. 34 The speed of the truck could not be accurately assessed. There was some evidence that as the truck turned from the Highway into Carroll Road it was travelling at 20 km per hour. However, his Honour treated the truck as travelling at "15 km per hour or a little more", or about 4.2 m per second. The parties at trial had accepted it was appropriate to proceed on the basis that the truck was travelling a constant 15 km per hour.

Legal principles 35 The SRA and the RIC each accepted that they owed a duty of care which, having regard to their statutory functions, they defined in the following terms: "1. SRA (as the owner and operator of the Xplorer train) owed a duty of care to take all reasonable precautions to ensure that the train was operated in a manner so as not to endanger members of the public, acting with due care for their own safety, when passing through the level crossing in their vehicles.

2. RIC (as the owner of the railway infrastructure comprising the rail track and level crossing at Breeza) owed a duty of care to take all reasonable precautions to ensure that the level crossing at Breeza would be safe for members of the public, acting with due care for their own safety, when passing through the level crossing in their vehicles." (At [211]) 36 The respondents relied upon the principles stated in South Australian Railways Commissioner v Thomas [1951] HCA 48; (1951) 84 CLR 84 at 89; Commissioner for Railways v Dowle [1958] HCA 15; (1958) 99 CLR 353 at 358; Commissioner for Railways v Murphy [1967] ALR 706; (1967) 41 ALJR 77 at 77(2), 78(2); and Wade v Australian Railway Historical Society (South Australian Division) (t/as Steamranger) (2000) 77 SASR 221, at [75], 233 as governing their respective duties. 37 In South Australian Railways Commissioner v Thomas, a case in which a collision occurred between a train and a car on a level crossing, the High Court made the following statements, (at 89), as to the duty of care of the Railways Commissioner: "In considering whether adequate warning was provided at a level crossing over a public road all the circumstances of the locality and of the traffic passing over it and the conditions prevailing at the relevant time must be taken into consideration: Alchin v. Commissioner for Railways . The duty of the commissioner is to do everything which in the circumstances is reasonably necessary to secure the safety of persons using the crossing: Cliff v. Midland Railway Co. at p. 261; Ellis v. Great Western Railway Co. , at p. 555; Liddiatt v. Great Western Railway Co at p. 550. This must include a duty to give reasonable warning of the approach of a train where the commissioner does not provide gates which are closed when a train is approaching. That duty is not fulfilled by providing means which would enable persons of acute vision and hearing exercising the most anxious care to avoid injury. The fact that all sorts and conditions of people use the highway must be taken into account, and, whilst the commissioner is not required to protect against their own carelessness people who proceed without any regard to their own safety, it is his duty to take every reasonable precaution to ensure that the level crossing will be safe for the members of the public generally who act with due care while exercising their rights of passing over it." 38 This statement was applied in Commissioner for Railways v Dowle at 358 and Commissioner for Railways v Murphy at 77(2) per Barwick CJ and at 78(2) per Kitto J (Owen J agreeing), which were also cases in which a vehicle collided with a train at a level crossing. 39 As previously explained, the focus of the appeal against the SRA was that the trial judge erred in finding that the SRA was not contributorily negligent. The question whether a party who has suffered damage is contributorily negligent is governed by the Civil Liability Act 2002, s 5R which provides: "5R Standard of contributory negligence

(1) The principles that are applicable in determining whether a person has been negligent also apply in determining whether the person who suffered harm has been contributorily negligent in failing to take precautions against the risk of that harm.

(2) For that purpose: (a) the standard of care required of the person who suffered harm is that of a reasonable person in the position of that person, and (b) the matter is to be determined on the basis of what that person knew or ought to have known at the time." 40 In Consolidated Broken Hill Ltd v Edwards [2005] NSWCA 380; (2005) Aust Torts Reports ¶81-815, Ipp JA (Giles JA and Hunt AJA agreeing) stated: "[70] Section 5R(1) was introduced into the Civil Liability Act by the Civil Liability Amendment (Personal Responsibility) Act 2002 (NSW). In the second reading speech introducing the bill which became the Civil Liability Amendment (Personal Responsibility) Act, the Premier said (Hansard 23 October 2002 at 5765) that many of its provisions were modelled on the recommendations in the final report by the panel appointed by the Commonwealth and State governments to review the law of negligence (review of the Law of Negligence Final Report September 2002 [the Ipp Report]). Section 5R(1) is in accord with Recommendation 30 of the panel. Paragraph 8.10 of the panel's final report states: '8.10 In the opinion of the Panel, there is in the Australian community today a widely-held expectation that, in general, people will take as much care for themselves as they expect others to take for them. This is an application of the fundamental idea that people should take responsibility for their own lives and safety, and it provides powerful support for the principle that the standard of care for negligence and contributory negligence should be the same.' Section 5R(1) reflects this 'fundamental idea', as well as the concepts expressed by Callinan and Heydon JJ in Vairy (at [220])." 41 The paragraph from Vairy v Wyong Shire Council [2005] HCA 62; (2005) 223 CLR 422 referred to in this passage adverts to the underlying philosophy of the common law which governs in this area. It reflects that which was said at para 8.10 of the Ipp report, set out above. Since Vairy, the High Court has considered the question of the duty of statutory authorities and the concepts of contributory negligence on a number of occasions, relevantly, in respect of the duty of a road authority. In that regard, Gummow J (Callinan and Heydon JJ agreeing), in Roads and Traffic Authority of NSW v Dederer [2007] HCA 42; (2007) 211 CLR 317 made the following observations: "[43] … duties of care are not owed in the abstract. Rather, they are obligations of a particular scope, and that scope may be more or less expansive depending on the relationship in question. Secondly, whatever their scope, all duties of care are to be discharged by the exercise of reasonable care. They do not impose a more stringent or onerous burden.

[45] … The result of [ Brodie v Singleton Shire Council [2001] HCA 29; (2001) 206 CLR 512] is that a road authority is obliged to exercise reasonable care so that the road is safe 'for users exercising reasonable care for their own safety' … the weight to be given to an expectation that potential plaintiffs will exercise reasonable care for their own safety is a general matter in the assessment of breach in every case, but in the present case it was also a specific element contained, as a matter of law, in the scope of the RTA's duty of care.

[46] A road authority such as the RTA is not obliged to exercise reasonable care in the abstract …

[47] The RTA's duty of care was owed to all users of the bridge, whether or not they took ordinary care for their own safety; the RTA did not cease to owe Mr Dederer a duty of care merely because of his own voluntary and obviously dangerous conduct in diving from the bridge. However, the extent of the obligation owed by the RTA was that of a roads authority exercising reasonable care to see that the road is safe 'for users exercising reasonable care for their own safety'. The essential point is that the RTA did not owe a more stringent obligation towards careless road users as compared with careful ones. In each case, the same obligation of reasonable care was owed, and the extent of that obligation was to be measured against a duty whose scope took into account the exercise of reasonable care by road users themselves." (Citations omitted) 42 The principles stated by the High Court in RTA v Dederer are to the same effect as its earlier statement in South Australian Railways Commissioner v Thomas upon which the respondents relied. The appellant did not submit that the respondents' formulation of the duty of care that they owed in the circumstances was erroneous. Its case was that, on the facts, findings of negligence and contributory negligence should have been made. Contributory negligence is determined objectively from all the facts and circumstances and the party alleging contributory negligence bears the relevant legal onus to establish the allegation: Joslyn v Berryman [2003] HCA 34; (2003) 214 CLR 552. There may be a question whether the appellant having been found not to have exercised reasonable care, it was not owed a duty of care at all.

Was the SRA negligent or contributorily negligent? 43 As I have already indicated, the appellant advanced three bases upon which the SRA should have been found to be contributorily negligent. First, that the train driver was negligent in failing to take timely action to avoid a collision. Secondly, that the SRA was negligent in constructing the level crossing on a curve. Thirdly, that the SRA was negligent in failing to implement a reduced speed limit at the crossing.

Was the train driver negligent? 44 The appellant's case at trial was that Mr Matthews did not sound the train's horn until the truck was on the crossing. It had been submitted that there was no reasonable basis for Mr Matthews to believe that the truck would stop at the stop sign and that he should have realised, in time to apply the brake to avoid a collision, that it was not going to stop and was proceeding, without slowing down, onto the crossing. Much of the force in the appellant's case was taken away when it was established that the train horn had sounded continuously for 9.25-10.75 seconds and that it first sounded some 11.25-12 seconds before the collision and that it could have been heard 300 m away and, therefore, by a truck travelling down Carroll Road. His Honour found that in the circumstances, Mr Matthews maintained a proper lookout and applied the train's brake immediately it became apparent to him that the truck driver had not complied with the stop sign and was proceeding across the level crossing. His Honour concluded that the train driver operated the train in a responsible and safe manner. 45 On the appeal, the appellant accepted that throughout almost the entire period in which the truck was travelling down Carroll Road to the level crossing, the train's horn was sounding. The appellant also accepted that Mr Stackman was travelling at about 15 km per hour and that he had not stopped at the stop sign. Nonetheless, the appellant contended that the train driver was contributorily negligent in failing to brake earlier than he did. In this regard, both the appellant and the respondents placed particular reliance on the fact that the train driver had sounded his horn for an unusually long period, although they took converse positions as to the import of his doing so. 46 The respondents submitted that the fact that the horn was sounded for approximately 9-11 seconds put to rest any case that Mr Matthews was negligent, as Mr Stackman had been well and truly warned of the approach of the train. The appellant contended, however, that this established its case, as the fact that he sounded the horn as long as he did demonstrated that he was aware of impending danger, namely, that it was apparent that Mr Stackman was not, or at least possibly was not, going to stop. Accordingly, Mr Matthews should have slowed down as he approached the crossing to take account of that possible risk and be in a position to bring the train to a stop before the crossing so as to avoid a collision. 47 The appellant's case on this issue involved a misconception, which although not determinative of whether Mr Matthews was contributorily negligent, is relevant to an understanding of the dynamics of the scene. The appellant's contention was that there was something unusual that had alerted Mr Matthews to cause him to sound his horn for approximately 9-11 seconds. This perception was gained principally from the evidence of Mrs Lawlor, who said that usually the train gave a short sound of the horn as it approached the level crossing and that the longer sounding of the horn that she heard on this day was unusual. This caused her to look to see what was happening. The appellant also relied on the evidence of Mr Southam who also said he looked to see what was going on when he heard the horn sounding longer than usual. 48 However, Mr Matthews sounded the horn because it was part of his practice when he approached a level crossing if there was traffic about, as was the case on this day. That evidence emerged from the following questions asked by the trial judge: "HIS HONOUR. … when you in the past had gone through this section of track approaching the crossing, the Carroll Road crossing, what was your custom in terms of blowing the horn? A. Well, the custom is to blow the horn approaching all crossings.

Q. What sort of signal do you give? A. If there's traffic approaching the crossing you give a fairly long blast but if there is no traffic there you just give a short blast and continue on." 49 The appellant's senior counsel was given leave to ask questions arising out of that evidence as follows: "Q. The short blast you have mentioned, how long does that take? A. Maybe a second.

Q. Could be two seconds? A. Could be."

This was undoubtedly the short sound of the horn to which Mrs Lawlor referred. 50 The appellant also placed reliance upon certain aspects of Mr Southam's evidence as establishing that Mr Matthews was negligent in not applying the brakes earlier. In particular, it contended that Mr Southam, who responded to the sound of the horn by looking out the window, saw the rear of the appellant's truck on Carroll Road, about 15-20 m down from the highway. The front of the truck, on these estimates, would have only been about 22.5 m from the railway tracks. The appellant contended that on Mr Southam's evidence, there was no braking at this point and he, Mr Southam, reacted to the probability of an accident occurring in the manner already described. The appellant's point was that if Mr Southam anticipated an accident at this point, Mr Matthews should also have anticipated a collision with the truck and commenced to brake earlier than he did. 51 However, Mr Southam's evidence does not advance the appellant's case in the way contended. Mr Southam estimated that from his position in the train when he looked out of the window, the level crossing was about 120 m away. On that estimate, the front of the train was about 70 m from the crossing. As Mr Matthews commenced to brake at approximately 100 m, or 4 seconds, from the crossing, the train was already braking when Mr Southam first saw the truck. 52 The appellant also relied upon Mr Matthews' evidence that the truck was on the crossing when the train's emergency brake was applied. Mr Matthews' evidence in this regard was as follows: "Q. As you approached the level crossing what did you observe the truck doing? A. I observed the truck coming down the grade towards the crossing and I also observed that he did not stop at the stop sign there but continued through the crossing.

Q. And do you have any recollection where the truck was in relation to the crossing when you first applied the brakes? A. The truck would have been entered the crossing at the time I first applied the brakes. Yeah, that's about, without saying my considerations that is the about the best way I can answer that.

Q. Yes. And just after you applied the brakes what happened then? A. Shortly after applying the brakes I came into contact with the rear of the semitrailer up in the vicinity of the rear three wheels. The front window came in, the cab started to collapse around me, unfortunately the load of the semi came in through the front window and yeah, it wasn't very pleasant.

Q. As you have described the movement of the truck, where were you looking? A. I was watching the truck the whole time."

The appellant contended that this, of itself, demonstrated Mr Matthews' negligence. If he was watching the truck the whole time, he should have realised before he did that the truck was not going to stop. 53 The appellant next submitted that had Mr Matthews applied his brake a fraction of a second earlier than he did, the collision would have been avoided. On the appellant's case, that was 0.3 seconds. The respondents accepted that the difference between hit and miss in this case was a fraction of a second, somewhere between 0.3-0.5 seconds. The respondents argued, however, that it could not be precisely ascertained whether the application of the train's brakes 0.3-0.5 seconds earlier would have made a difference. 54 On the trial judge's finding that the truck was travelling at about 15 km per hour, there was evidence from which it could be deduced that the truck could have been brought to a stop in approximately 5 m, or in just over 1 second. The calculation was derived from the Draft Standard, Table 6B.1: "Values of safe stopping distance, S1, for sealed and level surfaces". It is not known whether the calculations in Table 6B.1 took account of the downward slope of Carroll Road approaching the crossing, or of the load on the back of the truck. However, this was the only evidence on the subject of stopping distances and both parties accepted that it provided some guidance in assessing the braking distance and stopping time of the truck. It should also be remarked that the appellant did not call any evidence on the topic notwithstanding that it bore the onus on the question of contributory negligence: Joslyn v Berryman. 55 As I have noted earlier, his Honour's finding that the truck speed was about 15 km per hour was itself an estimate not based on any scientific or objective data. The respondent also submitted that the distance the truck travelled from the time the train horn first sounded was also somewhat conjectural. The experts had jointly undertaken a series of analyses of the distance the truck travelled from the point it turned off the highway down to the stop sign. The distance depended upon the radius of the turning curve and was likely to have varied from 66.95 m to 75.06 m. The respondent placed some reliance on this, although its relevance is marginal in determining where the truck was when Mr Matthews first observed it, or in determining when he should first have realised the truck was not going to stop. However, this evidence did serve to underscore that there were considerable difficulties in the appellant contending that Mr Matthews should have applied the brakes of the train earlier, when distances, speeds and times could only be assessed within various ranges. 56 The respondents next pointed out that although the speed of the train had been determined with reasonable accuracy, it was still only recorded within a range of 1-2 km per hour: see the agreement of the parties recorded at [48] of his Honour's judgment (at [31] above). Likewise, the point at which the brakes came on was also recorded by the data logger as being between 99 and 102 m. The respondents submitted that the 3 m differential was the difference between a hit and a miss. 57 The respondents further submitted that for the purpose of evaluating the reasonableness of Mr Matthews' action in not braking earlier, it was necessary to take into account reaction time, which his Honour accepted was 1-1.5 seconds and then to have regard to the distance the truck would travel in that time. Having regard to both the known data recorded by the train's data logger and the assumed information as to the speed of the truck and the train driver's reactions times, his Honour, at [250], engaged in the following analysis to establish the likely position of the train from the crossing in relation to the truck at the point the truck passed the stop sign (which was approximately 3 m from the first rail of the track): "(a) On the basis of the truck driver's account of a truck speed of 15 kilometres per hour at the time of accident (see Exhibit 2 - motor vehicle claim form dated 21 March 2001), that speed equates to 4.2 metres per second.

(b) The agreed distance between the stop sign to the point where the truck was hit by the train was 20.5 metres.

(c) Arithmetical calculation shows that it would have taken the truck 4.9 seconds to travel that distance of 20.5 metres at the above speed, ie, between the stop sign and the point of collision.

(d) Based on the same time interval of 4.9 seconds (rounded off to 5 seconds), the approximate position of the train can be established at the point that the truck moved forward of the stop sign. The train was travelling at 93 kilometres per hour or 25.8 metres per second. At an interval of 5 seconds before the collision, the train was 129 metres from the point of collision (5 seconds x 25.8 metres per second = 129).

(e) On the recorded data, the train horn was sounded when the train was at a greater distance than 129 metres from the point of collision, namely, at between 285 and 300 metres. The train driver's evidence was that he wished to give the truck driver, whose truck was within his sight, plenty of warning.

(f) At a train distance of 129 metres from the crossing the front of the truck was level with the stop sign.

(g) At a little under 5 seconds before the collision, namely, 4 seconds, the emergency brake of the train was applied. The truck at this point would have just moved past the stop sign and, when the emergency brakes were applied, the train was then 99 to 102 metres away from the crossing.

(h) On the basis of a truck speed of 15 kilometres per hour, the truck, in 1 second, would have moved 4.9 metres forward of the stop sign towards the crossing.

(i) Allowing for perception/reaction time by the train driver before activating the emergency brakes (of 2 seconds), in addition to the interval of 1 second referred to in (h), the truck would have advanced [4.2] metres when the train driver realised the truck was proceeding to enter the crossing. [The trial judge had said 4.9 m, but both parties accepted that the correct distance was 4.2 m.]

(j) The stop sign was 3.2 metres from the nearest rail. The truck, on the above basis, was on the crossing when the emergency brake was applied." 58 His Honour concluded, therefore, that the truck was on the track when Mr Matthews applied the emergency brake. This confirmed Mr Matthews' evidence on this point. The appellant contended that his Honour's conclusion, particularly in para (i), was erroneous because his Honour had failed properly to take account of the 2 second reaction time referred to in that paragraph. Had his Honour taken that into account, then, accepting that the brake was applied 4 seconds before the collision, his Honour should have found that the truck was 1 m past the first rail 6 seconds before the collision. The appellant, by extrapolation from this figure, reasoned that the truck would have cleared the track before the train reached the crossing, had Mr Matthews applied the brake earlier. The analysis was set out as follows in the appellant's original submissions: "8.61 Therefore at a constant speed of 15km/h (4.2m/sec) at 6 seconds out from the collision the front of the truck was 1m past the near side track line.

8.62 At 5 seconds prior to impact the front of the truck was 5.2m past the near side track line (ie 1m + 4.2m = 5.2m).

8.63 At 4 seconds prior to impact the front of the truck was 9.4m past the near side track line (5.2m + 4.2m = 9.4m).

8.64 At 3 seconds prior to impact the front of the truck was 13.6m past the near side track line (9.4m + 4.2m = 13.6m). The tracks were 1.5m apart. Therefore at this time the front of the truck was 12.1 metres past the far side track line.

8.65 At 2 seconds prior to impact the front of the truck was 17.8m past the near side track line and 16.3m past the far side track line.

8.66 At one second prior to impact the front of the truck was 22m past the near side track line and 20.5m past the far side track line.

8.67 At impact the front of the truck was 26.2m past the near side track line and 24.7m past the far side track line,

8.68 The truck is 16.3m long. Therefore at impact the rear of the truck would have been 8.4m past the far side track line. In other words the truck would have cleared the tracks by 8.4m." 59 The appellant submitted, therefore, that the train driver should have observed the truck earlier than he did, in which case the collision would have been avoided. 60 The respondents submitted that this reasoning was misplaced. They submitted that, properly understood, his Honour meant that the truck advanced at a speed of 4.2 m per second during the 2 second reaction time before the train driver activated the brake. Thus, the truck travelled a distance of 4.2 m when the train driver realised that the truck was proceeding to enter the crossing and a further distance of 4.2 m past the stop sign towards the crossing before the brake came on. Subject to what follows, I would agree with this analysis. 61 Both parties appeared to accept the factual basis of his Honour's reasoning in [250] (differing only as to whether his Honour should have found that Mr Matthews should have realised earlier than the truck was not going to stop). In particular, both parties appeared to accept the analysis based on an addition of 2 seconds reaction time, referred to in para (i). However, this was inconsistent with what his Honour accepted earlier in his judgment at [47], namely, that Mr Matthews' reaction time in applying the brake would have been in the range of 1-1.5 seconds. 62 Notwithstanding what I perceive to be an error in this aspect of the analysis, I do not consider the error assists the appellant's case. Given that the truck was travelling slowly and Mr Stackman could have brought it to a stop in about 5 m or 1 sec and it would have been reasonable for Mr Matthews to have assumed up to a point close to the stop sign, which was approximately 3 m from the track, that Mr Stackman was going to stop. 63 Various assessments can be made to demonstrate the point. For the truck to have stopped at the stop sign, Mr Stackman would have had to start to apply the brakes about 5 m away. The truck would have slowed gradually, by which time it would have travelled some distance, possibly 1, 2, 3 or 4 m, or some fraction of those distances. For example, in 0.5 seconds, the truck would have travelled about 2.5-3 m. The train, at this point, was about 55 m away. The truck, on the assumption that it was going at a constant 15 km per hour, was not in fact decelerating. Nonetheless, the difference in speed between travelling a distance of 4.2 m per second at a constant 15 km per hour and in travelling 5 m in a second, so as to bring the vehicle to a stop, would, at least in the first 2-4 m, be virtually imperceptible, particularly from the position of Mr Matthews, travelling at speed. 64 Once it became apparent that the truck was not going to stop, Mr Matthews would have taken 1-1.5 seconds to react before applying his brakes, by which time the truck would have travelled 4.2-6.3 m. If the truck was, for example, 2 m from the stop sign when Mr Matthews realised it was not going to stop, it would have travelled 4.2 m (on a reaction time of 1 second), or 6.3 m (on a reaction time of 1.5 seconds). In either case, it would have been on or very close to the track before Mr Matthews would have taken action to apply the brake. In the 0.2 seconds in which it took the brake to activate, the truck would have travelled another 0.8 m. 65 The calculations can also be done in reverse, that is, by working backward from the time the truck was hit. On this approach, counsel for the respondents advanced the following analysis: "i. The truck travelled a distance of 20.5 metres from the stop sign to the point where the back of the trailer was hit by the train. Arithmetic calculations show that it would take the truck 4.9 seconds to travel that distance from the stop sign at a speed of 15 kph (4.2 m/sec) …

ii. The emergency brakes came on (indicated by the increase in brake pressure) 4 seconds before impact and the brake controller was operated 0.2 seconds earlier …

iii. It follows that the brake controller was operated 0.7 seconds (4.9 sec - 4.2 sec) after the front of the truck reached the stop sign."

The analysis then continued by assuming: "… different values for the train driver's reaction time before he operated the brake controller enables the approximate position of the front of the truck relative to the stop sign to be determined when the train driver may have first perceived that the truck was not going to stop.

i. A reaction time of 1.0 second puts the front of the truck approximately 1.3 metres before the stop sign (4.2 m/sec x 0.3 sec).

ii. A reaction time of 1.25 seconds puts the front of the truck approximately 2.3 metres before the stop sign (4.2 m/sec x 0.55 sec).

iii. A reaction time of 1.5 seconds put the front of the truck approximately 3.4 metres before the stop sign (4.2 m/sec x 0.8 sec).

iv. A reaction time of 1.75 seconds puts the front of the truck approximately 4.4 metres before the stop sign (4.2 m/sec x 1.05 sec)." 66 Given the very slight difference in the perception of a truck travelling at 15 km per hour or 4.2 m a second and the same truck decelerating to a stop over 5 m in 1 second, as well as the inbuilt uncertainty in actual distances, speed and times, the appellant has not, in my opinion, established that Mr Matthews was unreasonable in not observing earlier that the truck was not going to stop. 67 The SRA and RIC also relied on the fact that that Mr Matthews was not cross-examined on the basis that he should have applied his brakes earlier; or that he should have realised that Mr Stackman had not seen him; or that he should have anticipated that Mr Stackman was not going to stop. Rather, the challenge to Mr Matthews in cross-examination was that he had not sounded the horn as he had said and that he realised too late that the truck would not get across the level crossing. Mr Matthews denied both propositions. The respondents contended that the omission to cross-examine on these matters was particularly significant, because the appellant had full notice of what Mr Matthews evidence was or was likely to be having regard to the evidence he gave in the Local Court. In particular, in those proceedings, he had said that he did not anticipate that the truck was not going to stop and that he applied the emergency brakes as soon as he realised that a collision was imminent. 68 The respondent submitted that in the circumstances the rule in Browne v Dunn (1893) 6 R 67 (HL) applied. The rule in Browne v Dunn is a discretionary rule that applies to evidence that may be called in a trial. Its content and operation was explained in Scalise v Bezzina [2003] NSWCA 362 where Mason P (Santow JA and Brownie AJA agreeing) said: "[95] … The rule is rooted in considerations of fairness. It is a rule of practice ' necessary both to give the witness the opportunity to deal with [contradictory] evidence, or the inferences to be drawn from it, and to allow the other party the opportunity to call evidence either to corroborate that explanation or to contradict the inferences sought to be drawn' ( Allied Pastoral Holdings Pty Ltd v Federal Commissioner of Taxation [1983] 1 NSWLR 1 at 16 per Hunt J)."

See also James Hardie Industries NV v Australian Securities and Investments Commission [2009] NSWCA 18. 69 As the rule in Browne v Dunn is a rule of practice, that point should have been taken at trial, but was not. However, the respondents' point now, as I understand it, is that the appellant seeks to make a different case on appeal than that made at trial. Although it was not argued that the principles in Coulton v Holcombe [1986] HCA 33; (1986) 162 CLR 1 applied, namely, that a new point cannot be raised on appeal, at least where to do so raises factual issues, nonetheless, the appellant is significantly hampered in seeking to establish contributory negligence by reference to matters that it did not put in issue at trial. 70 In Davis v Council of the City of Wagga Wagga [2004] NSWCA 34 Mason P (Beazley and Tobias JJA agreeing) remarked, at [34], that where factual matters which were seriously in issue were to be relied on, a challenge of such significance "ought to have been taken to the witness by cross-examination". The same comment is appropriately made here. 71 In circumstances where Mr Matthews said that he expected the truck to stop; where the evidence otherwise demonstrated that Mr Stackman could have brought the truck to a stop in about 5 m, or 1 second; where the horn was sounded 11.25-12 seconds prior to the collision for a total period of 9.25-10.75 seconds; and where Mr Matthews said that he applied the brakes as soon as he realised the truck was not going to stop, and there was no evidence challenging, let alone contradicting, any of that evidence, there was no evidentiary support for the appellant's assertion that Mr Matthews should have braked earlier. 72 In my opinion, the appellant has not established that Mr Matthews was negligent and this challenge to the trial judge's judgment should be rejected.

Should the SRA have reduced the speed limit at the level crossing? 73 Following the rebuilding of the crossing in June 1994, a sighting distance assessment was carried out, which revealed that the sighting distance looking in the direction in which Mr Stackman needed to look in this case, was 481 m. This was an insufficient sighting distance for a train travelling at 125 km per hour, but was sufficient for a train travelling at 98.4 km per hour. It will be recalled that Mr Matthews was travelling some 5-6 km per hour slower than this. The trial judge found that Mr Stackman had a sighting distance of approximately 400 m. The appellant submitted that for that sighting distance, the maximum speed at which the train should have been travelling was 75-82 km per hour. His Honour rejected this submission. His Honour held the available line of sight of 400 m was adequate, if the truck driver had stopped at the stop sign and made the necessary observations. This, however, does not deal with the appellant's point. The appellant contended that given a sighting distance of only 400 m, the speed limit for trains at this level crossing should have been reduced to 75-82 km per hour. 74 There were two sets of documentation that regulated, or at least guided, the operative conditions at level crossings: the RTA Standard TS 3901 (TS 3901) and the Draft Standard. At the time the crossing was rebuilt, TS 3901 was the current standard for new and reconstructed level crossings. However, it was based upon sighting distances and speed conditions of motor vehicles, not upon the speed capacity of semi-trailers or B-double vehicles and, therefore, on the appellant's argument, was inadequate for this crossing. At about this time the Draft Standard was under consideration by the SRA, but it was never formally adopted. However, the Draft Standard was recognised and used as a guide by the expert witnesses in the case as containing appropriate standards for regulating conditions at level crossings. 75 The appellant submitted that when the crossing was rebuilt in 1994, it should have been built so as to comply with TS 3901, which provided: "1.1 This Standard sets out the requirements for new Level Crossings and the reconstruction of existing Level Crossings.

1.6 New Level Crossings should not be located on curves of Main lines of any class." 76 The appellant submitted that the rebuilding of the level crossing brought it under the rubric of a new crossing and, accordingly, cl 1.6 applied and should have been observed and the crossing should not, therefore, have been located on a curve at all. Rather, the crossing should have been closed. There was another level crossing located 1.5 km back towards Werris Creek. This crossing had a clear view in both directions for a considerable distance and, on the appellant's argument, should have been used as a crossing for vehicles seeking to travel from the Kamilaroi Highway to Breeza. The appellant's submission was supported by Dr Wigglesworth, the appellant's expert, who stated in his supplementary report dated November 2005 that TS 3901 had first been introduced in 1986, but had not been observed by the SRA when reconstructing the level crossing in 1994. 77 Alternatively, the appellant submitted that if the crossing was to remain in use, the SRA should have taken steps to ensure that the sighting distance for trucks entering the level crossing was safe. As I have indicated, this, in effect, became the appellant's main argument, the appellant recognising that, on a case against the RIC based on failure to close the crossing, the appellant may not have been able to establish that such failure was causative of the accident. 78 The respondents submitted that the appellant had misunderstood cl 1.6 of TS 3901. Although, in cl 1.1, the Standard was expressed to extend to new and reconstructed level crossings, each specification had to be read according to its terms. By its express terms, cl 1.6 only applied to new crossings. It did not apply to the reconstruction of an existing level crossing which the respondents maintained was the case here. They submitted that the appellant had not established that this was a new crossing. 79 His Honour did not make a finding as to whether this was a new or a reconstructed crossing and given the appellant's formulation of its case, it is not necessary to enter that factual arena. The real point in contention is whether the SRA should have reduced the speed limit for the crossing, given the sighting distances. 80 The SRA knew of the inadequacy of TS 3901, as it applied to trucks using level crossings. On 17 August 1999, Ian Ackland sent an email to Owen Henry and others, regarding the Level Crossing Standard. (The precise identity and status of these personnel is not entirely clear. However, it can be inferred that they had relevant supervisory and/or management positions with the SRA.) The memo's importance was classified as "High". Mr Ackland said: "During our audit of the Hunter Valley & North West, a matter was drawn to our attention which I believe has serious safety implications. RSA field staff are having difficulty with the application of RAC Standard TS3901 for Level Crossing Design, in that it does not reflect current situations for sighting distances for large and heavy vehicles used in the country regions. While I am not an expert in the matter, I was told that the distances are based on the approach of a small passenger car to the crossing, where they really need to take account of semi-trailers such as B-doubles. Correspondence has been raised through John Boswell to our Standards people, with a response from John back in April this year. John quotes from a Peter Hicks' memo: 'The current standard is not up to date. However, until the new standard is approved and implemented, TS3901 is the standard to be applied to level crossings …' From what I can gather, there is an acknowledgment that the standard is inadequate, and that it is to be replaced, but so far this has not occurred.

Greg and I think this is a 'sleeper' at the moment, just waiting to ambush us. I hope we are wrong, but I draw this to your attention in the spirit of erring on the side of caution." 81 There were responses to this email to the effect that documentation was being worked on, addressing the issues raised. Following this, the Draft Standard was prepared, which, as I have indicated, both sides accepted provided a guide as to the conditions that should operate at level crossings. 82 Appendix 6C of the Draft Standard related to the "Assessment for the adequacy of stop signs". It provided, relevantly: "The driver of a vehicle stopped at the stop line needs a clear view for a distance (S 3 ) along the rail sufficient to see there is no approaching train travelling at the permissible track speed before the driver starts to drive across the crossing. The distance, S 3 , is the distance along the track that the fastest train travels in the minimum sighting time. The minimum sighting time is the time it takes a road vehicle to clear the crossing plus a five second safety margin.

A track length of S 3 and an approach road distance of 5 m are used in measuring the adequacy of stop-signed crossings …

The measured clearance times for different types of fully laden vehicles are shown in Table 6C.1.

Table 6C.1 Values of Measured Vehicle Clearance Times for Sealed Level Surfaces (Single Track)

TYPE OF VEHICLE CLEARANCE MINIMUM SIGHTING TIME(5) (sec) DISTANCE(1) (m) TIME MEASURED(2) (sec) CAR 16.8 10.6(6) RIGID TRUCK 22.4 9.3(3) 14.3 SEMI-TRAILER 28.9 12.6 17.6 B-DOUBLE 34.4 13.6 18.6 ROAD-TRAIN 46.4 21.3(4) 26.3

(1) approach side 5 m clearance + single track width 1.4 m (double track use 5.7m) + departure side 5 m clearance + length of vehicle (car, 5.4 m, rigid truck 11 m, semi-trailer 17.5 m, B-double 23 m and road-train 50 m)

(6) the value of 10.6 sec was calculated based on the following:

• 5 sec safety margin applied" 83 Appendix 6C also contained a table of values of minimum sighting distances for various train speeds. At a speed of 80 km per hour, the minimum sighting distance was 391 m for a semi-trailer (and 413 m for a B-double). At 100 km per hour, the minimum sighting distance was 489 m for a semi-trailer and 517 m for a B-double. At 120 km per hour, the distances were 587 m for a semi-trailer and 620 m for a B-double. 84 The appellant submitted that from these figures, a calculation could be made that the recommended speed for a sighting distance of 400 m was 82 km per hour for a semi-trailer and 77 km per hour for a B-double. The appellant submitted that it was thus apparent, from the SRA's own standards, that given its sighting and clearance requirements the speed for this level crossing was too high. 85 On these calculations, the appellant's case is a deceptively simple one. If the speed of the train had been regulated so that the maximum speed at this section of track was either 77 km per hour, or, at the most, 82 km per hour, this accident would not have occurred. A speed of 77 km per hour was the appellant's preferred figure, on the basis that the speed chosen had been assessed based on the heavier type of vehicle that used the crossing. The appellant submitted that as a survey had been done that included B-doubles, it would appear that this type of vehicle used the level crossing. 86 The appellant relied upon the report of Dr Wigglesworth to support its case that the maximum train speed at the level crossing should have been reduced. In this regard, Dr Wigglesworth referred to the survey conducted of the level crossing in November 1994 and the range of measures where the sighting distance indicated in the Draft Standard for stop signs could not be achieved. Those measures included reducing maximum train speeds. Dr Wigglesworth considered this was an "entirely appropriate" treatment. He noted that: "The suggestions to slow the train to 93 km/h in the Down direction (ie away from Sydney) … were in fact proposed in the Level Crossing Assessment Report on this crossing dated 19/11/94"

but added that this proposal had not been implemented at the time of the accident. 87 Dr Wigglesworth also noted that on the SRA's own assessment, there were inadequate sighting distances in three of the four quadrants and that the overall assessment of the level crossing was that sighting distances and vision angles were not acceptable. Dr Wigglesworth continued that: "… this internal Assessment Report gives a very clear warning that in three of the four quadrants (including the quadrant relevant to the collision in March 2001) the sight distances fell below the guidelines set out in the [Draft Standard]." 88 Dr Wigglesworth commented that at management level, the SRA was well aware of the problems with sighting distances and that the solution, if sighting distances could not be achieved, was to reduce train speed. Dr Wigglesworth's observation was well made, based as it was on the SRA's own internal documentation. 89 His Honour referred to Dr Wigglesworth's evidence as follows: "[282] … where sighting distances cannot be achieved, one strategy adopted by [the SRA] involved the reduction of train speeds which he considered to be an 'entirely appropriate' treatment of the issue (p.22 of his report). He noted that suggestions recorded in a file document to slow trains on the relevant section of track to 93 kilometres per hour in the 'down' direction had not been implemented with the maximum permissible train speed being fixed at 125 kilometres per hour. However, whilst that appears to have been the case, as noted earlier, the train in question was driven, prior to braking, at 92 to 93 kilometres per hour …" 90 The appellant contended that his Honour, at [282], apparently failed to appreciate that Dr Wigglesworth's evidence that the maximum speed for the train was 93 km per hour was based on a sighting distance of 481 m, whereas on his Honour's own finding the sighting distance was approximately 400 m. However, for the reasons I explain below, this submission is based on the premise, which the respondents challenged, that the "sighting distance" of about 400 m was a sighting distance that could be related to the requirements of the Draft Standard. The SRA did not dispute the various calculations referred to above. However, it submitted that the appellant had failed to establish that the sighting requirements and speed limits did not comply with the Draft Standard. 91 This argument involved two propositions. The first was based upon what the SRA contended was the proper way of interpreting the Draft Standard. The second proposition was that his Honour's finding of a sighting distance of approximately 400 m could not form the basis of the finding for which the appellant contended, namely, that the maximum speed for the level crossing should have been 75-82 km per hour. The SRA argued that the underlying factors upon which the sighting distance of 400 m, was found by the trial judge, were different from the underlying factors upon which the Draft Standard was based. In short, the argument was that that the appellant was seeking to make an apple do the work of an orange. 92 The SRA submitted that the table and notes set out at [82] above are to be interpreted in the following way. A semi-trailer 17.5 m in length (see note 1) approaching the level crossing from a distance of 5 m and crossing the track and clearing it by another 5 m, travels a total distance of 28.9 m in a time of 12.6 seconds. For design purposes, a safety factor of 5 seconds was then built into the equation (see note 5), so as to calculate the minimum sighting distance. The calculation was also premised on a vehicle starting from a stationary position (that is, having stopped at the stop sign) and gaining a speed of about 10-12 km per hour by time it had cleared the crossing (see note 6: vehicle clears crossing at 3 m per second). The various speeds were calculated on the basis that vehicles were fully laden (see note 2). 93 Further and more importantly on the respondents' case, sighting distances were premised upon lines of sight taken at a specific location and measured from and to specific heights. Thus, the sighting distances in the Draft Standard, for vehicles other than a car were based on a survey measurement taken from a viewing point located 5 m from the nearest rail (which the designated viewing point) at a height of 1.8 m above the ground looking to a point representing the train at a height of 3 m above the rail track. In the case of the Xplorer, which was 4.1 m high, that meant that the sighting distance was determined by measuring the distance from the designated viewing point to where the top 1.1 m of the train could be observed. It appears that this railway line was also used by other types of trains (the suggestion was that it was used by freight trains) in which case, the height and, therefore, the relevant sight distance, would be different. However, it was not suggested that any other type of train would travel at a speed material to the maximum permissible speed. 94 The respondents contended that it could be deduced from these various calculations that for a train travelling at 93 km per hour (being the speed of the train here), the sighting distance was 454 m. (The sighting distance in 1994 was, or course, measured to be 481 m.) Again, as I understand the case of the appellant, it did not cavil with the arithmetic calculation. Rather, it submitted that there was a finding that the sighting distance was about 400 m and it was in respect of that figure that the maximum permissible speed ought to have been calculated. That argument leads to a consideration of the second of the propositions raised by the respondents and arises directly from the analysis which has just been undertaken. 95 The respondents submitted that the sighting distance of about 400 m, as found by his Honour, was not referable to the sighting requirements of the Draft Standard. The respondents submitted that whereas the sighting distance of 481 m was established by survey and in accordance with the methodology just described, the sighting distance of about 400 m was not. Rather, it was an approximate distance, taken from a different location and estimated from a different height, than that specified in the Draft Standard and was based upon essentially generalised measurements taken by Constable Wheelhouse and by an RIC officer, Mr Sheehan. 96 Constable Wheelhouse's evidence was that he walked 300 m south of the crossing and from that point kept looking backwards to the stop sign and that it was not until he had walked 400 m that he could no longer see the stop sign, which he estimated to be about 6.5-7 feet high. Constable Wheelhouse was 190-192 cm, or 6 ft, 35 in tall. He estimated that from where he stopped at the 400 m mark, the top of the stop sign was roughly level with the train driver's line of sight. The lack of accuracy in these measurements is obvious. In any event, the line of sight is measured from the perspective of a driver of a vehicle, not from the perspective of a train driver. 97 In 2001, Mr Sheehan was employed by RIC as a track supervisor. He had previously been employed by the SRA when it still had the responsibility for the infrastructure of the level crossing, as outlined earlier. He was responsible for the railway track between Gunnedah and Werris Creek. One of his tasks as track supervisor was to oversee the spraying of grass and noxious weeds. He said major spraying was carried out three times a year, but that the track was under a regular, 72-hour inspection cycle. There was also evidence that grass and weeds could grow to a height of 1.5 m. However, there was no evidence of grass or weeds of this, or near this height, on the day of the accident. 98 Mr Sheehan's duties also included preparing an annual Track Examination System report on the condition of all level crossings in his area. One of the specific requirements of such an inspection was to check for any obstructions to sighting distances caused by grass and scrub. Mr Sheehan carried out the annual examination on this crossing on 14 March 2001, in which he assessed the sighting distances in quadrant a 1 to be 416 m. He described the manner in which he did this assessment in the following terms: "… the way that I always checked the sighting distance at Breeza for trains travelling on the down (that is, trains coming from the direction of Werris Creek) was to stand at the stop line on the up (north) and down (south) side of the level crossing and look for the kilometre peg in the up direction (that is, towards Werris Creek). The pegs are about 5 foot high and are located at the toe of the ballast about 2.5 metres from the nearest running rail … The nearest kilometre peg to the Breeza crossing on the down side … is about 416 metres to the east of the level crossing. If I could see the kilometre peg from the stop line on the down side of the level crossing I would write 'OK' or place a tick on the form, in the box under 'Grass and Scrub'." 99 The respondents submitted that Mr Sheehan's evidence could not form the basis of a finding as to the "sighting distance" either, as that term was used in the Draft Standard. Mr Sheehan's measurement was only taken to a specified distance without any evidence as to how far beyond the kilometre peg Mr Sheehan could see. Nor was the measurement taken from the height specified in the Draft Standard. 100 The respondents also pointed out that the appellant had failed to tender any survey evidence evaluating the visibility of the Xplorer train (or other trains) in accordance with the Draft Standard, despite relying extensively on survey evidence for other parts of its case. This submission was directed, in particular, to the inadequacy of the Bath Stewart survey evidence tendered by the appellant. The respondents made two particular criticisms of this survey evidence, in addition to the generalised one just mentioned. First, Bath Stewart did not evaluate the sighting distance in accordance with the Draft Standard. On the survey, the top 2.4 m of the Xplorer train would have been visible at a distance of 425 m from a viewing point positioned 1.8 m above the ground and 8 m from the nearest rail (instead of 5 m as required by the Draft Standard). Secondly, the survey failed to take account of the fact that Mr Stackman's actual line of sight was higher than the 1.8 m specified in the Draft Standard. 101 The respondents submitted that the trial judge concluded correctly that the Bath Stewart report did not provide a reliable guide for evaluating the issue of train visibility. In this regard, his Honour said: "[301] … It was not based upon accurate data relevant to the truck driver's eye height in determining the observability of the train and it was based upon scenarios calculated by reference to distances from the stop line not the stop sign ." (Emphasis original) 102 In my opinion, the respondents' submission that the appellant had not made out this aspect of its appeal should be accepted. His Honour's finding of a sighting distance of approximately 400 m was a non-technical measurement which could not be used by the appellant to establish what was fundamental to its case, namely, that the speed limit at this crossing should have been, at the very least, no more than 82 km per hour. There was no evidence of the technical sighting distance relevant to the assessment of a safe train speed, other than the November 1994 assessment of 481 m. There was no evidence that the top 1.1 m of an Xplorer train was not observable at a distance of 454 m, being the required sighting distance for a train travelling at 93 km per hour to give a semi-trailer a safe clearance time of 17.6 seconds, in accordance with the Draft Standard. As I have said, the train was travelling at about 93 km per hour. Nor was there evidence concerning the non-observability of other trains at a distance of 454 m. 103 There was no evidence that this sighting distance had been reduced or obscured by overgrown vegetation. The evidence accepted by his Honour was to the contrary. The configuration of the track had not changed. Accordingly, once the appellant's case based on overgrown vegetation and trees was rejected, the only evidence of the technical sighting distance at the relevant quadrant of the crossing was 481 m. It may be that there was an oversight in his Honour's reasoning in not explaining the difference in technical sighting distance measured so as to properly assess safe speeds for trains at level crossings. However, at trial, the respondents advanced the same argument and, in my opinion, it defeats this aspect of the appellant's case based on contributory negligence. 104 It will be apparent from my reasons in finding that the SRA was not contributorily negligent, that it did not fail to exercise reasonable care. Accordingly, to the extent that the appeal also related to the SRA's primary liability, that aspect of the appeal must likewise fail.

Failing to close the crossing 105 As I have said, this was but faintly argued. The appellant did not adduce any evidence of the cost of moving the level crossing, in accordance with Wyong Shire Council v Shirt [1980] HCA 12; (1980) 146 CLR 40. Nor was there any other evidence in the case from which this could be determined. The appellant submitted that there would not have been any cost involved in using the existing alternate level crossing 1.5 km away. That is not necessarily self-evident. In any event, the Draft Standard recognised that the specification that crossings should not be built on curves was not absolute and that if a crossing was on a curve, compensatory safety methods should be implemented, such as reducing the speed limit on the track. 106 In my opinion, the appellant did not establish negligence in failing to close the crossing. Although the speed limit had not been reduced to something below 100km per hour, the train was travelling at a speed which was appropriate for a sighting distance of 481 m. Accordingly, there was no causative effect on this accident in the failure to reduce the train speed at the crossing.

The result, other than quantum 107 It follows that the appeals in respect of liability and contributory negligence against the SRA and the RIC should be dismissed.

Ground 8: quantum 108 The appellant also challenged the trial judge's award of damages for the costs of the repair of the train on the basis that those costs were unreasonable. The challenge was directed to the "profit" component of the overall costs of repairs charged by the repairer, Bombardier, which the appellant contended were "extravagant". It submitted that it was not reasonable to pass such an extravagant profit margin on to the appellant. 109 The appellant's case at trial that the repairer's profit was extravagant was based on a report by Mr Olsen, engineer. 110 Both the appellant and the SRA advanced detailed written submissions on this issue. However, the essential point advanced by the parties is that to which I have referred. The appellant contended the profit was excessive and, therefore, not allowable. The SRA submitted that in circumstances where the appellant had not established that the work could have been carried out for a lower amount, it had not discharged its onus of establishing that the respondents had not mitigated their loss. 111 The approach that Mr Olsen took in providing his opinion was to undertake a "notional pricing estimate". This estimate was based on three documents, Project Full Cost Summary, the Financial Appraisal as at 31 October 2001 and the Financial Appraisal as at 31 January 2002. The objectives in undertaking the notional pricing estimate were to compare: " Actual costs and profits with pre-repair estimates of Bombardier.

To formulate a commercial viable alternate pricing structure with prices paid by SRA."

The SRA contended that this approach was flawed for two fundamental reasons: "Firstly, … that [Mr Olsen] relied upon the Project Full Cost Summary and the Financial Appraisal as at 31 January 2002 as primary evidence as to fact when there was no basis that established that they were, in fact, primary evidence. In other words, both documents, it was submitted, were utilised by Mr Olsen upon an assumption that they respectively represented the final pre-contract estimates and the actual costs on completion, he not having verified or validated his assumptions in those respects.

Secondly, Mr Olsen … did not, as an expert, identify the basis for the notional analysis [that he undertook] and there was not proved in evidence the factual sub-stratum for his notional pricing analysis" 112 The trial judge, at [454], observed that it was fundamental that an expert must identify the assumptions made, or the facts to be established by evidence, that are said to support the expert's conclusions or opinions. Having analysed the documents upon which Mr Olsen relied, his Honour concluded: "[463] Both the estimated and actual costs of the project in question are matters of historical record and such estimates and costs would be expected to be readily proved by Bombardier's internal project documents. Whilst the Project Full Cost Summary and the Financial Appraisal as at 31 January 2002 provide some information on estimates and actual costs, they do not, in my opinion, in themselves validate the primary facts essential to Mr Olsen's analysis. His analysis to that extent lacks a necessary primary evidentiary foundation." 113 In this regard, his Honour had earlier, at [450], considered that it was relevant that Mr Olsen had not relied on Bombardier's general ledger, which Mr Olsen himself had footnoted in his report as being: "… typically more accurate than at earlier stages when in the Financial Appraisal or a similar tool is also commonly relied upon for reporting of work in progress and costs accrued." (At [449]) 114 His Honour also considered that, when regard was had to the Financial Appraisal as at 31 October 2001, it appeared that Bombardier's financial appraisal documents were, at least in part, forecast records and not records of actual costs. There was also evidence that the Financial Appraisal as at 31 October 2001 may have been reviewed and altered after it was created. His Honour discussed this as follows: "[465] Additionally, the evidence suggests the possibility that the Financial Appraisal as at 31 October 2001 may have been reviewed and altered after it was created. In this respect, the Financial Appraisal as at 31 October 2001 'forecast total' was at a later time reproduced in 'Financial Forecast as at 31 January 2002' under the heading 'Previous Appraisal Forecast 31/10/01' . Some of the amounts as originally stated, changed. Thus, the amounts shown for 'labour & overhead subtotal' ($137,802) is not the figure recorded in the appraisal as at 31 October 2001 ($249,252) and similarly the 'materials total' of $147,939 had previously been shown as $202,832 in the Financial Appraisal as at 31 October 2001. The remaining sub-totals correlate with the Financial Appraisal as at 31 October 2001. These variations in sub-totals had a significant bearing on the total gross profit figures shown in the two documents (23.19% versus 42.78%).

[466] The documents are also consistent with the proposition that at some point after 31 October 2001, original forecasts for 30/09/01 and for 31/10/01 were revised due to a reduction in labour and overhead and materials sub-totals, generating a higher level of profitability. The reason for the ability to save or reduce costs on previous forecasted figures during the course of the repair project was not explored in evidence." 115 His Honour, at [467], considered whether, even if Bombardier recovered a net return well in excess of 20 per cent (being towards the top of the range suggested by Mr Olsen), such a net return would render the contract costs "extravagant" or "exorbitant". His Honour remarked, at [470], that a "particularly profitable" contract, as Mr Olsen described it, did not mean it involved "extravagant or exorbitant pricing". 116 His Honour concluded that the appellant had not established by way of primary evidence the necessary substratum for the assumptions Mr Olsen had made in relation to the two documents, Project Full Cost Summary and Financial Appraisal as at 31 January 2002. The result was that there was no primary evidence to establish the actual costs after the whole contract had been completed. 117 His Honour thus concluded: "[474] Even if the Financial Appraisal as at 31 January 2002 document may be taken as indicating a revision of forecasts and the reduction of costs to Bombardier as the project proceeded below what had been estimated, that does not lead to the conclusion that the costings, as at the date of the contract, were excessive or unreasonable. There may, for example, have been techniques or methodologies employed that enabled Bombardier to reduce costs as the project proceeded. The position remains, as earlier stated, that the tendering and associated procurement processes were established as having been conducted on a proper basis in accordance with appropriate standards and with independent verification. The cross-defendant did not establish that the cross-claimant acted unreasonably in entering into the contract with its specialist sub-contractor." 118 It followed on his Honour's reasoning that the appellant had not established that there was any evidence of unreasonable repair costs. Nonetheless, his Honour proceeded to examine Mr Olsen's analysis. On his analysis, Mr Olsen determined that "actual costs" should have been $460,095, as compared to Bombardier's estimated costs of $762,180. On this approach, the net margin was 46.1 per cent. 119 However, the appellant did not adduce any evidence of Bombardier's usual rates and margins or rates and margins charged by industry competition. As his Honour pointed out, at [476], rates and margins may vary depending upon the nature of the work and the other relevant technical and commercial factors, such as contractor's capacity and facilities. In this regard, it was relevant that Bombardier was a specialist manufacturer and had produced the proprietary design for the Xplorer. Mr Olsen acknowledged that only Bombardier would have been in possession of the detailed drawings and design information relevant to Xplorer trains and that: "[379] … at the manufacturing level, it was likely that that company possessed jigs and tooling specific to the design that would be of use during the repairs to the Xplorer car (report, Exhibit AC, paragraph 3.3.1.1).

[380] Mr Olsen further acknowledged that Bombardier were in the process of undertaking retrofit work on the Xplorer fleet. It, accordingly, had the advantage of detailed current familiarity with the cars at a shop floor level as well as having the benefit of supervisory experience from the manufacturing stage." 120 In my opinion, his Honour's conclusion, at [481], that Mr Olsen's "notional pricing" did not provide a basis for establishing that the actual repair costs (including the profit margin) were outside a reasonable range, was correct. 121 In the end result, as I understand it, the appellant's argument was limited to the submission that a profit margin of approximately 45 per cent was extravagant and therefore it ought to be relieved of paying damages in an amount that was excessive. The appellant contended that it was not arguing that the SRA had failed to mitigate its loss and thereby the appellant did not bear any relevant onus to establish this. Rather, by analogy to the 'penalty' cases, the Court had the overriding power and jurisdiction to intervene. 122 A case 'analogous' to penalty was not run at trial, but in any event, the analogy is inappropriate. In brief terms, a penalty clause is a clause in a contract which specifies the damages payable on breach of a contract, but which is not a genuine pre-estimate of the damages which would flow from breach. Such a provision usually specifies a quantum of greater and often exorbitant damages: see generally Amev-Udc Finance Ltd v Austin [1986] HCA 63; (1986) 162 CLR 170. The SRA had entered into a tender process for the carrying out of the repairs. The SRA proceeded with a sole tender process on the basis that Adtranz was the original rollingstock manufacturer and the custodian of the drawings required to perform the repairs, which would remove concerns regarding intellectual property rights. It is not clear on the evidence whether detailed consideration was given to other repairers to undertake the repair work before the decision was made to proceed with a sole tender process, but the appellant accepted that no other repairer was available, on my understanding meaning that the choice of Adtranz was a reasonable and proper choice. 123 Prior to the submission of the tender, Adtranz was acquired by Bombardier, who then submitted the tender to the SRA. The appellant accepted that a sole tender process was appropriate given the prevailing market conditions and the familiarity that Bombardier had with the train cars. The issue in dispute at trial was whether or not the SRA had taken reasonable steps to protect itself in the sole tender process. His Honour found that the tender process was carefully undertaken and as I have already stated, there was no challenge to that process on appeal. 124 The SRA was required to prove the reasonable costs of the repairs to the train. It adduced detailed evidence of not only the actual costs, but the methodology whereby the tender process was conducted, the type of contract entered into and Bombardier's expertise in carrying out the repair work. The accuracy or appropriateness of these matters was not in contest on the appeal. The appellant's challenge to the profit margin was not made out at trial for the reasons given by his Honour, and the appellant has not shown and did not attempt to show, any error in his Honour's analysis in this regard. It did complain that Mr Olsen was not relevantly cross-examined on his opinion. However, in circumstances where the SRA was able to establish that Mr Olsen's assertions were unsupported by primary source material, the absence of cross-examination is not sufficient to require a Court to accept an expert's evidence: see M & EM Hull Pty Ltd v Thompson [2001] NSWCA 359 at [21] per Rolfe AJA (with whom Sheller JA and Davies AJA agreed). At the end of the day, the appellant's case did not rise above a number of unproved assumptions. The reasoning of the trial judge has not been shown to be incorrect. Indeed, his Honour's analysis shows a detailed and reasoned approach to the evidence which was soundly based.

The result 125 I propose that the appeal should be dismissed with costs. 126 GILES JA: I agree with Beazley JA. 127 IPP JA: I agree with Beazley JA. **********

DISCLAIMER - Every effort has been made to comply with suppression orders or statutory provisions prohibiting publication that may apply to this judgment or decision. The onus remains on any person using material in the judgment or decision to ensure that the intended use of that material does not breach any such order or provision. Further enquiries may be directed to the Registry of the Court or Tribunal in which it was generated.