GOVERNMENT INSURANCE OFFICE OF NEW SOUTH WALES v BOWER SUPREME COURT OF NEW SOUTH WALES COURT OF APPEAL KIRBY P, PRIESTLEY and HANDLEY JJA 22 March 1991 and 10 April 1991, 3 July 1991 [1991] NSWCA 107 APPEAL — review of fact finding — findings based, in part, on credibility and impression of witnesses — whether Court of Appeal authorised to set aside judgment — whether findings inconsistent with uncontroverted or incontrovertible facts held: (1) Errors of fact finding were disclosed in certain findings; (2) (per Kirby P) Other findings did not support the conclusions stated or were irreconcilable and unreconciled in the reasons given; (3) (per Handley JA, Priestley JA concurring) The errors showed that the trial judge failed to appreciate that contributory negligence was a live issue; (4) The Court of Appeal was therefore authorised to conclude that an error had occurred warranting the setting aside of the judgment entered; (4) But the Court of Appeal was unable to determine the matter for itself; (5) Accordingly, the matter should be returned for retrial limited to the issues of negligence and contributory negligence, those two issues being irreparably related. Warren v Coombes (1979) 142 CLR 531; Brunskill v Sovereign Marine and General Insurance Co Limited (1985) 59 ALJR 842; 62 ALR 53; Jones v Hyde(1989) 63 ALJR 349; Abalos v Australian Postal Commission (1990) 65 ALJR 11 and Chambers v Jobling (1986) 7 NSWLR | applied.; Winter and Anor v Boynton and Anor, Court of Appeal (England) unreported, The Times, 6 February 1991 referred to. MOTOR VEHICLES - negligence - collision at intersection controlled by traffic lights - whether motor vehicle or motorbike ignored red light - conflict of evidence - proper approach by appellate court to review of fact finding by trial judge - inadequacy and incompleteness of reasoning at first instance - inconsistency of finding demonstrated held: The Court of Appeal was authorised to set aside the judgment at first instance and to require that the matter be retried and should do so. Mifsud v Campbell, Court of Appeal, unreported, 12 February 1991; (1991) NSWJB 29 applied. Kirby P In this appeal by way of rehearing, the appellant challenges fact finding by the trial judge (Yeldham J) and the conclusions which led to the judgment which his Honour entered in favour of the respondent. Because Yeldham J expressed his findings to depend, in part, on his impression of witnesses (and because that is in any case inherent in his findings) the appellant faced the not inconsiderable difficulty of establishing, in the somewhat ungainly language of the authorities, that the trial judge had misused the advantage which he derived from conducting the trial and seeing the witnesses. The appellant argued that, following fully for such advantages and also for the "subtle influences of demeanour" upon his Honour's fact finding, uncontroverted facts were available from the evidence to demonstrate that his Honour had erred in the conclusions which he expressed. See generally Paterson v Paterson (1953) 89 CLR 212, 221; Warren v Coombes (1979) 142 CLR 531, 551; Brunskill v Sovereign Marine and General Assurance Co Limited (1985) 59 ALJR 842, 844; 2 UNREPORTED JUDGMENTS 62 ALR 53, 56; Jones v Hyde (1989) 63 ALJR 349, 351; Abalos v Australian Postal Commission (1990) 65 ALJR 11, 16. The principles in the foregoing authorities are familiar to this Court. They are invoked in virtually every case where an appeal against fact finding is involved. Views may differ about the extent to which judicial decisions can safely be grounded upon the impression which witnesses give in the artificial circumstances of the witness box. Science may warn against lawyerly hubris in that regard. But it is the duty of this Court to conform to the authority of the High Court on this question. That authority is, in turn, derived from earlier English decisions. In England like principles are regularly applied to this day. See eg Powell v Streatham Mann Nursing Home [1935] AC 243, 249 (HL); Winter and Anor v Boynton and Anor, unreported, Court of Appeal, The Times 6 February 1991. In this Court, different and other reasons have been recognised as to why trial judges will often enjoy advantages over the appellate court. Such reasons may rest on firmer foundations than "the subtle influence of demeanour ". See eg Lend Lease Developments Pty Limited v Zemlicka (1985) 3 NSWLR 207, 209f; cf T Bingham, "The Judge as Juror: The Judicial Determination of Factual Issues" (1985) 38 Current Legal Problems, 1. Nevertheless, the Court cannot escape its statutory duty to conduct a rehearing as the Supreme Court Act 1970 s75A requires. If the preconditions are established, it must shoulder the responsibility to correct a wrong done at first instance. That is a function conferred upon it by Parliament. Where authorised to do so, it may correct a finding at first instance, even where this has been based, in whole or part, expressly or by implication, upon the impression as to credibility which witnesses at the trial made upon the trial judge. See eg Chambers v Jobling (1986) 7 NSWLR 1, 11, 20. But where, as here, demeanour has played a part in the trial judge's conclusions, the appellant faces a significant difficulty. Put generally, the appellant must be able to demonstrate, by reference to uncontroverted facts, the error which authorises the appellate court to overturn the decision. That is the burden which the appellant assumed in this appeal. In my opinion it has succeeded. A motor cyclist is involved in a collision The central facts of the case were not disputed. On 5 October 1984 at about 4.30 am a collision took place at the corner of William Henry Street and Harris Street, Ultimo. Mr Craig Bower, then aged 21, was proceeding on his motorbike in William Henry Street towards the intersection. He was travelling east on his way home after a party which he had attended with friends between midnight and 4 am. During that time, by his own account, he consumed a considerable amount of alcoholic liquor. Travelling towards the intersection in a southerly direction in a blue Di Tomoso coupe was Mr Clive Stuart, a company director. Stationary at the intersection, but in Harris Street facing north, and about to make a right-hand turn into William Henry Street, was a truck driven by Mr Douglas Piper. That truck had, shortly before it reached the intersection, been overtaken by another motor vehicle, also travelling north in Harris Street, driven by Mr Bernd Ade, a dental technician. A collision took place in the middle of the intersection between the motor vehicle driven by Mr Stuart and the motorbike driven by Mr Bower. Mr Bower was seriously injured. He brought proceedings against the Government Insurance Office of New South Wales, as the licensed insurer of Mr Stuart's vehicle. Those URGOVERNMENT INSURANCE OFFICE OF NEW SOUTH WALES v BOWER (Kirby P) 3 proceedings came before Yeldham J in May and June 1989. The hearing was devoted to evidence - both oral and in statement form - concerning the recollections of the dramatis personae just named. Evidence was also given by police officers, including evidence in relation to the changing sequence of the traffic lights which control the intersection where the collision occurred. Another issue which was explored at the trial concerned the effect on Mr Bower of the alcohol which was found in his bloodstream by a test taken soon after the accident. Yeldham J resolved the controversies of the trial by entering judgment in favour of Mr Bower. He ordered the insurer to pay him $262,500 as damages together with costs. From his Honour's judgment, the insurer has appealed to this Court. No complaint is raised concerning the quantification of damages. The attack in the appeal was that his Honour erred: (a) In finding that Mr Bower had proved that he had been injured because of Mr Stuart's negligence; and (b) In finding that the insurer had not proved that Mr Bower had contributed by his own negligence to the happening of the collision. The trial was conducted upon the basis that one or other of Mr Bower and Mr Stuart had proceeded through a red light. The evidence accepted by Yeldham J suggested that neither of them had slowed down as they proceeded through the intersection. Each of them claimed to have had the benefit of a green light. On the evidence it was impossible that this was so. One of them was mistaken. The trial judge held that the mistake was that of Mr Stuart and hence that his insurer was liable in law for Mr Bower's damages. The trial judge's finding Yeldham J's finding on liability was stated briefly. Correctly, he posed as the "critical question" whether the traffic control lights were in favour of Mr Bower or Mr Stuart. The insurer relied upon two matters to establish that Mr Stuart had the green light in his favour. First, it pointed to the high concentration of alcohol in Mr Bower's blood. Mr Stuart, on the other hand, was on his way to work not returning home after a party. It was therefore argued that the probabilities favoured a mistake on the part of Mr Bower rather that on the part of Mr Stuart. To support this argument the appellant called an expert on the effects of alcohol on human judgment and motor skills, Professor Graham Starmer. Secondly, it urged his Honour to accept Mr Stuart's evidence as consistent with the other testimony given at the trial. Yeldham J dealt with these arguments in turn. As to the high concentration of blood alcohol, his Honour said: "There is no doubt that the plaintiff had consumed a reasonably substantial amount of beer at a party which he had attended after finishing work at about 11.30 pm. He arrived at the party at about midnight and left just after four o'clock. His blood alcohol content as later ascertained was 0.0146% and hence he was moderately affected by intoxicating liquor, as Professor Starmer made plain in his evidence. That still leaves the issue of whether or not the traffic lights controlling the plaintiff's approach were green. If they were he was entitled to assume that he had the right of way and, in the absence of any indication from the other driver that he was not going to give way to obey the traffic control lights, the plaintiff would be entitled to succeed." (emphasis added) Yeldham J gave two reasons for finding in favour of Mr Bower. They were: 1. That he accepted the truthfulness of Mr Bower; and 4 UNREPORTED JUDGMENTS 2. That his testimony was corroborated by an independent witness Mr Ade. This is what Yeldham J said: Mr Stuart said in evidence that as he approached and entered the intersection the lights were green and in his favour. This is supported by a statement from one Piper, which I admitted pursuant to s14B of the Evidence Act but I did not have the advantage of seeing him or hearing him cross-examined. The plaintiff himself claimed that as he approached the intersection, riding east, the lights in his favour changed to green and this was about 20 metres before the intersection. I regard the plaintiff as a decent young man and I accept the substance of his evidence, despite his consumption of alcohol. It is corroborated by the evidence of an independent witness, Mr Ade who drove north along Harris Street, in the opposite direction to Mr Stuart and who said that, some distance after passing through the intersection, he heard the collision between a blue sports car, which had earlier passed him, and the plaintiff's cycle, and he immediately looked in his rear vision mirror and observed that the lights that he had gone through, and which would control the entry of Mr Stuart into the intersection, were red. If the plaintiff's evidence had stood alone, having regard to the alcohol which he had consumed, I would probably not have found in his favour on the issue of the colour of the lights. But I was impressed with the evidence of Mr Ade, and I am satisfied that he did look immediately he heard the impact and that the lights at that time were red and probably therefore had been red at the time when Mr Stuart entered the intersection in his sports car. In these circumstances, and notwithstanding the alcohol that the plaintiff had drunk, he is entitled to a verdict. There is no contributory negligence." (emphasis added) Alcohol consumption: errors of fact finding There are serious errors of fact finding in Yeldham J's summary of the evidence as to alcohol consumption. The first is that the blood alcohol concentration of Mr Bower was 0.146 and not 0.0146 as stated by his Honour. The true level could even have been higher, having regard to the delay which took place between the impact and the test. The evidence on this point was given by Professor Starmer. It was uncontradicted and substantially unchallenged. In a report of 24 May 1989, Professor Starmer recorded that the blood sample was taken at 5.30 am. This was about an hour after the crash. Having regard to Mr Bower's given height and weight, Professor Starmer expressed the following opinion: "Because the body disposes of alcohol at a constant rate once the peak.... has passed, it is possible to calculate the blood alcohol concentration at some previous time and to be reasonable confident in the estimate. Based on an average alcohol elimination rate... Mr Bower's BAC (blood alcohol concentration) at the time of the crash is likely to have been of the order of 0.161 g/100ml with a range of 0.1156-0.166g/100 ml - which addresses the likely BACs of very slow and very fast alcohol metabolises... Australian crashes of the type in which Mr Bower was involved are about 75% alcohol-related and represent impairment of attention and control movements and, particularly, the inability to perceive and to react correctly to an emergency situation when under the influence of alcohol. These effects are increased in those whose driving experience is relatively limited.... I would expect that there would have been deficits in a number of functions which are important in this context. Peripheral awareness would have been reduced. Speed and distance judgment would have been impaired. Dark adaptation would be greatly extended. Co- ordination would also have been greatly reduced. In short, Mr Bowers ability to notice other vehicles at the URGOVERNMENT INSURANCE OFFICE OF NEW SOUTH WALES v BOWER (Kirby P) 5 intersection would have been adversely affected as would his capacity to judge whether it was safe to negotiate the intersection. Increased aggression and risk-taking behaviour, which are frequently associated with consumption of alcohol and expressed as speeding and/or traffic violations, appear to be factors here. Epidemiologically, crash risk is increased some twenty- fold at a BAC of 0.146g/100ml." Professor Starmer's oral evidence was equally emphatic: "Q: From your qualifications can you give an opinion as to whether or not a person would be affected by alcohol at a reading of.112, or I think you said, to.166? A: Well... I would expect the state of intoxication to range somewhere between substantial and gross. Q: When you say it depends on a person's ability to relate to alcohol? A: Yes. There would be a difference, which would largely relate to the individual's tolerance of alcohol, but I could not accept that there would be anybody in the population who would not be measurably affected." Pressed in cross-examination, Professor Starmer estimated the range of likely blood alcohol concentration at the time of driving at between. 1156 to.166. It was suggested to him that he did not have enough data to express any but a tenuous opinion on the impact of alcohol on Mr Bower. However, he replied: "A: No, I wouldn't agree with you that they are tenuous. I would agree with you if you suggested to me that if I had more information on these matters I would be able to be more precise in my opinion. But I have tested literally thousands and thousands of people over the last however many years it is, and I know that I have never come across an individual with blood alcohol concentration of.146, which is the bottom-line figure, who was not demonstrably impaired. (emphasis added) Yeldham J appears to have accepted the "bottom-line figure". But with the addition of a zero. This produced a figure of.0146 or one tenth of the correct reading supported by the evidence. This may have led him to a view about "moderate" affectation. His Honour used the word "hence" which suggests that he was influenced by the reading recorded by him. None of the other evidence supported his Honour's view. The evidence of Professor Starmer which was not relevantly shown to be unreliable and not attacked in cross-examination, was that the proper description of Mr Bower's affectation was "between substantial and gross". It was such that he would have been "demonstrably impaired". It is to be emphasised that these are apparently reliable expert opinions based upon the incontrovertible fact of the blood alcohol reading taken of Mr Bower one hour after the accident. With every respect to Yeldham J, his treatment of this evidence is unsatisfactory. His statement of the blood alcohol level is wrong. His characterisation of the intoxication as "moderate" is contrary to the expert testimony. There was no given or apparent reason for his Honour to disregard that testimony. Professor Starmer's evidence explains why such a bike rider could have proceeded through traffic signals showing red against him. Alternatively, it explains the rider's state of confusion and even a genuine mistaken recall about the colour of the lights. Whilst Yeldham J was undoubtedly correct to say that even an intoxicated driver is normally entitled to rely upon traffic lights in his favour, a real question is posed by the incontrovertible evidence of intoxication, and the expert testimony based upon it, as to whether the evidence of Mr Stuart was not, in the circumstances, more likely to have been correct. The "corroboration" of the passing motorist 6 UNREPORTED JUDGMENTS That leaves the evidence of Mr Ade. Yeldham J regarded the evidence of Mr Ade that he looked in his rear-vision mirror after the collision and saw that the lights were red "both ways", ie to the north-south traffic in Harris Street as corroborating the evidence of Mr Bower. From this, the inference was drawn that they had been red when Mr Stuart had gone through them. When the detail of Mr Ade's evidence is scrutinized, however, it becomes clear that it is not consistent with the evidence given by Mr Bower. Mr Bower said: "Q: As you went over the hill [in William Henry Street] and passed those traffic lights at the crest of the hill at what speed were you travelling? A: About 40 to 50 kilometres and slowing down. Q: Why were you slowing down? A: The lights at the end [ie at the intersection with Harris Street] were red. Q: When you say 'the lights at the end'? A: At the intersection. Q: This is the intersection between Harris Street and William Henry Drive? A: Yes Q: So what did you do? A: I was slowing down for them. Q: Over what distance were you slowing down? A: From the top of the hill, as I went through the lights at the top of the hill I slowed down to come towards the lights, approximately 20 metres before the intersection the lights went green and I proceeded to accelerate through the lights. His Honour: Q: You would say then that anything that hit you if it came from your left or your right would have gone through the lights against them? A: Yes, your Honour. Counsel: Q: You say you were about 20 metres away when the lights changed to green? A: That is correct... Coming down the hill I was looking to my right-hand side at the traffic lights facing north and I noticed they were red as mine were green and that's why I proceeded to go through." The traffic lights on the right-hand side were those facing Mr Piper's truck. He was not called as a witness. However, in a statement admitted under s14B of the Evidence Act he said that the lights were green to him at the time of impact. Mr Ade said that he had overtaken a truck (obviously that of Mr Piper) about 100 or 150 metres away from the intersection. He was travelling at 60 to 70 kilometres an hour. He went through the intersection. But as he approached it, he noticed "the bike on the left-hand side coming down the hill". At that time the lights were green in his direction and therefore red against the bike. When first seen, the bike was 20 to 30 metres back from Harris Street, just before the contacts (sensors) for the intersection traffic lights. Mr Ade then proceeded across the intersection. About 100 metres further on he passed a blue sports car (obviously that of Mr Stuart) in Harris Street travelling in the opposite direction. He said: "Q: Towards you? A: Yes, and I continued further and then I heard the crash and when I looked through the rear vision mirror I only could see the traffic light on red. Q: How long after the crash did you look back? A: Straight away. Q: And the light you had previously gone through was then red facing? A: Both ways. Q: That is north and south? A: Yes. Mr Ade estimated that, when the crash occurred, he had travelled a further 100 metres north in Harris Street. The difficulty with his evidence is that it really does not corroborate the evidence given by Mr Bower of having a green light entitling him to proceed across the intersection without stopping. URGOVERNMENT INSURANCE OFFICE OF NEW SOUTH WALES v BOWER (Kirby P) 7 The problem is illustrated in the attached diagram showing the intersection. The diagram is not drawn to scale. Mr Ade said that he first saw what must have been Mr Bower on his motor cycle when it was a distance back from the southern edge of the intersection. At that stage Mr Ade estimated that Mr Bower was about 20 to 30 metres back from the intersection. The motor cyclist was travelling at a normal speed. This may have been about 20 kilometres an hour. Mr Ade had the green light. He proceeded through the intersection. Indeed, after the intersection, according to his evidence, he proceeded a further 200 metres before the impact occurred. This means that he proceeded approximately [Editors Note: The diagram on page 13 is unable to be represented on Info-One System, please refer to the hard copy.] 207.5 metres from the point at which he had first seen Mr Bower. In the same time Mr Bower proceeded about 40 metres to the point of impact. This distance is derived from his distance from the intersection when first seen and the approximate point of impact in the middle of the intersection. The evidence showed that William Henry Street was approximately 15 metres wide and Harris Street approximately 19 metres wide. In the same time on Mr Ade's evidence, Mr Stuart, after passing Mr Ade, would have travelled 107.5 metres to the point of impact. But if Mr Bower were travelling at about 20 kilometres per hour, in 4 seconds he would have covered 22 metres. In 7 seconds he would have covered 39 metres. Police testimony given at the trial showed that the traffic control lights governing the intersection went through a cycle from green to amber to red in a space of between 4 and 7 seconds. Upon this basis, if Mr Bower's evidence were correct, and the lights jumped as he approached the intersection, even at his given speed of 20 kph he would have been far beyond the intersection in the time Mr Ade took to traverse the distance which he described in his evidence. Since we know that Mr Bower did not get past the centre of the intersection where the collision occurred, some parts of Mr Ade's evidence must be inaccurate. One or more of his estimates must be wrong: his estimate of (i) his own speed; (ii) the point at which he first saw Mr Bower; (iii) the speed at which Mr Bower was travelling; (iv) the distance past the intersection when he crossed Mr Stuart's vehicle; or (v) the total distance from the intersection when he heard the impact. Counsel for Mr Bower ultimately conceded that the estimates of distances given by Mr Ade must be wrong. But he submitted that the errors of estimation did not undermine Mr Ade's positive evidence that he looked into his rear-vision mirror immediately after the crash, and saw the lights in Harris Street showed red. That was enough, so it was said, to support Yeldham J's judgment in favour of Mr Bower. Yeldham J did not attempt to reconcile the evidence of Mr Ade with that of Mr Bower. Instead, he accepted both. The foregoing analysis of Mr Ade's evidence shows that the evidence of Mr Ade and Mr Bower simply cannot stand together. The only evidence that would have been consistent with Mr Bower's proceeding over such a short distance (whilst Mr Ade and Mr Stuart covered much greater distances) would have been evidence that Mr Bower actually stopped at the traffic signal facing him, and then proceeded through green lights when they changed. Only in this way would his movements be reconciled with Mr Ade's evidence. But this was not Mr Bower's evidence. No satisfactory attempt was made by Yeldham J to analyse these differences. Instead, he appears to have come to his conclusion for the reason which he stated, namely that Mr Bower presented as "a decent young man". With every respect to his Honour, this was unconvincing and 8 UNREPORTED JUDGMENTS unsatisfactory. It falls short of the reasoning which this Court has held that litigants are entitled to receive in the resolution of factual disputes. See Mifsud v Campbell, Court of Appeal, unreported, 12 February 1991; (1991) NSWJB 29. Independent witness - a truck driver There is a further reason which confirms me in my conclusion that this action must be retried on the issue of liability. It is the evidence contained in the statement by Mr Piper which, as I have said, his Honour admitted over objection. His Honour did not refer to this statement in reaching his conclusion on liability. However, when this statement is closely examined, I regard it as supporting the evidence of Mr Stuart. Mr Piper describes how he was standing at the intersection of Harris Street with William Henry Street, in Harris Street with his right-hand blinkers turned on "just moving" and waiting to make a turn. He then describes how the motorbike proceeded at a normal speed straight through the intersection and Mr Stuart's vehicle also went straight through. He said: "I do not think the motorcyclist put his brakes on because he kept going straight and hit the car... at right angles. The car, that is the rear of it, went to its left as if it had been pushed. It kept coming across the intersection and past the truck.... I do not know if the car driver applied his brakes... Both kept going straight ahead. When I first saw the motor cycle it would have been roughly 50 feet from the intersection. After the motorbike hit the car and spun around it finished up under the front of the truck I was driving. I was stationary at this stage... A car stopped at the scene and the driver approached me [clearly Mr Ade] and handed me a card. He told me he was a dentist and could not wait. He said he saw what had happened. He did not say what he had seen... The street lights were turned on. I would say the street lighting there is no good. Although the green lights were showing in my direction as I approached the intersection, I cannot say what colour they were for the car approaching the intersection or for the motor cycle rider. I do not know that they were green in my direction when the collision occurred. I do not know when the lights facing me changed to red as I was too busy with the car driver and motor cycle rider." (emphasis added) The crucial passage in this statement is the assertion that the lights were showing green in the direction of Mr Piper's truck and that he was stationary waiting, with the blinker on, for an opportunity to turn right. The inference readily drawn from this evidence is that the light was also green in the opposite direction in Harris Street, ie facing Mr Stuart as he went through the intersection. The evidence suggests that it was impossible for the light to be red against Mr Stuart and green in favour of Mr Bower yet also green in favour of Mr Piper. Thus, if Mr Piper's statement is truthful, it contradicts Mr Bower's evidence. Although he cannot affirmatively say that there was a red light against Mr Bower (for that was not his concern) the fact that there was a green light in Mr Piper's favour right up to the point of collision makes it impossible, or at least extremely unlikely, that the light was green in favour of Mr Bower. Mr Stuart said that he was particularly alert to his right to proceed through the intersection because he had his eye on the truck (obviously Mr Piper's truck) which had its blinkers showing to turn right. This provided an added reason why he would have been careful to retain the green light when proceeding forward. Yeldham J did not refer to this. Conclusions and orders URGOVERNMENT INSURANCE OFFICE OF NEW SOUTH WALES v BOWER (Kirby P) 9 It is basic to the just resolution of disputed issues of fact at trial that the reasons given to support the judge's conclusion in favour of one party, and adverse to another, should disclose no relevant error of law, no significant error of fact finding and should deal fairly and accurately with at least the most important evidence supporting the case of the party which has lost. The reasons given are subject to scrutiny not only for these qualities but for the internal consistency of the findings which are made. These are features of a rational justice system. They are put to the test by the facility of appeal. It would be a negation of the rationality of our system and of the important right to an appeal by way of rehearing, if such critical features could be swept aside by an incantation about the impression of a witness or by a passing reference to demeanour. The facility of appeal provided by Parliament is to prevent that most unacceptable form of wrongdoing an erroneous decision reached in a court solemnly endorsed by the court's judgment. On some occasions wrongs suspected by the appellate court cannot be righted. On others, the appellate court should accept the possibility that its own assessment may be wrong and that of the trial judge who enjoys undoubted advantages, right. But if error is clearly shown, the error should, if possible, be righted. The erroneous judgment should be set aside. That is the whole point of providing a second look at cases on appeal. The facility which Parliament has provided should not be negatived by the rigid application of formulary qualifications. This Court has not seen any of the crucial witnesses in this case. It is therefore not in a position to undertake the more detailed fact finding which is necessary to reach a supportable conclusion about how this accident occurred and, relatively, who went through the red light. What is needed is a more painstaking consideration, without factual error, of the evidence of Professor Starmer and its relevance, if any, to the happening of the accident. It is then necessary to examine the apparent inconsistencies in the evidence. Mr Bower's evidence of what happened as he approached the intersection must be considered as it appears to conflict with that of Mr Ade. The evidence of Mr Piper must be considered as it appears to contradict that of Mr Ade. At the very least there must be a more convincing demonstration of liability in the insurer than has been afforded in the reasons of Yeldham J. Yeldham J's judgment in favour of Mr Bower cannot stand. It is not possible, at least in this case, to sever the apportionment for any contributory negligence ultimately found from the finding of negligence. However, as the damages assessed by Yeldham J were not contested in this Court, it is sufficient to send the matter back for trial on the issue of liability alone. To this end, the orders which I favour are: 1. Appeal allowed; 2. Set aside the judgment in favour of the respondent/plaintiff; 3. Return the proceedings to the Common Law Division for redetermination of the issues of: (a) the alleged negligence of the driver for which the defendant is liable; (b) the alleged contributory negligence of the plaintiff; and, if necessary (c) the apportionment of responsibility; 4. Order that any judgment for the plaintiff in the new trial be based on the assessment of the plaintiff's damage at the first trial in the sum of $262,500 and that such judgment take effect from 28 June 1989; 10 UNREPORTED JUDGMENTS 5. The respondent to pay the appellant's costs of the appeal but to have, in respect thereof, a certificate under the Suitors' Fund Act 1951; and 6. The costs of the first trial reserved for determination by the judge hearing the second trial. Priestley JA I agree with Handley JA. Handley JA In this matter I have had the advantage of reading in draft form the reasons for judgment of the President. His summary of the evidence and the findings of the trial judge make it unnecessary for me to cover the same ground again. I agree with his analysis of the evidence of Professor Starmer and his conclusion that the trial judge made a number of significant errors in his fact finding based on this evidence. I also agree with his conclusion that the finding by the trial judge that the evidence of the independent witness Bernd Ade "corroborated" the evidence of the plaintiff cannot be supported. I also agree that this Court is not entitled to interfere with the trial judge's finding, based on his acceptance of the evidence of Mr Ade, that immediately after he heard the crash he looked in his rear vision mirror and saw that the lights facing traffic in Harris Street were red. The trial judge found the lights in Harris Street "probably therefore had been red at the time when Mr Stuart entered the intersection in his sports car". On this basis the judge held that the plaintiff was entitled to a verdict and concluded that there was no contributory negligence. His Honour's decision that there was no contributory negligence must have been based on his earlier finding that the evidence of Mr Ade corroborated the plaintiff's evidence that he entered the intersection when the traffic lights facing him were green. The conclusion reached by Kirby P, with which I agree, that the evidence of Mr Ade did not corroborate the plaintiff's evidence in this respect, requires further consideration to be given to the state of the lights when both parties entered the intersection. The evidence of Mr Ade was that he looked back at the traffic lights immediately after the crash and saw that they were red for traffic in Harris Street. This does not establish the state of the lights facing the plaintiff and Mr Stuart when they entered the intersection a few seconds earlier. There was undisputed expert evidence concerning the manner in which the traffic lights at this intersection operated. After turning from green to amber they would remain amber for 3-4 seconds, then change to red. At that stage the lights in all directions would be red. After a further 1-2 seconds the lights in the cross direction would turn to green. The evidence of Mr Farrelly from the Road Traffic Authority was that the time that elapsed between the end of the green light in one direction and the start of the green light in the other direction was between four and seven seconds. It follows from this evidence that Mr Stuart could have entered the intersection on a green, amber or red light at any time within a period of four seconds or more before the lights facing the plaintiff turned green. If the plaintiff entered the intersection from William Henry Street during this period he would have done so contrary to red lights facing in his direction. It is possible that the lights for traffic in Harris Street seen by Mr Ade had changed to red the instant before he looked back and the lights in Prince Henry Street were then still red. In that event the plaintiff would have entered the intersection against a red light and Mr Stuart with either a green or an amber light. \ROVERNMENT INSURANCE OFFICE OF NEW SOUTH WALES v BOWER (Handley JA)I On the other hand the lights In Harris Street may have been red for some seconds before Mr Ade looked back. In that event the plaintiff would have entered the Intersection with the lights in his favour, and Mr Stuart with a red light against him. Depending on the precise findings the conclusion could be that the plaintiff entered the intersection with the lights in his favour or with a red light against him and Mr Stuart entered the intersection with the light.s against him or in his favour. The trial judge found that Mr Stuart had probably entered the intersection when the lights in his direction were red. This is the most favourable finding for the plaintiff that his Honour could have made. I would add at this point that the acceptance of Mr Ade's evidence as to the state of the lights when he looked back, on its own, could not establish on the balance of probabilities what the state of those lights were the instant before he looked back. In these circumstances the judge's finding that it was "therefore" probable that Mr Stuart had driven into the intersection contrary to a red light may not be supportable. However even on this finding it does not follow that the plaintiff himself entered the intersection with a green light in his favour. In my opinion therefore even on the most favourable inference that could be drawn from Mr Ade's evidence there remained a live issue about contributory negligence on which the plaintiff's evidence was uncorroborated. His Honour's error in finding that Mr Ade's evidence corroborated that of the plaintiff, and his failure to analyse the interrelationship between Mr Ade's evidence and the uncontested evidence about changes in the traffic lights means that he failed to appreciate that contributory negligence was a live issue. As I have sought to demonstrate that issue cannot be decided on the basis of Mr Ade's evidence alone. Moreover this Court cannot possibly decide it on the basis of the written record. It follows in my opinion that at the very least the issue of contributory negligence must be remitted for re-trial. Once this conclusion is reached it follows that the judge's finding on the issue of negligence must also be set aside. The issues of negligence and contributory negligence in a case such as this are inseverable. If the judge hearing the new trial found that contributory negligence was established he or she would then be faced with the issue of apportionment which requires a comparison to be made between the culpability and causative potency of the negligence of the defendant and the contributory negligence of the plaintiff. A trial judge who had not decided both issues could not possibly undertake such an apportionment. In my opinion therefore the issues on liability must be re-tried and the action should be remitted to the Common Law Division for this purpose. However since the assessment of damages by the trial judge was not attacked there is no need for that issue to be re-tried. In these circumstances I agree with the orders proposed by Kirby P. 1. Appeal allowed; 2. Set aside the judgment in favour of the respondent/plaintiff; 3. Return the proceedings to the Common Law Division for redetermination of the issues of: (a) the alleged negligence of the driver for which the defendant is liable; (b) the alleged contributory negligence of the plaintiff; and, if necessary UNREPORTED JUDGMENTS (c) the apportionment of responsibility; 4. Order that any judgment for the plaintiff in the new tiral be based on the assessment of the plaintiff's damage at the first trial in the sum of $262,500 and that such judgment take effect from 28 June 1989; 5. The respondent to pay the appellant's costs of the appeal but to have, in respect thereof, a certificate under the Suitors' Fund Act 1951; and 6. The costs of the first trial reserved for determination by the judge hearing the second trial. Counsel for the Appellant: CT Barry Counsel for the Respondent: BF Murray QC / D Lukins Solicitors for the Appellant: Government Insurance Office of New South Wales Solicitors for the Respondent: Sharah Henville and Co