HIGH COURT OF AUSTRALIA GLEESON CJ, McHUGH, GUMMOW, KIRBY, HAYNE, CALLINAN AND HEYDON JJ Matter No S513/2004 THE WATERWAYS AUTHORITY APPELLANT AND DANIEL GERARD FITZGIBBON & ORS RESPONDENTS Matter No S98/2005 MOSMAN MUNICIPAL COUNCIL APPLICANT AND DANIEL GERARD FITZGIBBON &ORS RESPONDENTS Matter No S131/2005 MIDDLE HARBOUR YACHT CLUB APPLICANT AND DANIEL GERARD FITZGIBBON & ORS RESPONDENTS The Waterways Authority v Fitzgibbon Mosman Municipal Council v Fitzgibbon Middle Harbour Yacht Club v Fitzgibbon [2005] HCA 57 5 October 2005 S513/2004, S98/2005 & S131/2005 ORDER 1. The time within which the second and third respondents may apply for special leave to appeal is extended, special leave to appeal is granted, the applications are to be treated as appeals and heard with the appellant's appeal. 2. Each appeal is allowed with costs. 3. Set aside orders 3 and 4 of the Court of Appeal of the Supreme Court of New South Wales made on 3 December 2003 and: (a) In place of order 3, order that there be a new trial generally; (b) In place of order 4, order that the respondents in the Court of Appeal pay the costs of the appeal of the appellant in that Court and that the costs of the trial be costs in the new trial. On appeal from the Supreme Court of New South Wales Representation: Matter No S513/2004 J L Glissan QC with J S Whyte for the appellant (instructed by McCabe Terrill) D F Jackson QC with D A Wheelahan QC and E G Romaniuk for the first respondent (instructed by Paul A. Curtis & Co) B W Walker SC with R J Cheney for the second respondent (instructed by Riley Gray-Spencer) M T McCulloch SC with S P W Glascott for the third respondent (instructed by Phillips Fox) Matter No S98/2005 M T McCulloch SC with S P W Glascott for the applicant (instructed by Phillips Fox) D F Jackson QC with D A Wheelahan QC and E G Romaniuk for the first respondent (instructed by Paul A. Curtis & Co) J L Glissan QC with J S Whyte for the second respondent (instructed by McCabe Terrill) B W Walker SC with R J Cheney for the third respondent (instructed by Riley Gray-Spencer) Matter No S131/2005 B W Walker SC with R J Cheney for the applicant (instructed by Riley Gray-Spencer) D F Jackson QC with D A Wheelahan QC and E G Romaniuk for the first respondent (instructed by Paul A. Cutis & Co) J L Glissan QC with J S Whyte for the second respondent (instructed by McCabe Terrill) M T McCulloch SC with S P W Glascott for the third respondent (instructed by Phillips Fox) Notice: This copy of the Court's Reasons for Judgment is subject to formal revision prior to publication in the Commonwealth Law Reports. CATCHWORDS The Waterways Authority v Fitzgibbon Mosman Municipal Council v Fitzgibbon Middle Harbour Yacht Club v Fitzgibbon Appeal and new trial – Practice and Procedure – New South Wales – Powers of court – Sufficiency of trial judge's reasons – Miscarriage of fact-finding process – Exercise of power to order new trial – New trial ordered by Court of Appeal on a limited factual basis – Substituted factual finding decisive on issue of causation – Whether Court of Appeal was right to order a new trial on this limited basis. Torts – Negligence – Causation – Person suffered injury when entering water from jetty – Absence of a handrail – Public authorities and yacht club responsible for design or approval of jetty – Whether person entered the water deliberately – Contributory negligence. Supreme Court Act 1970 (NSW), s 75A. Supreme Court Rules 1970 (NSW), Pt 51 r 23. 1. GLEESON CJ. The facts of the case appear from the reasons of Callinan J. Three issues were argued before this Court. First, was the Court of Appeal in error in reversing the decision of the trial judge to dismiss the first respondent's action on a factual basis that went to the issue of causation? Secondly, if the answer to the first question is in the negative, did the Court of Appeal err in the exercise of its power to send the matter back for a new trial on a certain, limited, basis? Thirdly, did the Court of Appeal err in its order as to costs? It was the second issue that was the principal focus of attention in this Court. Nevertheless, it is necessary to begin by addressing the first issue, because that sets the context in which the form of order for a new trial was made. 2. The first respondent and a number of other young people, while attending a social function at the Middle Harbour Yacht Club, engaged in some horseplay on the jetty outside the club. One of their number, Nathan Wilmot, was thrown from the jetty into the harbour. A witness, Luke Molloy, said the first respondent was "in the middle" of the group that escorted Nathan Wilmot out of the club onto the jetty, to be thrown in. There was a dispute about that. Soon afterwards, the first respondent, to use a neutral expression, entered the water himself. At the point from which he left the jetty, it was about one and a half metres above water level, and the water was about eight inches deep. He hit his face on the sand and suffered catastrophic injury. 3. The first respondent sued the appellant, and the second and third respondents, for damages for negligence. The jetty (at which boats would tie up, and onto which people would step from boats) had no handrail. The appellant and the other two respondents were, in one way or another, said to be responsible for the design, or the approval, of the jetty and, in particular, the absence of a handrail. The first respondent's case was that he was jostled or pushed by some of the other people on the jetty, lost his balance, and fell into the water. The absence of any handrail was said to be a cause of his injuries. (There was also a complaint about a toe-board that may have contributed to his instability.) There were issues as to whether the absence of a handrail involved negligence, and as to contributory negligence and damages. As will appear, none of those issues were decided at trial. 4. There was a dispute about how the first respondent came to enter the water. On one possible view of the facts, the design of the jetty, and the absence of the handrail, had nothing to do with what occurred. It was part of the defence case at trial that the first respondent had deliberately dived or plunged into the water, not because he overbalanced, but because he thought Nathan Wilmot was in trouble and he intended to go to his assistance. This view of the facts was based on the evidence of Dr Trevithick, who was the emergency registrar at the hospital to which the first respondent was taken by ambulance. The doctor said he recalled that the first respondent, who was lucid, told him that he dived into the water because he thought his friend was at risk of drowning, and that, when he dived, he hit his face on the bottom. Dr Trevithick's evidence was supported by the notes he made in the hospital records. The notes said: "Visiting Sydney with friends for sailing regatta. Had several alcoholic drinks with friends. One friend was thrown into the water at the Spit and pretended to drown so Daniel dived 1.5m into shallow water, striking his mouth on the bottom - he was paralysed immediately." 1. A number of the first respondent's companions, including Luke Molloy, gave evidence about the occurrence. As would be expected, some of them did not profess to be able to give a clear account of how the first respondent came to enter the water, and their evidence was in some respects inconsistent and contradictory. It did not support the account of events recorded in Dr Trevithick's note. The first respondent, who could not recall what he told Dr Trevithick, denied that anything of the kind recorded by Dr Trevithick happened. He said he was jostled or pushed, and lost his balance. 2. The proposition that the first respondent "dived" into the water was not necessarily inconsistent with the evidence that he and his companions gave at trial. For example, one of his companions, Matthew von Bibra, made a statement in which he said: "I was looking in Dan's direction and I saw him fall forward towards the water as if he had for some reason lost his balance. At this time I could see that there were people either side and behind him. As he fell forward I could see his arms outstretched trying to regain his balance. But at the 'point of no return' Dan put his arms in front of him like a person diving and then he hit the water." 1. What was, however, completely inconsistent with the first respondent's case, and, if accepted, destructive of his case on causation, was the proposition that he entered the water deliberately. Again, the explanation recorded by Dr Trevithick, which was that he wanted to help Nathan Wilmot, and the proposition that the first respondent entered the water voluntarily, did not necessarily stand or fall together, although as a practical matter no other reason was given as to why the first respondent would have jumped or dived in. 2. The trial judge, Newman AJ, accepted the evidence of Dr Trevithick. That evidence was supported, not only by Dr Trevithick's note, set out above, but also to some extent, although less cogently, by an ambulance record and by records made by other members of the hospital staff. The trial judge said: "[The defendants] all alleged that the plaintiff did not enter the water as a result of falling because he had been either jostled or pushed but rather as a result of him deliberately entering the water by diving from the jetty's edge. In other words, the defendants contended that the plaintiff's entry into the water was the result of his own deliberate act and that the presence or absence of a handrail on the jetty's edge was irrelevant to his mode of entry." Having recorded in that form a question of fact relevant to one of the issues in the case (causation), Newman AJ went on to resolve that issue in favour of the defendants on the basis of his acceptance of the evidence of Dr Trevithick and his preference for that evidence over the evidence of the first respondent and his companions. 1. The first respondent appealed to the Court of Appeal. The grounds of appeal included complaints that Newman AJ had failed to give adequate reasons, and had failed adequately to consider the whole of the evidence. The notice of appeal sought either a resolution of the issue of liability in favour of the plaintiff and a new trial on damages or, alternatively, a new trial on all issues. In the course of argument in the Court of Appeal, brief, although rather dismissive, reference was made to an intermediate possibility. It was "in play", but very little argument was devoted to it. 2. In the Court of Appeal, Foster AJA, with whom Meagher JA and Santow JA agreed, undertook a close examination of all the evidence. He concluded that Newman AJ had failed to give adequate reasons for his decision. In particular, he concluded that Newman AJ had failed to examine with sufficient care the evidence of the first respondent's witnesses, or to explain why important parts of that evidence were apparently rejected. As to Dr Trevithick, he said his evidence was largely dependent on his notes, that the notes were made for purposes of medical management, that Dr Trevithick's record of the purpose of the first respondent's entry into the water was not supported by any other evidence, and that, having regard to the rest of the evidence as to what happened, the supposed account of an attempted rescue was "nonsensical". He accepted Newman AJ's finding that Dr Trevithick was an honest witness, but concluded that his notes were unreliable. The finding that the first respondent deliberately entered the water was "glaringly improbable". 3. What has been said so far explains how the first issue in the appeal to this Court arises. On that issue, as the judgment of Foster AJA demonstrated, the reasoning of the trial judge failed to deal adequately with important aspects of the evidence called on behalf of the first respondent at trial. Indeed, much of that evidence was neither referred to nor analysed. Foster AJA concluded that "the body of such evidence ... could not and should not have been rejected on the basis of the evidence given by Dr Trevithick and the material in the notes." 4. It was one thing for the Court of Appeal to conclude that the trial judge had not done justice to the first respondent's case, and that there should be a new trial. The further conclusion, that the evidence of the first respondent's witnesses should have been accepted, and that a finding based on Dr Trevithick's evidence and notes was "glaringly improbable", is another matter. A puzzling feature of the case is that, on the evidence called by the first respondent, there appears to be no rational explanation of how Dr Trevithick could have conceived the idea, which he recorded in his notes, that the first respondent thought his friend in the water was in trouble and went in to assist him. Neither the first respondent nor any of the people who took him to the hospital acknowledged telling Dr Trevithick that. It may be that the doctor misinterpreted or misunderstood something that was said to him by the first respondent. The first respondent did not claim to know the depth of the water where he entered it. The "glaring improbability" of the conclusion that he entered the water deliberately was thought to arise mainly from the fact that, on the evidence of the first respondent and his witnesses, there was nothing about the behaviour of Nathan Wilmot to indicate that he had any need for assistance. 5. As to the second issue, it was obvious that, unless the Court of Appeal was going to uphold the decision of Newman AJ, there would have to be a new trial. On any view, there were unresolved issues concerning liability, contributory negligence, and damages. All the evidence relating to those issues was called before Newman AJ. As to liability, for example, there was a lively dispute about whether there should have been a handrail on the jetty. As to contributory negligence, the dynamics of the situation in which the first respondent was involved, and the nature and extent of his participation, were the subject of conflicting evidence. 6. The Court of Appeal decided that the matter should go back for a new trial, but with a positive finding in favour of the first respondent on the question of fact identified by the trial judge. The Court of Appeal ordered: "That there be a new trial of the action conducted on the basis that it is established in favour of the appellant that, through being jostled or pushed, he lost his balance and fell from the jetty into the water." 1. The Court of Appeal had wide power under s 75A of the Supreme Court Act 1970 (NSW) and the Supreme Court Rules Pt 51 r 23. It could, for example, direct admissions to be made by any party for the purpose of a new trial. The question is not the width of the power, but the appropriateness of the manner of its exercise in the circumstances. There are two closely related considerations. Should the Court of Appeal have substituted for the finding of Newman AJ a positive finding as to how the first respondent came to enter the water? Should it then have ordered a new trial "on the basis of" that finding? 2. There was some argument in this Court about the precise effect of the order made by the Court of Appeal. There is little doubt about what was intended. The Court of Appeal took Newman AJ's formulation of a question of fact that arose from the way the case was argued at trial. It was a fact relevant to a fact in issue, that is to say, causation. If the first respondent deliberately leapt into the water, then the absence of a handrail, even if one was required by a duty of care owed to the first respondent by one or more of the defendants, had no causal connection with the damage. The Court of Appeal considered that it should both reverse the trial judge's finding of fact, and preserve for the first respondent the benefit of that reversal at the new trial. 3. Conducting a new trial on the basis of a certain view of the primary facts is not impossible. Expressing the order in terms of the Act and the Rules, it is as though the defendants were required to make an admission of fact. However, the trial judge's finding of fact was based upon consideration of the credibility of the first respondent, and the witnesses called in his case, as well as Dr Trevithick. Furthermore, it may have implications, unexplored by Newman AJ or the Court of Appeal, with respect to the credibility of those witnesses which touch upon other unresolved issues in the case. 4. The evidence about the circumstances in which the first respondent entered the water was contradictory and confused. The dispute as to whether the first respondent deliberately entered the water to assist Mr Wilmot was not the only area of disagreement or uncertainty. The witness Luke Molloy gave evidence that was significantly different from that of the first respondent in some respects, although on the question of seeking to assist Nathan Wilmot it supported the first respondent. Luke Molloy said the first respondent was in the middle of a group of 10 people who took Nathan Wilmot out of the Club to throw him in the water. If, as Luke Molloy's evidence might suggest, and contrary to the first respondent's evidence, the first respondent was an active participant in the dunking of Nathan Wilmot (a matter about which the Court of Appeal made no finding), and a small crowd of young people, including the first respondent, had been milling around on the jetty, then the problem for a trial judge was not so simple or clear-cut as asking: did he jump or was he pushed? The situation could have been more complex, and issues both of causation and contributory negligence could be affected by a decision, at a new trial, as to exactly what went on. 5. It is apparent that the Court of Appeal made the focus of its attention the trial judge's finding on a specific factual question. However, the facts relevant to the issues that arose on the pleadings were wider and more complex. In ordering a new trial on both liability (including contributory negligence) and damages, and yet at the same time seeking to preserve its own finding on a particular factual question, the Court of Appeal did not in its own reasons examine the implications for the practical conduct of the new trial. 6. The fact which the Court of Appeal found (that the first respondent, through being jostled or pushed lost his balance and fell into the water), and then by its order made the basis on which the new trial was to be conducted, is only one of a number of facts relevant to facts in issue in the case, and it is not a fact that can be isolated from all other facts that remain in controversy. Newman AJ isolated it because, if decided one way, it meant that the first respondent's case must fail on the issue of causation, and no other issues arose for decision. The converse, however, does not hold. There remain for decision a number of unresolved questions which will depend upon the reliability of the evidence of other witnesses, including that of the first respondent. At a new trial, the trial judge will have to hear evidence, and make findings about, the circumstances of the accident. The evidence will not necessarily be the same as the evidence at the first trial. It could be significantly different. It is in the interests of justice that the judge hearing the second trial should be in a position to make a fresh appreciation of the whole of the relevant evidence, unconstrained by an artificially isolated assumption that reflects the first respondent's forensic success in the Court of Appeal. The appellant should succeed on the second issue. 7. The third issue concerns costs. The Court of Appeal, despite ordering a new trial, ordered the other parties to pay the first respondent's costs of the first trial, notwithstanding that all the other issues in the case were litigated before the judge. 8. This unusual costs order is consistent with, and reflects, the Court of Appeal's approach to the second issue. In a sense, it makes plain what the Court of Appeal was seeking to achieve. It was regarded as the logical corollary of what was done in relation to the order constraining the basis on which the new trial was to be conducted. It can only be justified on the basis that the first trial is to be treated as a trial of a separate issue, being the factual question formulated by Newman AJ. Having reversed Newman AJ on that question, and substituted its own finding, which was to become the basis for the continuation of the litigation, the Court of Appeal ordered the defendants to pay the costs of the first trial. That, however, simply exposes the problem in another form. The question formulated by Newman AJ could never properly have been identified as an issue for a separate trial. It is not a separate issue. It is one of a number of interconnected questions of fact relevant to facts in issue. If resolved in one way, it was decisive on the issue of causation, but if resolved differently it was not decisive of any issue, and the manner of its resolution could affect, and could be affected by, the approach taken to other questions. 9. I respectfully disagree with the Court of Appeal's decision on the third issue for the same reason as I disagree with its decision on the second issue, but in my view they are really only two sides of the one coin. 10. I would make the following orders. The second and third respondents, who seek special leave to appeal out of time (the appellant having previously been granted such special leave), should have extensions of time to apply for special leave to appeal, such special leave to appeal should be granted, and the applications should be treated as appeals and heard with the appellant's appeal. The appeals should be allowed with costs. Orders 3 and 4 of the Court of Appeal should be set aside. In place of order 3 it should be ordered that there be a new trial of the action generally. In place of order 4 it should be ordered that the respondents in the Court of Appeal pay the costs of the appeal of the appellant in that Court, and that the costs of the trial be costs in the new trial. 11. McHUGH J. I agree with the orders proposed by the Chief Justice in this matter. 12. I agree with his Honour's reasons and also with the additional reasons of Hayne J. 13. GUMMOW J. Orders should be made as proposed by the Chief Justice. 14. I agree with the reasons of the Chief Justice and of Hayne J. 15. KIRBY AND HEYDON JJ. At Christmas time in 1996 Mr Nathan Wilmot won a national championship yachting race in Hobart. It was a tradition in yachting circles to throw winners of such races into the water, but it proved impossible to comply with this tradition at that time in Hobart. On the evening of 29 March 1997, a yachting party at the Middle Harbour Yacht Club took place in connection with a regatta held in Sydney. Mr Wilmot attended. The plaintiff, a successful and promising yachtsman, also attended. He had never been there before, and he arrived after dark. Some guests decided to throw Mr Wilmot into the water in belated compliance with the tradition not observed in Hobart. This they did. The plaintiff saw him in the water. The plaintiff then himself entered the water from a jetty headfirst. He was rendered quadriplegic. 16. The plaintiff described his injury as having happened because he was accidentally nudged from behind by other watchers. He lost his balance. He could not recover it because his feet were "butted up against" a toe-board about 200 millimetres high on the edge of the jetty. He began to fall. There being no handrail to restrain his fall, he tried to counteract it by swinging his arms. However, he hit the shallow bottom of the harbour. 17. According to the defendants, on the other hand, the plaintiff's injury happened because he deliberately dived into the water. 18. The defendants' case was not supported by any eyewitnesses. Its prime support came from an alleged admission recorded by Dr Trevithick, who treated the plaintiff on his arrival at hospital, to the effect that he had dived because he thought Mr Wilmot was drowning. The plaintiff's case was supported by several eyewitnesses, both to the fact that he did not enter the water voluntarily, and to the fact that Mr Wilmot, far from drowning, was standing up in shallow water. The plaintiff's case was also supported by a complaint of being pushed which he made immediately after the injury. 19. The trial judge found for the defendants in reliance on Dr Trevithick's evidence. The Court of Appeal, while accepting Dr Trevithick's honesty and in large measure his reliability, found the trial judge's finding that the plaintiff had intentionally entered the water to be glaringly improbable. It reversed the trial judge's finding, and ordered that a new trial be conducted on the limited basis that it was established in favour of the plaintiff that, through being jostled or pushed, he lost his balance and fell from the jetty into the water. The consequences of the trial judge's approach 1. The trial in this matter took place over 10 days – a working fortnight. But it was not heard in a fortnight. For reasons which may have been good but are not clear, it was heard over more than two months – on 5-6, 9-10, 12-13, 16 and 19 September, 17 October and 15 November 2002. Thus, although it took 10 days, it proceeded in a staggered manner. This was likely to have disrupted continuity. It was also likely to have inflated costs. 2. The first defendant was an agency of the Government of the State of New South Wales. It owned the land, and was the authority having the responsibility of approving the structures on it. The second was the yacht club; it was lessee and occupier of the land, and had organised the party. The third was the local council. As between the plaintiff and the defendants, the defendants put in issue duty, breach, causation, contributory negligence and quantum. However, the case between the plaintiff and the defendants, although turning on detailed evidence, some of it difficult to reconcile, was relatively simple. As between the defendants, each of the defendants' cross-claims contended that if the plaintiff succeeded on the basis that there should have been a handrail, the legal responsibility for failure to provide it lay with some other defendant. The issues between the defendants were of considerable legal complexity. 3. The trial judge was an Acting Justice of the Supreme Court of New South Wales. He delivered a reserved judgment on 20 December 2002. That judgment dealt with only one of the questions in the case. It was not a question isolated by the pleadings. The trial judge described it several times as the "prime factual issue" or "prime factual case". Indeed it was, in the sense that if the trial judge found against the plaintiff on that issue, and the plaintiff enjoyed no success on appeal, it was inevitable that the plaintiff would have to lose. It was an issue which went to an aspect of causation, and to an aspect of the defendants' contributory negligence case. But the trial judge dealt with no other issues. All the members of this Court are of opinion that there must be a new trial at least on those other issues. Ordering a new trial is "[i]n all cases ... a most deplorable result … ."[1] In this case it is a scandalous result. In our view the Court of Appeal so regarded it and properly turned its attention to how, within the large powers that Parliament has conferred on that Court, such a scandalous outcome might properly be palliated. 4. The trial judge had power to order a question to be tried separately from other questions pursuant to the Rules of the Supreme Court of New South Wales, Pt 31 r 2, which was then in force. He did not exercise it. It is notorious that the course of ordering that a preliminary separate question be tried, and deciding the case on that question, rather than deciding the case on all issues, is a course which can create graver difficulties than those which it is intended to solve. This case illustrates that even graver difficulties can flow from deciding a case on a single issue isolated by the trial judge without instigation by or consent from the parties. 5. The course adopted by the trial judge preserved to the plaintiff an appeal to the Court of Appeal as of right, since an appeal from a decision on a question decided separately pursuant to a Pt 31 r 2 order would have required leave: Supreme Court Act 1970 (NSW), s 103. But in other respects the course adopted has had calamitous effects, and not only for the plaintiff. The point of a Pt 31 r 2 order is that usually it relates to some critical preliminary question which, if decided one way, will terminate the proceedings and save the parties the costs of litigating other questions. Yet here the parties litigated all the issues at the trial, they have not been saved any costs, and as a result of the trial judge's failure to decide the other questions in the case the parties are now exposed to the costs and other pains of a second trial which a different course on his part may have rendered unnecessary. That is because had the trial judge decided all the issues in the case, treating each conclusion adverse to the plaintiff as a separate ground of decision, it might be that in the Court of Appeal or this Court it would have become apparent that the plaintiff could not succeed in a second trial whatever errors the trial judge made in deciding the "prime factual issue". 6. Instead the plaintiff, and each of the three defendants, are now to be exposed to a new trial on all issues (on the view adopted by Gleeson CJ, McHugh J, Gummow J and Hayne J) or on all issues but the "prime factual issue" (on the view adopted by the Court of Appeal and proposed by Callinan J). Whatever form it takes, the second trial could well last almost as long as the first. It may last longer if the defendants adopt different tactics at the second trial from those adopted at the first, as from time to time in their submissions in this Court they suggested they would. 7. The proceedings caused, and continue to cause, each defendant to be exposed to the risk of a multi-million dollar verdict. The Court knows nothing of whether any of the defendants can rely, in whole or in part, on the benefit of a contract of insurance. Nor does the Court know whether the other party to that contract is, or will remain, solvent. Even if the matter is approached in non‑monetary terms – that is, whether or not there are solvent insurers in place, and whether or not each defendant can face their absence, if they are absent, with equanimity – there will be individuals within the camps of the three defendants to whom the first trial caused distress, and to whom a second trial will cause additional distress. That distress will take the form of "strain" and "anxiety" – for they, like the plaintiff, must, before the trial started, have had a "legitimate expectation that the trial [would] determine the issues one way or the other."[2] If, on the other hand, the matter is approached at the crude level of the defendants' financial self-interest, although the Court of Appeal's order that they pay the costs of the first trial will not stand, they will be much worse off. They will suffer financial costs and lost opportunity costs not capable of being repaired by a costs order even if they win the second trial. They will have doubled their exposure to that part of their costs not covered by a costs order in their favour. And an order that the plaintiff pay the costs of both trials will not be much comfort. The Court knows nothing of the plaintiff's financial position, but ordinarily the chances of a man rendered quadriplegic at the age of 20 being able to meet an order to pay the defendants' costs of one trial, let alone two, nearly a decade later are extremely remote. 8. In turn, the plaintiff is in an even worse position than the defendants. For him the strain and anxiety of a second trial will be very great. The legitimacy of his expectation that the trial of his action would determine all the issues was very high. Quite apart from the ordinary emotions he would have experienced on losing the case, his disappointment about the obstruction of that expectation must have been very painful. He did not ask the trial judge to decide the "prime" issue in isolation. Indeed, the "prime factual issue" did not emerge over the years between the accident in 1997 and the start of the proceedings in 1999. Thus each of the defendants pleaded that the damage suffered by the plaintiff had been contributed to by his negligence, but none of the particulars relied on alleged that he had intentionally entered the water. Nor did the "prime factual issue" emerge over the years between the start of the proceedings in 1999 and the trial in 2002. Thus it did not appear in differential case management documents filed by the defendants before the trial. Rather, in those documents the first and third defendants did not deny or seek to qualify the plaintiff's contention in his differential case management document that he simply "fell … and struck his head on the sandy bottom" in circumstances where there was no hand railing or warning as to the depth of the water. The second defendant went further, admitting the correctness of the evidence which the plaintiff eventually gave: "[T]he Plaintiff was pushed from the jetty area and struck his head on the sandy bottom." It is not clear whether that significant admission was ever withdrawn. 9. The "prime factual issue" only emerged for the first time, as an issue between the parties, during the cross-examination of the plaintiff, and over the unavailing protest of counsel for the plaintiff. The contribution which the defendants made to what the trial judge did was to urge the contention that the plaintiff had to fail if the "prime factual issue" were decided in their favour. At trial, the plaintiff pointed to the absence of a handrail, while, as the trial judge recorded, "the defendants contended that the plaintiff's entry into the water was the result of his own deliberate act and that the presence or absence of a handrail on the jetty's edge was irrelevant to his mode of entry." The plaintiff, like the defendants, has been left with the consequences of the "prime factual issue" being decided against him by the trial judge in a manner which, in the opinion of the Court of Appeal and of this Court, is so erroneous that there must be a new trial. 10. The quality of, and the gaps in, the trial judge's reasoning, as Foster AJA said while delivering the leading judgment of the Court of Appeal, "could not … have failed to leave the [plaintiff] with a significant sense of grievance. He would be not only disappointed in the result but disturbed by it."[3] 11. Counsel for the plaintiff took up these terrible words in the following submission to this Court: "[T]he judge was dealing with someone who had suffered, very suddenly, injuries which changed him from being a prominent young Australian athlete to someone who is massively crippled. The case did not, with respect, merit being dealt with by, in effect, seizing on a basis which would involve the necessity to resolve the fewest issues or a basis which would obviate the need to determine more complicated questions. Your Honours, one is left, and I say so with respect, with the distinct and somewhat disturbing impression that the primary judge's approach was tailored … to arrive at a situation which had the result that the more complex issues did not have to be dealt with. That may well account for the underlying tone of a degree of distress which one sees, we would submit, in the reasons of the Court of Appeal and a distress which reflects a perception that the Supreme Court had not really performed its function and that the approach to the resolution of the case at first instance had been one which was expeditious rather than that which was proper. In our submission, the Court of Appeal adopted the correct approach in deciding the issue itself." 1. In our view, this was entirely correct. It would, incidentally, have been equally correct if the plaintiff had not been a prominent athlete, or not young, or not Australian, or less than massively crippled. Unless there are very good reasons to the contrary, personal injuries cases at least merit treatment of the kind advocated by counsel for the plaintiff. The result of that treatment not being given to the plaintiff in this case is completely unsatisfactory. The approach of confining the decision to the "prime factual issue" overlooked, or betrayed an indifference to, the extreme and well-known difficulties which injured plaintiffs without assets have in mounting complex litigation against defendants who, without impropriety, are seeking to take every step the law affords them to preserve their positions. They may have to marshal lay witnesses not necessarily sympathetic to them. They may have to seek documents from the defendants, or from third parties who may not be amenable to that course. They may have to find expert witnesses and persuade solicitors to pay them. They may have to appeal to the charity of legal advisers prepared to fund litigation without any certainty that either the just fees of the unpaid advisers will ever be paid, or the other expenditures which have been made by those advisers will ever be reimbursed. The need for a second trial necessarily involves complete or partial repetition of each of these steps. The willingness of witnesses and others to assist is likely to decline. The capacity of witnesses to remember events nearly a decade ago will be reduced. Litigants in Australian courts in this position – indeed any litigants in those courts – deserve better treatment than the parties in this case received. 2. There are, at least potentially, five issues. Did the defendants obtain natural justice in the Court of Appeal? 1. The first issue is whether the defendants were denied natural justice in the Court of Appeal on the question of whether a new trial limited to only some issues should be ordered. There is no utility in examining the point, since the parties have had a full opportunity in this Court to debate what order, if any, should be made about a new trial. The defendants kept trying to link this issue to other issues, but in truth it had no relevance to those other issues. Should a new trial of any kind have been ordered? 1. Errors of the primary judge. The second issue is whether the Court of Appeal erred in concluding that a new trial of any kind should be ordered. It did not err for two reasons. First, the trial judge made numerous errors in the reasoning he did set out. Secondly, he failed fully to set out the reasoning which might have supported the conclusions he reached. No party contended that the Court of Appeal should have decided the issues which the trial judge had not decided. The only possibility left was a new trial. 2. The evidence of Mr Treharne. The first reason centres on errors of the trial judge in not dealing with some key evidence; and, while recording other evidence, in not analysing the totality of it and explaining how, so far as it was apparently contradictory, parts of it could be reconciled or rejected. A good example concerns one important witness whom the trial judge did not mention, Mr Treharne. In his evidence he denied that the plaintiff dived in and said that he saw the plaintiff enter the water "with his arms waving to try and stop himself" before extending his arms to the front. Counsel for the second defendant conceded in this Court that the evidence could not be dismissed as "trivial". That evidence was strongly contradictory of the conclusion favoured by the trial judge that the plaintiff had "deliberately [entered] the water by diving". Mr Treharne was not cross-examined to suggest that the plaintiff's arms were not waving in the manner described, nor that his denial of diving was false, nor that the plaintiff intentionally entered the water. 3. Counsel for the first defendant contended that it was wrong for the Court of Appeal to rely on Mr Treharne's evidence to reverse the trial judge. Counsel cited the following passage[4]: "[W]here a trial judge has made a finding of fact contrary to the evidence of a witness but has made no reference to that evidence, an appellate court cannot act on that evidence to reverse the finding unless it is satisfied 'that any advantage enjoyed by the trial judge by reason of having seen and heard the witnesses, could not be sufficient to explain or justify the trial judge's conclusion'." However, the Court of Appeal was satisfied of this in view of the trial judge's other omissions and errors, and rightly so. 1. Analysis of eyewitness evidence. The plaintiff called evidence of the relevant events from other eyewitness observers (Mr James, Miss Roberts‑Thompson, Mr Roberts-Thompson, Mr Molloy and Mr von Bibra). Their evidence suggested that the plaintiff's entry into the water was not deliberate. The trial judge failed to explain how this eyewitness evidence could be reconciled with his conclusion. Nor, if he thought the eyewitness evidence could not be reconciled with his conclusion, did he explain why the eyewitness evidence was to be rejected. In fact, the trial judge cannot have rejected this oral evidence sub silentio. It would have been extremely difficult to do so in view of the way in which the witnesses were cross-examined. As the Court of Appeal pointed out of these witnesses[5]: "No attack was made on their honesty in cross-examination. No suggestion was put to them that their evidence was fabricated in order to assist the [plaintiff]. Nor was it suggested, in cross-examination, that they were grossly mistaken in their observations of the events of the evening or in their recall of them. … [T]here were no questions in cross-examination to the eye-witnesses, to the effect that what they had observed was, throughout, a deliberate dive. ... [T]he question most likely was not put because the [tenor] of the testimony of each witness was such that it was obvious that such a proposition would have been unequivocally denied. There is, after all, an obvious difference in appearance between an unintended fall and a deliberate dive; this is so, even where one who has accidentally fallen towards water is seen attempting to recover his situation by endeavouring to convert his uncontrolled fall into a dive, before entering the water." 1. These conclusions of the Court of Appeal are correct. Thus while it was suggested to Mr James that his vision of the plaintiff's entry into the water was obscured, he denied this, and the matter was taken no further. Miss Roberts‑Thompson was not cross-examined on her observations, in particular on her impression that the plaintiff did not appear to be diving. Mr Roberts‑Thompson was only cross-examined to show that he saw only part of the plaintiff's entry into the water. Mr Molloy was cross-examined to suggest that he could not see the plaintiff's entry into the water clearly, but without avail. Mr von Bibra agreed that he had not mentioned the account he gave in evidence in chief when interviewed by lawyers on 5 March 2002, but said that he had since refreshed his memory from an earlier statement he had made on 17 April 2000, of which he did not have a copy on 5 March 2002. Mr von Bibra, who gave evidence pursuant to a subpoena, denied fabricating his evidence, and the 17 April 2000 statement confirmed it. The trial judge did not reject the evidence of either Mr von Bibra or any of the others. 2. The second defendant, the Middle Harbour Yacht Club Ltd, being the organiser of the evening, was the best positioned of the defendants to call any evidence from the 50-150 people present of how the plaintiff entered the water if it were truly contended that the consistent evidence of the eyewitnesses was to be rejected. It called one person present to prove that those attending the function were well behaved, but that witness did not observe the plaintiff's entry into the water. The trial judge did not refer to the failure of the second defendant, or the other defendants, to call evidence contradicting the eyewitness evidence called by the plaintiff, or to the significance of this failure. 3. Indeed, the trial judge appears to have accepted the honesty and reliability of the eyewitness evidence called by the plaintiff (apart from that relating to one aspect of Mr Molloy's) by saying that he would have found for the plaintiff on the "prime factual issue" but for Dr Trevithick's evidence and the medical notes. The trial judge certainly did not criticise the demeanour of any of these witnesses, or point to any significant unreliability or improbability in their evidence. 4. The first defendant described this eyewitness evidence as being "internally inconsistent". Whether or not Mr Molloy was not wholly consistent with other witnesses on the subject of the plaintiff's involvement with Mr Wilmot's entry into the water, there were no inconsistencies in his account of the plaintiff's entry into the water. Indeed, counsel for the second defendant was right to concede that so far as there were differences between the eyewitnesses in their accounts of the plaintiff's entry into the water, those differences tell in favour of their testimonial reliability: they are what one would expect of honest witnesses doing their best without having put their heads together. 5. The first defendant referred to this evidence as "utterly incomplete" and as "snippets". While it may be that the perception of some witnesses was incomplete in the sense that each saw particular aspects of the plaintiff's entry into the water, these perceptions were not "utterly" incomplete, and Mr von Bibra's were not incomplete at all. The first defendant also said that most of this evidence stemmed from statements made about a year after the event. Nothing was made of this in cross-examination, and it was not suggested that the oral evidence did not reflect actual recollections of what had been observed. 6. Failure to deal with Mr Moon's evidence. The trial judge referred to Mr Moon's evidence that when he approached the plaintiff and assisted him immediately after he had been dragged to the shoreline, the plaintiff asked "Who pushed me in?". But the only discussion of it related to the question whether the evidence should have been given in the plaintiff's case in chief or in reply. That is surprising, in view of the fact that the debate during the trial about the admissibility of what Mr Moon said did not turn on the difference between a case in chief or a case in reply, but on whether s 108 of the Evidence Act 1995 (NSW) applied, or whether s 64 applied, and, if it did, whether reception of the evidence was barred by the failure of the plaintiff's legal advisers to comply with the requirements of prior notice imposed by s 67. 7. Nothing was made of any admissibility point in this Court, save for reminders that Mr Moon's evidence was hearsay. Since the plaintiff (the person who "made the representation" reported by Mr Moon) gave evidence, Mr Moon's evidence was in fact plainly admissible under s 64(3), to which s 67 does not apply[6]. At least for that reason, the trial judge's rulings that the evidence was admissible was correct. In view of the time at which and the circumstances in which the plaintiff spoke, Mr Moon's evidence in chief about the plaintiff's question was powerful evidence. It was repeated in cross-examination. Although in cross-examination counsel for the first defendant suggested reasons to Mr Moon as to why his powers of perception and recollection might have been affected by the events of the evening, there was no direct challenge to the relevant evidence. Counsel for the second defendant did not cross-examine Mr Moon on that topic, and counsel for the third defendant did not cross-examine Mr Moon at all. 8. In these circumstances it would have been difficult for the trial judge to reject Mr Moon's evidence. Indeed he did not say that he rejected it. Nor did he attempt the impossible task of reconciling it with his conclusion that the plaintiff deliberately entered the water. 9. The failure to deal with the evidence of Mr Wilmot. The trial judge mentioned Nathan Wilmot early in his reasons. He said that he would "return to the evidence relating to Wilmot later". He did not do so. 10. Mr Wilmot's evidence was hearsay – a statement signed on 20 February 1998. The key elements of it were received without objection, because Mr Wilmot was in Sardinia. Despite the inability of the defendants to cross-examine Mr Wilmot, his evidence had considerable power. He was in the water when the plaintiff entered it, he was the first person to assist the plaintiff, and he dragged the plaintiff ashore. His evidence was inconsistent with any idea that after he had been thrown into the water he was in any distress or was feigning it, and there is no other firsthand evidence that supports that idea. 11. The plaintiff's evidence in chief was to the effect that Mr Wilmot was standing no more than knee deep in the water. This was repeated in cross‑examination. He denied that Mr Wilmot was lying in the water as if pretending to drown, or that the plaintiff had ever told anyone that. Counsel did not cross-examine on that denial, which was repeated in re-examination. The plaintiff's evidence that Mr Wilmot was standing in water which was no more than knee deep without conveying or feigning distress was corroborated by Mr James, Miss Roberts-Thompson, Mr Roberts-Thompson, Mr Molloy, Mr von Bibra, Mr Treharne and Mr Moon. 12. There were factors pointing strongly to the reliability of Mr Treharne, in particular, on this issue: as he was Mr Wilmot's cousin, he was likely to have observed Mr Wilmot in the water closely, so as to ensure that he was safe. There was either no cross-examination of these witnesses on the point or, in the case of Mr Molloy, no effective cross-examination. Even in Mr Wilmot's case, it was not suggested that his behaviour was indicative of some form of distress, whether actual or feigned. 13. Of Mr Molloy's evidence the trial judge said: "Molloy's evidence is clearly contradictory of the plaintiff's evidence relating to his part in the manner in which Wilmot entered the water. Molloy has the plaintiff as an active participant in the dunking of Wilmot whereas, of course, the plaintiff maintains his part was no more than a spectator. Not only that, Molloy deposes that Wilmot's actions in the water involved him carrying out a range of activities which were much more extensive than him merely standing up as all other witnesses called by the plaintiff deposed to. The contradictory nature of Molloy's evidence is of some importance when the evidence called by the first defendant in the matter is taken into account." 1. This passage is hard to understand. Mr Molloy did say that Mr Wilmot was "floundering" in the water; but it is obvious that all he meant by that was that he was walking around in it. He denied Mr Wilmot gave the appearance of being in difficulties. 2. The Court of Appeal rightly pointed out that Mr Molloy's evidence stopped "fairly short of asserting that the plaintiff was 'an active participant in the dunking'". Mr Molloy said only that the plaintiff was "with the 10 people involved in escorting him out". The Court of Appeal also rightly pointed out that Mr Molloy did not assert a "much more extensive" range of activities than those deposed to by other witnesses. The Court of Appeal was also correct to say that the trial judge did not explain the extent to which Mr Molloy's evidence contradicted other evidence, and what that other evidence was. Nor did the trial judge find who, if anyone, was wrong, and, if so, why. Further, the trial judge did not explain what importance Mr Molloy's evidence had in relation to that called by the first defendant – that is, the evidence of Dr Trevithick. 3. Mr Wilmot's unsworn evidence, confirmed by much sworn evidence of other witnesses, was inconsistent with any possibility that the plaintiff deliberately dived into the water because he thought Mr Wilmot was at risk of drowning, and was, in turn, inconsistent with passages in Dr Trevithick's notes and testimony to the effect that the plaintiff had deliberately dived for that reason. With respect, the trial judge's failure to refer to Mr Wilmot's evidence reflected his failure to refer to any of the other evidence which corroborated it, apart from that of Mr Molloy. 4. Medical notes unsupported by testimony. For the reasons given by the Court of Appeal[7], no significance attaches to the statements in the ambulance officers' report that the plaintiff "dived" 1.5 metres. Nor, for the reasons it gave[8], does any significance attach to the nurses' notes recording the plaintiff as "jumping [crossed out and replaced by "diving"] into water off Spit Bridge": the trial judge was right to say that the Spit Bridge "is not far from the Middle Harbour Yacht Club", but the fact remains that it is a totally different place. 5. Nor, for the reasons which the Court of Appeal gave[9], does any significance attach to a note made by Dr Liston the morning after the admission of the plaintiff to hospital to the effect that the plaintiff's injury was "secondary to diving into shallow water". 6. The trial judge found that these three statements – to the ambulance officers, the nurses and Dr Liston – were made by the plaintiff. There was no testimony to this effect, since the plaintiff had no recollection of this and the makers of the statements did not give evidence. It is certainly very difficult to conclude from the notes that the plaintiff admitted to the makers of the documents that he deliberately entered the water. 7. That is also true of another note, made by Dr Sew Hoy one and a half hours after the plaintiff was admitted to the hospital ("No recollection of having fallen? dived into the water.") This note was described by the trial judge as equivocal, but, as the Court of Appeal explained, if anything it supported the plaintiff's testimony that he had no recollection of speaking to anyone at the hospital. And it supported the conclusion that whatever was said about the plaintiff's mode of entry into the water to those at the hospital was vague. 8. Aspects of Dr Trevithick's unreliability. The trial judge said that Dr Trevithick was "a most impressive witness". That expression evidently encompassed both his honesty as a witness and his reliability as a witness. The defendants' arguments blurred the distinction between his honesty as a witness and his reliability as a witness. Whatever inhibitions exist against the New South Wales Court of Appeal overturning the findings of trial judges based on honest testimony, they cannot extend to findings based on non-credit based reliability: for to reason that some common law rule precludes the latter findings from being reviewed would not be giving full effect to the statutory mandate to that Court to conduct a rehearing. And that is clearly inconsistent with the authority of this Court in Warren v Coombes[10], Fox v Percy[11] and many other cases. 9. The Court of Appeal analysed Dr Trevithick's evidence much more closely than the trial judge had. It accepted the trial judge's "finding, based on demeanour," that he was truthful. But it noted that that did not necessarily make him reliable in his understanding and recollection of the information coming to him in the hour and a half between first seeing the plaintiff and making his notes. The Court of Appeal demonstrated four elements of unreliability in his testimony. 10. First, initially Dr Trevithick suggested that the history was recorded solely as a result of interviewing the plaintiff. Later he said it was secured after speaking to four of the plaintiff's friends. He said thereafter that in addition the history was based on what the ambulance officers told him. 11. Secondly, initially he said that when, an hour and a half after first seeing the plaintiff, he made his notes about the history he took from the plaintiff, it was very unlikely that he looked at the nurses' notes. He later contended that he may have looked at the front sheet of those notes as a means of obtaining the telephone numbers of the plaintiff's relatives. It is likely that he did, because although his oral evidence, given without recourse to the notes he had made, placed the plaintiff's accident at the Middle Harbour Yacht Club, his notes placed it at "the Spit". The front sheet of the nurses' notes contain, among other items of information including telephone numbers, a note referring to "jumping [crossed out and replaced by "diving"] into water off Spit Bridge". That suggests two possibilities for the source in Dr Trevithick's notes of the reference to Spit Bridge: either the nurses' notes, or some mistaken informant on whom both he and the nurses were relying. On Dr Trevithick's evidence the only other possible sources of the information were the plaintiff's friends and the ambulance officers, and none of them could have been mistaken about the scene of the accident. That excludes the second possibility and leaves only the first. 12. Thirdly, after Dr Trevithick gave evidence that he had obtained the telephone number of the plaintiff's mother from the plaintiff, he conceded the possibility that he obtained it from the plaintiff's older brother. 13. Fourthly, in relation to the absence from his notes of the fact, noticed by Mr Moon, that after the accident the plaintiff was going into and out of consciousness at the scene of it, Dr Trevithick initially said that while he remembered it, it was not recorded because the notes were only a summary, but later said he could not remember what was said on that subject. 14. In view of these four elements, it is not entirely clear what the trial judge was referring to when he said that Dr Trevithick's "answers in cross-examination confirmed his evidence in chief." 15. The ambiguity of Dr Trevithick's evidence. Putting aside the above instances of the unreliability of Dr Trevithick's evidence, even if the trial judge was correct in concluding not only that Dr Trevithick was a truthful witness, but also that he was a reliable one, the trial judge failed to consider whether anything the plaintiff said to Dr Trevithick was in truth not an admission of deliberate entry into the water, but only of entry into it. 16. The trial judge paid no attention to the ambiguity of the evidence given by Dr Trevithick and found in various hospital and ambulance records to the effect that the plaintiff "dived" into the water. To "dive" into water is to "descend or plunge" into it, usually headfirst[12]; but the word does not by itself indicate whether the descent or plunge is deliberate or not. Nor do the reasons for judgment record whether consideration was given to whether the authors of the word "dived" other than Dr Trevithick were influenced by the unhappy experience of medical professionals that those who sustain spinal injuries in water often do so because of a deliberate plunge into it. Nor did the trial judge consider the limited significance to be attributed to the word "dived" when used, if it was used, by a man who, according to all the other relevant firsthand evidence in the case, had lost his balance, had attempted to regain it, and had then entered the water headfirst while attempting to place his arms and hands in front of him in the manner which a person intentionally diving into water might. In short, the trial judge did not consider and deal with the possibility that if the plaintiff told Dr Trevithick that he had "dived" into the water, that was not inconsistent with what the primary witnesses had observed the plaintiff doing. 17. Counsel for the second defendant contended that Dr Trevithick's notes and testimony recorded an admission of deliberate entry into the water for the following reasons. The note which Dr Trevithick made about an hour and a half after first speaking to the plaintiff was: "One friend was thrown into the water at the Spit and pretended to drown so Daniel dived 1.5 m into shallow water." (italics added) Dr Trevithick's evidence in chief about what the plaintiff told him was: "I don't think I can actually use his exact words after this length of time … he dived into the water because ... he thought his friend was at risk of drowning." (italics added) Counsel for the second defendant submitted that the words "so" and "because" indicated that the plaintiff was saying he intentionally entered the water. 18. That linguistic point has no substantive force. As the Court of Appeal said, "the suggested attempt at rescue" of Mr Wilmot recorded in Dr Trevithick's notes and repeated in his testimony was "nonsensical". That was not a description challenged by any defendant in this appeal. The only explanation offered for it was that offered by counsel for the plaintiff: that Dr Trevithick had got matters the wrong way around, since in truth Mr Wilmot had, being the first person to aid the plaintiff, in a sense tried to rescue him. It would have been highly improbable, even if the plaintiff had thought Mr Wilmot was in difficulty, and even if he had not seen him standing in the water, that the plaintiff would have deliberately dived in. He could have entered the water as speedily without diving, he had never been to the yacht club before and was unaware of the depth of the water and the nature of the surface beneath it, and an experienced sailor would have appreciated the foolhardiness of diving in such obviously dangerous circumstances. 19. There was simply no evidentiary basis whatever for the existence of any pretended or actual danger to Mr Wilmot. Indeed it was contradicted by the evidence of primary witnesses whom the trial judge also apparently found to be truthful and reliable. It was also contradicted by the evidence of the same primary witnesses that the plaintiff did not intentionally enter the water. 20. For the above reasons the Court of Appeal was right to order a new trial. 21. Lack of reasons. The Court of Appeal recorded a submission by the plaintiff that there should be a new trial because of the failure of the trial judge to give adequate reasons. It analysed the authorities and concluded that the trial judge had failed to comply with the duty to give adequate reasons. It therefore concluded that a new trial would have to be ordered unless the Court of Appeal were in a position to decide the matter for itself. 22. This conclusion was a free-standing reason for ordering a new trial; it was not merely a passing dictum. It was belatedly invoked by the plaintiff in this Court. And it is not a reason which was attacked in the defendants' grounds of appeal to this Court. Any attack on it must fail. With respect, the Court of Appeal was right to hold that the trial judge had not complied with the duty to give adequate reasons. While it is not necessary for trial judges to set out and analyse the totality of the evidence, in a case like the present, in which the trial judge rightly or wrongly identified certain inconsistencies which he saw as crucial, it was necessary to explain why they were inconsistencies and why one body of evidence was to be preferred to another. 23. That ground too supports the Court of Appeal's order for a new trial. Did the Court of Appeal have power to order a new trial on a restricted basis? 1. The third issue is whether the Court of Appeal had power to make the qualified new trial order it made. Counsel for the plaintiff devoted considerable energy to demonstrating the amplitude of its powers. However, the defendants did not deny that the Court of Appeal had the necessary power, and for that reason there is no utility in debating the issue. As Callinan J has demonstrated[13] the power was to be found in the Supreme Court Act 1970 (NSW), s 75A and the Rules of the Supreme Court of New South Wales, Pt 51 r 23. There is no need to repeat that demonstration. It is agreed in by all other members of this Court[14]. The power of the Court of Appeal to act as it did is clearly established. Did the Court of Appeal exercise its discretion to limit the new trial soundly? 1. The fourth issue was whether or not the Court of Appeal was right to limit the new trial in the way it did. This is the point upon which this Court is divided. Upon this point, we agree in the conclusion reached by Callinan J. 2. The nature of the arguments. The second defendant advanced three groups of arguments against the limitation which the Court of Appeal order placed on the second trial. The first was that there were unresolved credibility disputes. The second turned on the proposition that the Court of Appeal had pre‑empted a finding of the trial judge turning on his observation of the demeanour of witnesses whom the Court of Appeal had not had the opportunity of observing. The third was that the conduct of the new trial would be unsatisfactorily hampered by the order. 3. Unresolved credibility disputes. Counsel for the second defendant submitted in relation to the question of whether the information appearing in Dr Trevithick's notes was derived from the "four" friends who came to the hospital that three of them – Messrs Wilmot, Moon and Roberts-Thompson – "all rejected the possibility that they were the source". He also submitted that this meant "there remains a lively, unresolved dispute of credibility … that should have been left to be determined by a proper trial" (ie a second trial without conditions). 4. This submission is ill-based both factually and in logic. 5. There is, in fact, evidence that there were five friends at the hospital – Simon Hartigan (who Mr von Bibra remembers as being there), Mr von Bibra, Mr Moon, Mr Roberts-Thompson and Mr Wilmot. Of them, there was no evidence of any kind from one – Mr Hartigan. Although Mr Moon stated that he saw Mr Wilmot at the hospital, Mr Wilmot's statement did not deal with any hospital visit he made, and indeed it suggests that there was none. Hence the constant reference by the parties to "four" friends may in fact be correct. Only three of the friends testified: Mr von Bibra, Mr Roberts-Thompson and Mr Moon. Of these, only Mr von Bibra and Mr Roberts-Thompson observed the plaintiff's entry into the water. Mr von Bibra was asked no questions about what he or others said to medical staff at the hospital. It would be surprising if there really were a lively unresolved dispute of credibility about what happened at the hospital, since in Mr von Bibra's statement of 10 April 2000, in which he gave an account of the plaintiff's entry into the water which was strongly confirmatory of the plaintiff's evidence, and said that Mr Wilmot did not appear to be feigning drowning or actually drowning, he said that he and other friends had conversations with some of the medical staff. He continued: "There was no discussion by any of us to any medical staff as to what had actually happened at the Club to [the plaintiff] which caused his injury. Certainly there was no mention to the medical staff by any of us to the effect that [the plaintiff] had dived into the water to save a drowning mate." It is plain that if Mr von Bibra had been asked by the medical staff what the cause of the plaintiff's accident was, he would have been unlikely to say what Dr Trevithick's notes record. 1. Mr Roberts-Thompson and Mr Moon did not reject the possibility that they were the source of the information in Dr Trevithick's notes: they merely said that they could not remember being the source. If there were any dispute about the credibility of those answers, the time for it to be dealt with was in cross-examination of the witnesses at the first trial. In fact there has never been any suggestion either that there was a dispute about their credibility or, if there was, that it is unresolved. 2. The absence of demeanour-based findings against the plaintiff. Whether or not an order for a new trial on a restricted basis should be reserved for exceptional cases – and the enactments giving power to do this say nothing about exceptional cases – this was certainly an exceptional case. 3. The factor ultimately turning the trial judge against the plaintiff's case was the admission which the plaintiff supposedly made to Dr Trevithick. That called for an assessment of whether the admission was made, and, if so, what its reliability was; and it called for that assessment to be made against a background of a mass of evidence not turning on demeanour. 4. The assessment of whether the plaintiff made the admission, and what its significance was in the light of the other evidence, depended on taking into account the following matters. Dr Trevithick was not asked to recall the events of the night he treated the plaintiff until some weeks prior to the trial, over five and a half years later. He had not prepared a statement at any stage. He relied on the notes he prepared about an hour and a half after the plaintiff was brought to the hospital. He experienced an extremely busy evening, having to deal with five patients with catastrophic injuries. He composed his notes after speaking not only to the plaintiff but also to his friends (only two of whom observed the plaintiff's entry into the water) and the ambulance officers; and he referred to the nursing notes. He did not record, and could not remember, which pieces of information came from which source. The plaintiff was in a distressed condition, and had been moving into and out of consciousness. The supposed admission depended on a conclusion that an ambiguous statement by the plaintiff was intended to be taken in one particular way. Even if Dr Trevithick is assumed to have been both honest and reliable (subject to the qualifications in the latter respect demonstrated by the Court of Appeal), the supposed admission, when considered against all the other evidence, the honesty of which was not challenged and the reliability of which was either not challenged or scarcely questioned, cannot negate the plaintiff's case that he did not enter the water voluntarily, but did so as a result of losing his balance after a nudge from behind. 5. Either the plaintiff deliberately entered the water or he did not. If he did not, in the particular circumstances of this case as it was run, he can only have entered it by being jostled or pushed. The Court of Appeal's analysis concluded that, allowing for and accepting the trial judge's findings as to Dr Trevithick's truthfulness, the plaintiff did not deliberately enter the water. The only other possibility open is that he was jostled or pushed. 6. In the particular circumstances of this unusual case the Court of Appeal was right to substitute that conclusion for the trial judge's conclusion. 7. One of those particular circumstances is the way in which the defendants conducted it. For example, while in this Court the defendants sought to make something of a contrast between Dr Trevithick's positive recollection of what the plaintiff said to him, and the plaintiff's professed inability to remember speaking to any doctor on the night of the accident, at the trial the plaintiff was not cross‑examined to suggest that that professed inability was merely convenient self-serving and that he did in truth remember saying what it was suggested he said. The defendants did not cross-examine the plaintiff to suggest he was fabricating his story. So far as the defendants cross-examined the eyewitnesses to the accident about their evidence in chief on that subject, they only did so to suggest a want of reliability because of poor opportunities for observation. 8. These tactics flow from an understandable forensic choice. The cross‑examiners did not call evidence to contradict the eyewitnesses, and evidently lacked any material from which to cross-examine them to suggest that their evidence was false. It would certainly have been wrong to have cross‑examined them to suggest fabrication without proper supporting material. 9. In any event, undue aggression towards either a quadriplegic plaintiff or his witnesses can be counterproductive. But understandable though the tactics were, the consequence of their adoption was to narrow very greatly the opportunity for the trial judge to base his reasoning on his observations of witness demeanour. His reasoning had to turn very largely on inferences from primary evidence which, the defendants by their conduct of the case accepted, might be unreliable but was given sincerely. Similarly, since the criticism by the Court of Appeal of Dr Trevithick's evidence turned on factors going to reliability rather than credibility, it was open to that Court to reason as it did. 10. That conclusion is not undermined by the following famous words of Lord Sumner[15]: "[N]ot to have seen the witnesses puts appellate judges in a permanent position of disadvantage as against the trial judge, and, unless it can be shown that he has failed to use or has palpably misused his advantage, the higher Court ought not to take the responsibility of reversing conclusions so arrived at, merely on the result of their own comparisons and criticisms of the witnesses and of their own view of the probabilities of the case. The course of the trial and the whole substance of the judgment must be looked at, and the matter does not depend on the question whether a witness has been cross-examined to credit or has been pronounced by the judge in terms to be unworthy of it." 1. The Court of Appeal rightly thought that the trial judge in part failed to use and in part palpably misused his position. Not only were the eyewitnesses not cross-examined as to credit, they were not cross-examined as to the issue. The judge did not pronounce them unworthy of credit, but his acceptance of the truth of what Dr Trevithick wrote down and testified to about how the accident occurred could only have been arrived at if he had explained how the eyewitnesses could be ignored. This he did not do, and the cross-examinations gave him no material with which to do it. The first defendant submitted that appellate courts are not "entitled to dismiss [a trial judge's] findings where they are supported by exposed reasoning". The findings of this trial judge were not. 2. In short, the Court of Appeal, having been invited to overturn the trial judge's findings of fact, had a duty to examine the trial judge's reasoning and the evidence to see whether it should. In that examination it employed the following unimpeachable methods. First, it concluded that the trial judge's reasoning was unsatisfactory – and the defendants devoted very little effort to defending it either in the Court of Appeal or in this Court. Secondly, the Court of Appeal, as it was entitled to: (a) looked at all the probabilities; (b) took into account the trial judge's findings so far as they were favourable to the plaintiff and not open to rational criticism; (c) refrained from reaching conclusions adverse to the plaintiff based on propositions in medical records which were made by persons who were not called as witnesses, which derived from unknown persons, and which were contradicted by all the eyewitness testimony in the case; (d) while giving full weight to Dr Trevithick's notes and testimony, and to the trial judge's conclusions from them so far as they were demeanour-based: (i) departed from his evidence where, although honest, it was not reliable; (ii) alternatively, accepted Dr Trevithick as having accurately recorded and recollected what he was told, but rejected the reliability of what he was told. 1. None of these steps involved the Court of Appeal reversing demeanour‑based findings of the trial judge. 2. Obstructing the second trial. The third group of arguments against restricting the retrial advanced by the defendants turned on various disadvantages and inconveniences which, it was said, the restriction would create. Counsel for the plaintiff, on the other hand, submitted that, in view of the unrestricted and complete way the original trial had been conducted, it was incumbent on the defendants to point in a realistic way to the parts of the evidence called at the trial which might be affected by the Court of Appeal's order. This they did not do. 3. The second defendant argued that to conduct the second trial on the basis that the plaintiff was "jostled or pushed" was unsatisfactory: "Jostled or pushed by whom? Someone whose propensity to be doing just that should have been within [the plaintiff's] cognizance for the purposes of contributory negligence or somebody that he did not need to be taking into account? Is this going to be an inadvertence or misjudgment answer to contributory negligence by the plaintiff or not? Jostled or pushed is quite a spectrum." 1. Similarly, the second defendant argued that the Court of Appeal's order amounted to a "partial determination" which "prevents or substantially hinders" a second trial on "questions of fact intimately associated with how the plaintiff came to be standing on the jetty, how he came to enter the water and the circumstances surrounding those matters." 2. It is true that the Court of Appeal's order prevents inquiry into the limited issue of how the plaintiff came to enter the water. These submissions did not explain why the Court of Appeal's order made it impossible for the defendants to investigate the scene on the wharf in order to advance the balance of their contributory negligence case, whether by inquiring into how the plaintiff came to be standing on the jetty, into who jostled or pushed him, or into any other circumstances surrounding that matter. 3. The second defendant also submitted that it was impossible to separate out the issue which the trial judge had decided one way and the Court of Appeal another, because it turned on questions of credibility, and those questions might affect other areas of the case as well. An illustration was given: the credibility of the plaintiff on damages could be affected by a conclusion as to his credibility in describing his entry into the water. Another illustration was given: the credibility of the eyewitnesses on the plaintiff's contributory negligence could be affected by a conclusion as to his credibility in describing his entry into the water. So far as damages are concerned, at the start of the trial, the issues on damages were identified as turning on the plaintiff's life expectancy, the amount to be awarded as general damages, the future accommodation to be provided to the plaintiff, and the appropriate level of computer facilities for him. 4. With all respect, none of these matters seemed likely to raise a credibility question. Certainly this Court was taken to no part of the plaintiff's cross‑examination on damages which revealed any challenge to his credibility – or any other challenge. The first and third defendants did not cross-examine on damages. The second defendant's cross-examination on damages consisted of a calm process of eliciting various uncontroversial facts about the plaintiff's then position and future prospects. No doubt these facts were seen as being useful in the second defendant's case, but nothing in the cross-examination suggested any challenge to credibility. There are no signs of acrimony or disagreement between counsel and witness. So far as contributory negligence is concerned, it was said that the credibility of the eyewitnesses as to the plaintiff's entry into the water "is going to be critical". Perhaps the defendants would like it to be so at the second trial, but they did not make it so at the first trial. They should not be afforded a second chance that reopens a factual question which, on proper analysis of the evidence in the first trial, conducted by the Court of Appeal as its duty required and its powers permitted, was resolved in favour of the version presented by the plaintiff's case. 5. It was said that the Court of Appeal's order that the second trial be conducted on the basis that the plaintiff, through being jostled or pushed, lost his balance would raise difficulties in relation to dealing with the contradiction between Mr Molloy and other witnesses on whether the plaintiff had helped to throw Mr Wilmot into the water. That was a particular of contributory negligence, but the submission did not say how that contradiction bore on the basis for the new trial, which assigned a particular mode by which the plaintiff entered the water. The defence tactics in cross-examination at the first trial, too, did not support any inference that the contradiction was related to the plaintiff's mode of entry into the water. 6. There were signs that these submissions had an armchair character. They gestured at tactics which the defendants might have used at the first trial, but did not. They suggested that the defendants would wish to conduct a second trial differently if the question of how the plaintiff entered the water was still in issue. This was a suggestion the more easily made in view of the fact that none of the counsel who led for the defendants in this Court led at the trial; in appellate litigation the absent are often wrong. Thus the first defendant contended that, at the second trial, it might be a question whether the eyewitness evidence had been contaminated by the fact that the statements on which it may have been based were prepared a year later, after the witnesses had congregated at the Middle Harbour Yacht Club for counselling, and whether they were not impartial witnesses. The first defendant also referred to the fact that at the first trial the quantity of alcohol drunk by the plaintiff had "not [been] very much explored with him in cross-examination". 7. But the first trial was not a rehearsal, warm-up or dummy run. It must have been conducted against the possibility of all available factual findings being made. To suggest that the defendants will be hampered at a second trial in relation to the employment of tactics they did not employ in the first trial is unconvincing. 8. There is another respect in which the defendants are attempting to alter the consequences of their conduct of the first trial. Accepting that they did not instigate or agree to the trial judge deciding the case on the "prime factual issue" and not all other issues, the terms of his reasons for judgment suggest that they did point out that if the plaintiff's case failed on that point, the whole case would fail. It does not lie well in their mouths now to protest that the Court of Appeal's order that a new trial be conducted on the basis that the issue had been decided the other way makes a second trial unworkable. Their perceptions of that issue as being the "prime factual issue" did not make their conduct of the first trial unworkable. What they effectively tendered as an issue for separate resolution to their own advantage, they should now be required to accept as having been properly concluded against them, with the forensic consequence that follows for the second trial. Otherwise, the value and outcome of the thorough review of, and conclusion on, the facts, recorded in the reasons of the Court of Appeal, are thrown away. 9. The third defendant contended that the Court of Appeal's order was unsatisfactory in foreclosing questions of causation and contributory negligence; made it difficult for the parties to prepare for and conduct the second trial; and would make it difficult to deal with any objection by the plaintiff to evidence of the circumstances surrounding the accident. These are speculative contentions. All otherwise admissible evidence will be receivable unless it goes to the one issue foreclosed. The judge conducting the second trial may have to rule on objections, but the difficulties were not shown to be insuperable. As counsel for the plaintiff said in this Court, it is not necessary now to give advisory opinions on what questions counsel for the defendants could or could not ask at the second trial, nor is it necessary to consider whether the judge hearing the second trial might make some mistake in ruling on objections to those questions. The third defendant said that the difficulties were probably incapable of articulation until after the second trial had been conducted. Certainly they were not articulated convincingly in argument in this Court. 10. When this category of the defendants' arguments is examined, it can be seen that it was not unfair of counsel for the plaintiff to describe them as no more than "a great deal of huffing and puffing". 11. Conclusion. In these circumstances the Court of Appeal was correct to conclude that there should be no re-litigation – perhaps employing different methods – of an issue which the defendants had already had a full opportunity of litigating. In Pateman v Higgin[16] Kitto J, discussing the Common Law Procedure Act 1899 (NSW), s 160, a precursor to Pt 51 r 23, said: "it remains … a sound general proposition from which to start in the consideration of each particular case according to its own circumstances that if there is to be a new trial it ought to be of the case as a whole unless the Court thinks that 'they shall do more injustice by setting the matter at large again'." 1. To have ordered a new trial on all issues would have produced more injustice than the order the Court of Appeal made. It would have called for a re‑litigation of an issue fully litigated at the first trial. It was an issue on which the correct outcome, given the conduct of the first trial by the defendants, was plainly that at which the Court of Appeal arrived. The trial judge's expressed conclusion in the first trial prevented the attainment of finality on all issues. It was right of the Court of Appeal to seek to achieve finality on the one issue he did deal with, however unsatisfactorily. 2. Upholding the special order. We have taken the trouble, at greater length than is usual in this Court, to explain the facts and to recount the evidence given at the trial. We have done so for several reasons. We accept that, when a trial has miscarried, the usual order for the appellate court to make is for a new trial generally, upon all issues. That is what a majority of this Court favours in this appeal. It is the course that will now follow. 3. However, it is important to demonstrate that this was not a usual case. Moreover, the Court of Appeal is not open to criticism for making the order effecting the retrial in the terms that it did. Only by appreciating the detailed evidence, and the error of the first trial, can the order of the Court of Appeal be understood. It is now accepted that the order made was within the powers of the Court of Appeal. The terms of that order involved a discretionary judgment that an appellate court, like this, should be slow to disturb. It is only when the deplorable result that will now follow is fully appreciated that the justification for the Court of Appeal's order is made clear. At the first trial, the defendants tendered an issue on the "primary factual question". At trial they succeeded, but on appeal they failed on that issue. They were willing to take its fruits. Properly, the Court of Appeal, reversing the trial judge, required them to wear its burdens. In our view this Court errs in disturbing the Court of Appeal's orders in this respect, which we would affirm. Did the Court of Appeal err in ordering the defendants to pay the costs of the first trial? 1. The fifth issue arises from the Court of Appeal's order that the defendants pay the costs of the first trial in any event – that is, even if they succeed against the plaintiff in a second trial on some point not decided in the first trial. 2. The Court of Appeal was asked to change that order, but declined, without giving reasons, to do so. The absence of reasons makes it difficult to identify the justification for the order. Counsel for the plaintiff sought to justify it by saying that the defendants, having not taken the point on which they succeeded before the cross-examination of the plaintiff, and having urged on the trial judge the proposition that if they succeeded on that point that was the end of the matter, were responsible for the wasting of all the costs incurred in the first trial. 3. The materials before this Court do not demonstrate that the responsibility of the defendants for the course which the trial judge took was so great that they should be left with a liability to pay the costs of the first trial in any event. Indeed, counsel for the plaintiff, in a note circulated after oral argument, made it plain that the idea of deciding one issue to the exclusion of all others was not urged by any party. Accordingly, the costs order should be set aside, and the costs of the first trial should abide the decision of the judge who hears the second trial. 4. However, it is undesirable to make any order that the defendants have their costs on this issue in this Court, and indeed they did not in terms seek it. The matter is de minimis because it occupied so small a part of the argument in this Court. There is a real risk that the expense of working out what fraction of the overall costs went to this issue would exceed the value and justification of attempting to do so. Orders 1. We agree with the orders proposed by Callinan J. 2. HAYNE J. I agree with Gleeson CJ. 3. Reference was made in argument to the "sufficiency" of the primary judge's reasons. When it is said that a judge did not give "sufficient" reasons for a decision there may be some doubt about what principles are engaged. Reference may be being made to the duty of a judicial officer "to make, or cause to be made, a note of everything necessary to enable the case to be laid properly and sufficiently before the appellate Court if there should be an appeal [including] not only the evidence, and the decision arrived at, but also the reasons for arriving at the decision"[17]. To fail to make or cause to be made such a note may invoke principles of procedural fairness and constitute a failure to exercise the relevant jurisdiction[18]. 4. In the present case, however, reference to the "sufficiency" of the primary judge's reasons is not to be understood as seeking to invoke only those principles. Rather, because the primary judge was bound to state the reasons for arriving at the decision reached, the reasons actually stated are to be understood as recording the steps that were in fact taken in arriving at that result. Understanding the reasons given at first instance in that way, the error identified in this case is revealed as an error in the process of fact finding. In particular, it is revealed as a failure to examine all of the material relevant to the particular issue. 5. The primary judge's reasons stated his conclusion that the evidence of Dr Trevithick was to be accepted and preferred to that of other evidence but disclosed no reasoning supporting that conclusion. No analysis was made of the competing evidence and no explanation proffered for rejecting it. The most that might be inferred from what was said was that some special significance was attached to the existence of the written record upon which Dr Trevithick founded his oral evidence. But what significance was to be attached to the existence of that record might well be thought to have turned critically upon the source or sources of the information recorded in it. That was not a matter examined in the reasons. The absence of explanation for, and reasoning in support of, the conclusion expressed in the primary judge's reasons reveals that the process of fact finding miscarried. It miscarried because, so far as the reasons reveal, no examination was made of why Dr Trevithick's evidence was to be preferred to that of other witnesses. 6. When it is understood that the primary judge made this error it becomes apparent that there must be a new trial of this issue as well as all other issues in the matter. The task of the Court of Appeal is prescribed by statute[19] and the applicable rules of court[20]. The nature of that task was recently considered in Fox v Percy[21] and its importance emphasised. The Court of Appeal is bound to rehear the case and the Court, in doing that, may draw inferences and make findings of fact. But it was as much an error for the Court of Appeal to make the finding of fact which it sought to specify in its order as it was for the primary judge to make the erroneous finding he made. 7. As Fox v Percy recognises[22], the Court of Appeal "must, of necessity, observe the 'natural limitations' that exist in the case of any appellate court proceeding wholly or substantially on the record". The defect in the primary judge's fact finding lay in the failure to evaluate all of the evidence bearing upon the relevant issue of fact. The Court of Appeal could not substitute its finding when that too was based on only part of the material which ought properly to have been considered by the primary judge. Yet that is what the Court of Appeal did. 8. The Court of Appeal made a finding of fact that was necessarily founded upon only part of the material that had been available to the primary judge: the transcript of what the witnesses had said and the documentary evidence that was received. The witnesses called to give evidence at the trial had given divergent accounts of what had happened. The Court of Appeal could make no evaluation of the credibility of those witnesses[23]. Without itself seeing and hearing the witnesses, and without any relevant finding by the primary judge about the probable accuracy and reliability of the testimony given by the witnesses other than Dr Trevithick, the Court of Appeal could not decide whether the manner in which any of those other witnesses gave their evidence bore upon the finding to be made. It may be that the manner in which those witnesses gave their evidence would be of no assistance in deciding what finding should be made. For present purposes, however, the critical thing is that the Court of Appeal had no basis upon which it could treat it as irrelevant. 9. It may readily be accepted that retrial is a remedy that inflicts great hardship on parties and witnesses. But the Court of Appeal rightly concluded that there had to be a new trial in the present matter. The Court of Appeal erred, however, in deciding that the new trial should be confined in the manner described. The orders proposed by Gleeson CJ should be made. 10. CALLINAN J. The first respondent suffered serious injuries when he was either pushed, jostled or jumped from a jetty under the control of, or constructed or approved by, one or more of the appellant, the Middle Harbour Yacht Club (the second respondent) and the Mosman Municipal Council (the third respondent). He sued the appellant and the second and third respondents in the Supreme Court of New South Wales. Each of the defendants at the trial cross-claimed against the others and alleged contributory negligence on the part of the first respondent. The action was dismissed by the trial judge (Newman AJ). Because of that it was unnecessary for his Honour to determine the issues between the appellant and the other respondents. The first respondent appealed to the Court of Appeal. That Court ordered a retrial upon a restricted basis. Who, if any, of the appellant and the second and third respondents should be liable to the first respondent remains undetermined. 11. The principal question in the appeal to this Court is whether the New South Wales Court of Appeal erred in the exercise of its powers under s 75A of the Supreme Court Act 1970 (NSW) ("the Act") by ordering, that although there should be a new trial of the action, it be conducted on the basis that it was established in favour of the first respondent that, through being jostled or pushed, he lost his balance and fell from the jetty into the water. Facts 1. The first respondent on 29 March 1997, was 20 years old and a successful recreational sailor. He lived in Queensland but was in Sydney on that date to compete in sailing races at Rose Bay. On the evening of that day he was present at a ball conducted at the Middle Harbour Yacht Club. The clubhouse is constructed on the foreshore of the Harbour. On each side, and directly in front of it is a jetty. 2. A white "toe-board" approximately 200 mm in height, but no railing, was constructed on the seaward side of the jetty. It presented no barrier to entry into the water. 3. The evidence as to the number of people in attendance at the ball was imprecise. It is clear however that it was not a small function. Alcoholic drinks were served, but not consumed in excessive quantities. Although the first respondent had partaken of them his judgment was not impaired: it was his intention to compete in an important sailing race the next day. 4. Mr Nathan Wilmot who had won the National titles in Hobart during the previous year for the class of boats in which the first respondent sailed was present at the ball. Sailors, or at least this group of them, had the practice at the conclusion of a regatta of "dunking" the winner in the waters in which the race was conducted. Mr Wilmot had not been dunked in Hobart the preceding year. Some of those present at the ball resolved to dunk him during the ball. They took hold of him in the club, carried him to the jetty and dunked him in shallow water off the beach. 5. Some further description of the jetty is required. Two other jetties ("the mooring jetties") jut out at right angles from the club. Boats are moored from time to time in the water to the sides of the mooring jetties and close to the jetty from which the first respondent entered the water. The depth varied with the tide. The tide was in on the evening of the ball. The sand immediately below the jetty was covered by shallow water at the time. The distance between its surface and the jetty was about one and a half metres. Its depth in the vicinity of the jetty was only about one fifth of a metre. The jetty was lit but the lighting did not illuminate the seafloor. There was no impediment to access from the clubhouse to the jetty. Access to the mooring jetties could only be gained by passing through lockable gates. The trial 1. A suggestion that the first respondent participated in the dunking of Mr Wilmot was not made out. He did however say that he watched it from the jetty on the fringe of a group of people. His feet were then resting, he said, against the toe-board on its edge. It was from there that he entered the water. When he did his head struck the sandy bottom. He was rendered, in consequence, a quadriplegic. 2. The appellant was the owner of the land on which the clubhouse and jetties were built. The second respondent was the occupier of the premises, and the third respondent was the local government authority in whose municipality the clubhouse and the jetties were situated. 3. The first respondent's case included that while he was standing on the jetty he was jostled or bumped into the shallow water. 4. The second and third respondents, neither of which had direct knowledge of the events, made generalised allegations of contributory negligence on the part of the first respondent. However, the second respondent's allegations were more specific: "(a) Failing to keep a proper lookout; (b) Failing to take due care in all the circumstances; (c) Failing to take adequate precaution for his own safety; (d) Failing to observe the environment in which he was walking; (e) Failing to take care whilst on the wharf/jetty; (f) Standing too close to the edge of the wharf/jetty; (g) Failing to observe that he was in a position of peril; (h) Exposing himself to a risk that he could be forced off the edge of the wharf/jetty; (i) Failing to move back from the edge of the wharf/jetty; (j) Rushing towards the edge of the wharf/jetty without due care; (k) Failing to have regard to the depth of the water beneath the wharf/jetty; (l) Participating in activities with other people which exposed himself to a risk of injury; (m) Engaging or participating in activities which led Nathan Wilmot to be thrown into the water; (n) Failing to prevent Nathan Wilmot from falling into the water". There was no allegation by any of the appellant and the second and third respondents in their pleadings that the first respondent had deliberately dived into the water. Notwithstanding this, whether he did became a central issue at trial. 1. The trial judge having found that the first respondent had deliberately dived into the water dismissed the action. 2. In evidence-in-chief at the trial, the first respondent said this in answer to these questions: "Q. ... what [did] you experience? A. Well, it was a nudge from behind as I was on the edge which put me off balance and it – and I couldn't recover because my feet were butted up against this raised timber edging, I couldn't put my feet out to stop my momentum so I just fell forward. Q. Do you recall anything about the way in which you fell? A. With regard – sorry? Q. You have told his Honour that there was this nudge from behind? A. Yes. Q. You lost your balance, you went forward? A. Yep. Q. Do you recall what inclination you adopted as you went down? ... A. ... I tried to counteract it by putting – by swinging my arms back. Q. Did that do any good? A. No, no, I just toppled there. Q. What did you strike? A. The bottom, the sandy bottom it was. Q. With what? A. My head. Q. What did you notice about yourself? A. I didn't notice anything at the time. I did black out. The next thing I remember, I – it was quite a bit of commotion and water and – and looked up and there was Nathan Wilmot looking down at me and I noticed I couldn't control – like I couldn't stand up or couldn't control what was happening." 1. Later the first respondent gave the following evidence about being pulled out of the water by Mr Wilmot: "A. ... once I was on the beach, they just kept me there, just comforting me and I said I couldn't move or anything and just – at the edge – at the edge of the water and, yeah, I just waited. They said they called an ambulance and we just waited for it. Q. Did you realise at that stage what was the problem with you? A. I just wanted to get to the hospital. I knew I just couldn't – I couldn't move, and, yep. It was serious." In cross-examination, the first respondent made these responses: "Q. … do you have any recollection of Nathan lying in the water as if he was pretending to drown? A. No. Q. Have you ever told anyone about that? A. About Nathan? Q. Have you ever told anyone that Nathan was pretending to drown? A. No. Q. You never have? A. No." The basis for this cross-examination was a statement attributed to the first respondent by the doctor who treated him at Royal North Shore Hospital on his arrival there soon after he was injured. The cross-examination continued: "Q. Can I suggest to you that you told at least one and perhaps two of the ambulance officers that you dived approximately 1.5 metres, landing face first in shallow water. Did you tell anyone that? A. I don't remember. Q. Did you tell any of the ambulance officers that you'd been drinking rum and Coke and beer? A. I didn't drink any beer. Q. So it may be you told the ambulance officers you'd been drinking rum and Coke? A. I don't remember saying that, no. Q. Do you have a recollection now of being taken or actually being admitted to Royal North Shore Hospital? A. When I first got there? Q. Yes? A. I don't remember. It's very hazy. Q. Do you remember whether anyone from the yachtsman's ball either went to hospital with you or was at the hospital after you were admitted? A. Do remember Ben Moon being in the ambulance with me." 1. In re-examination the first respondent persisted in his denial that his entry into the water was voluntary: "Q. Did you have any intention of entering that water that night? A. None at all. Q. Did you know how high the deck was from the water? A. I had not been to the yacht club before so I wasn't aware of the heights of anything. Q. Did you know how deep the water was above the sandy beach? A. Yeah, I had no idea of the surroundings. Q. Did you know what the nature of the bottom was? A. No. I'd never been to the yacht club before. Q. Did you know what the nature of the bottom was off that point of the timber deck? A. No. Q. Did you know if it was sand? A. No. Q. Or rock? A. I didn't know of anything. Q. Did you see Mr Wilmot at any time face down in the water? A. No. Q. What was the deepest into the water that you saw him at any stage? A. At ankle deep. Q. In what position was he when you saw him at that depth? A. He was standing in the water. Q. Did he convey to you the appearance of any difficulty in the water? A. No. Q. Were you affected by liquor at the time you entered the water? A. No. Q. Do you recall any conversation in the ambulance? A. No. I was – I'd just broken my neck. I had no sense." (Emphasis added) 1. I turn to the evidence given by the witnesses called in support of the first respondent's case. Mr James gave the following account in evidence-in-chief: "Q. What became of Mr Fitzgibbon? A. I saw Mr Fitzgibbon standing there looking in towards Nathan, and then I was talking back to my friends, turned around and I saw him overbalanced. … Q. How would you describe the method or plane in which he entered? A. I would say he entered, he was falling over headfirst, and he entered the water in a forward motion. … Q. The picture that you saw was one of him attempting to regain his balance? A. Yes, sir. Q. Did he achieve that outcome? A. No, sir." 1. Another witness, Ms Roberts-Thompson, said that she saw the first respondent falling into the water: "Q. Where was he in relation to the edge? A. I couldn't tell you in relation to the edge. I could tell you – I only saw him when he was falling into the water. Q. Well, what did you see then? A. I saw him falling into the water and as he got closer, right to the end – as he was falling I saw his arms go forward but he landed in the water before any of that, before he could finish anything. Q. Before his arms moved in the way that you have described how would you describe their position with regard to his body generally before he put them out as you've described it? A. Just, it would be like a falling forward motion. Q. Did he appear to you to be diving? A. No. Not at all. Q. How did he enter the water? A. Well, as I said before, he was trying to get his hands out but he hit the water. It was basically headfirst." 1. Her brother, Mr Roberts-Thompson, was also a witness. He said this: "… Daniel I identified was falling, pushed, but certainly was making his way into the water and it certainly didn't look as though it was something he was meaning to do on purpose … … … It wasn't that he was purposefully diving … he was … caught off balance, or fell, or basically that it was not his intention to be heading towards the water." 1. Evidence was given by a friend of the first respondent, Mr Molloy. He said that he saw the first respondent overbalancing with his arms raised beside him and falling. While his evidence in this respect was consistent with the first respondent's and other witnesses', contrary to the first respondent's assertion, Mr Molloy said that he had a clear recollection that the first respondent was personally involved in the "dunking" of Mr Wilmot. 2. The trial judge made no reference in his reasons to the evidence of another witness, Mr Treharne, which the Court of Appeal thought to be of "considerable importance". In chief Mr Treharne said that he was standing "shoulder to shoulder" with the first respondent on the jetty at the time of the Wilmot dunking. He did not see Mr Wilmot enter the water but saw him standing in it about "knee deep". Later Mr Treharne gave this evidence: "Q. Whilst you were standing there with him in that position, did you notice something happen to him? A. Yes, I did. Q. What did you observe? A. I noticed the top half of Dan's body propel forward. Then I noticed his feet get tangled up on a thin white painted piece of timber around the edge of the wharf, and then with his arms waving to try and stop himself, and then proceeded forward towards the water, and then had his arms out in front of him before he hit the water. Q. How did he strike the water? A. Head first. ... Q. Other than you being shoulder to shoulder with Mr Fitzgibbon at the point that you have indicated, and at the time that you have indicated, were you aware of the presence of other people around you? A. Yes. Q. Where were those other persons located apropos Mr Fitzgibbon? A. Congregated out at the end of the wharf and obviously behind us, because we were out on the edge of the wharf. Q. Did you see anyone push him? A. No. Q. Did you see a hand applied to him? A. No. Q. Or a shoulder, or any other propelling force? A. No. Q. Did he dive in? A. No." 1. Mr Treharne also said that the point at which the first respondent entered the water was very near to where Mr Wilmot was. The water there would therefore have been about knee deep. 2. In cross-examination Mr Treharne was not asked about the entanglement of the first respondent's feet in the toe-board. He said that as soon as the first respondent's upper body moved forward, his immediate reaction was to try to regain his balance. No attack was made by the cross-examiner upon Mr Treharne's emphatic denial that the first respondent had dived into the water. 3. Mr Moon, another friend of the first respondent, gave evidence that he had participated in the "dunking" of Mr Wilmot. He saw Mr Wilmot standing in the water. Next he heard a clamor behind him, turned around and noticed somebody else in the water. He saw Mr Wilmot wade out to that person, the first respondent, and pull him into the shallows. He heard the first respondent say: "Who pushed me in?" 4. The last of the witnesses on liability called by the first respondent was Mr von Bibra. He too was a competitor in the sailing competition. His evidence was that he saw a group of people escorting Mr Wilmot on to the jetty. He said: "I noticed [the first respondent] – [he] caught my attention when he lost balance on the end of the boardwalk ... He was right on the edge of the boardwalk so his feet would have been next to [the toe-board]". He added that the first respondent had gone forward and put his arms out futilely to try to stop himself, and that he could not regain his balance and fell into the water. He entered the water with his arms stretched out but below his shoulders. He saw Nathan Wilmot move to help him. In doing so he had to move a little bit further out, a distance of a metre and a half to two metres. While they waited for the ambulance he noted that the first respondent "was going in and out of consciousness", despite that he and the others were trying to keep him awake. 1. Mr von Bibra was cross-examined on a statement that was tendered and which he had given to the lawyers for the club. He had relevantly said this in it: "18. Nathan was still being half carried half walking to a point on the wharf where the water was on the sand ... I moved along the wharf to the point ... standing near a post. 19. It appeared to me Nathan was half pushed half thrown into the water from about waist high. From memory there were two at Nathan's feet and two at his shoulders. It was like one gentle movement into the water. 20. I can't remember seeing Nathan standing in the water, but it never run [sic] 'through my head' Nathan was drowning either authentically or in 'horse play'. At this stage Nathan was the only one in the water. 21. At this stage I was standing on the wharf at the point ... I saw Dan Fitzgibbon standing on the wharf at the point ... Dan was surrounded by about 8 to 10 people. I noticed that Dan's feet were hard up against the white painted timber on the edge of the wharf. 22. I was looking in Dan's direction and I saw him fall forward towards the water as if he had for some reason lost his balance. At this time I could see that there were people either side and behind him. 23. As he fell forward I could see his arms out-stretched trying to regain his balance. But at the 'point of no return' Dan put his arms in front of him like a person diving and then he hit the water." 1. The witness in the case whose evidence the trial judge came to regard as decisive was Dr Trevithick, the emergency registrar at the Royal North Shore Hospital on duty on the night of the first respondent's accident. 2. He said that he well recalled many of the events of the night. It was Easter Sunday. It was memorable for the need to attend to five major trauma cases. It was the busiest night of any that he had encountered until then in any emergency department in which he had worked. Dr Trevithick was the doctor "specifically responsible" for the first respondent's care in the emergency department that night. 3. The first respondent was, the doctor said, capable of conversing upon arrival at the hospital. He smelled of alcohol but did not appear to be affected to such a degree as to cause concern in his management. The conversation that the first respondent and the doctor had was described by him in this way: "[The first respondent] said to me that he had been on a wharf at Middle Harbour Yacht Club and when one of his friends was pushed into the water during – I don't think I can actually use his exact words after this length of time, but some skylarking perhaps, or high jinks – it's a bit hard to describe what I am trying to say – some episodes of hilarity on the wharf and enjoyment following winning a sailing regatta – when one of his friends was pushed into the water, he dived into the water because he was – he thought his friend was at risk of drowning and he dived in to the water from a height of about 1.5 metres, hitting his face on the bottom of the – on the ground – when I say – on the bottom of the water because the water depth was quite shallow at the time and he was – then felt completely numb in his limbs and had to be rescued by onlookers." 1. The doctor's contemporaneous note was more brief but was to a similar effect: "Visiting Sydney with friends for sailing regatta. Had several alcoholic drinks with friends. One friend was thrown into the water at the Spit and pretended to drown so Daniel dived 1.5m into shallow water, striking his mouth on the bottom – he was paralysed immediately, rescued by onlookers – Ambulance attended and placed cervical collar." 1. There was other evidence capable, albeit slightly, of supporting the doctor's evidence. One such piece of evidence was a note made by a nurse at the hospital (who was not called to give evidence). It read: "Patient BIBA [brought-in-by-ambulance] after diving [the word 'jumping' was crossed out] into water off Spit Bridge 1.5 metres head first into H20 ... Patient alert & orientated." 1. Three other documents admitted into evidence were also said to lend support to Dr Trevithick's evidence. The first was a note made by a Dr Sew Hoy who examined the first respondent at 12:45 am on 30 March 1997. It read: "no recollection of having fallen? dived into the water". Another was a note made by a Dr Liston who also saw the first respondent in the intensive care ward after his accident. It relevantly said: "20-year-old male; previously well; acute spinal injury secondary to diving into shallow water under the influence of ETOH". The final document was a "patient report" prepared by the New South Wales Ambulance Service which read: "? Spinal injury. 20-year-old male dived approximately 1.5 metres landing face first in shallow water". Neither Dr Sew Hoy nor Dr Liston was called to give evidence. Of all this evidence for the defence the trial judge said the following[24]: "None of the arguments advanced on behalf of the [first respondent] in any way reduces the view I have formed as to the credibility and reliability of Dr Trevithick's evidence. I am of the view that Dr Trevithick was not only a truthful witness but also a reliable one. I accept fully that he had a clear recollection of what he was told that evening by the [first respondent]. As I have already indicated his evidence does not stand alone, it is supported by notations made by others who had the care of the [first respondent] on that evening. In my view it overcomes the probative value of the evidence given by the [first respondent] and those called in his case. I should add that even if those other notations did not exist, my view as to the probative value of Dr Trevithick's evidence would be unchanged." Some of the other findings of the trial judge should be noted[25]: "Molloy's evidence is clearly contradictory of the [first respondent's] evidence relating to his part in the manner in which Wilmot entered the water. Molloy has the [first respondent] as an active participant in the dunking of Wilmot whereas, of course, the [first respondent] maintains his part was no more than a spectator. Not only that, Molloy deposes that Wilmot's actions in the water involved him carrying out a range of activities which were much more extensive than him merely standing up as all other witnesses called by the [first respondent] deposed to. The contradictory nature of Molloy's evidence is of some importance when the evidence called by the [appellant] in the matter is taken into account. The evidence called by the [first respondent] as to the events surrounding him entering the water is of such nature that in the absence of any evidence called on behalf of the [appellant] I would have come to the conclusion that the [first respondent] had established his case that he had either tripped on the raised board at the eastern edge of the jetty or was pushed or jostled by those surrounding him or indeed a combination of both factors and thus he would be entitled to have the matter determined on that factual finding. While it is true that certain of the evidence given by witnesses called on his behalf when viewed alongside his evidence has contradictory elements in it, those contradictory elements would not be sufficient in my view to displace my base view that the [first respondent] had established his case on a balance of probabilities. However, the matter does not end there. Why it does not is because of oral evidence given by a Dr Trevithick, the emergency registrar at Royal North Shore Hospital on the evening in question. Furthermore, that oral evidence was supported not only by Dr Trevithick's notes on the evening but also by notes made by other personnel at Royal North Shore Hospital and by ambulance officers of statements made to them by the [first respondent] when he was being transported to Royal North Shore Hospital and after his arrival at that institution. ... It is true that when the [first respondent] was seen at 12.45am on the morning of 30 March 1997 by a Dr Sew Hoy at North Shore the note he took of the history given to him by the [first respondent] was equivocal. That note was 'no recollection of having fallen? dived into the water'. However, when he was seen in intensive care later on that morning a Dr Liston recorded the following history: '20-year-old male; previously well; acute spinal injury secondary to diving into shallow water under the influence of ETOH.' ... Putting aside any considerations of onus of proof, looking at all the evidence dispassionately I conclude on a balance of probabilities that the [first respondent] entered the water because he dived from the jetty. In terms of onus of proof this in turn means that I am not satisfied that the [first respondent] has established his prime factual case on a balance of probabilities." Not only did the trial judge fail to make reference to Mr Treharne's evidence but he also omitted to deal with the significant piece of evidence given by Mr Moon that the first respondent had asked, immediately after he was rescued: "Who pushed me in?". It is also relevant that the second respondent called no eye-witnesses or persons present at the ball, and that the trial judge failed to deal with a submission by the first respondent about that based on Jones v Dunkel[26]. The Court of Appeal 1. The first respondent appealed to the Court of Appeal (Meagher and Santow JJA and Foster AJA). He argued there that the trial judge attributed too much weight to the evidence of Dr Trevithick and gave insufficient weight to the substantial body of evidence led on his behalf. In that regard particular attention was drawn to the fact that the trial judge failed to advert to, or consider in any way, the evidence of Mr Treharne. The argument in this respect was that the reasons of the trial judge were appealably deficient. 2. Their Honours conducted their own thorough review of the evidence. It was the opinion of Foster AJA, with whom Meagher JA and Santow JA agreed, that the trial judge failed to provide adequate reasons for the rejection of the large body of evidence that the first respondent did not dive into, or enter the water voluntarily and that it was inevitable that the first respondent would not only be "disappointed" but "disturbed" as well by this omission[27]. 3. Dealing first with the notes from the hospital nurse and the ambulance officer, Foster AJA said this[28]: "[Newman AJ], in his reasons for judgment, has made a finding, without discussion, that both these descriptions [of the first respondent diving] were statements made by the [first respondent] to the relevant ambulance and hospital personnel. With respect, I am unable to agree. In my opinion, the evidence falls short of establishing that these entries recorded admissions made by the [first respondent] to the effect that he deliberately dived into the water." Turning then to the notes of Drs Liston and Sew Hoy his Honour said[29]: "I am of the same view in relation to the notes made after the [first respondent] passed from the care of Dr Trevithick. Dr Liston was not called, nor was Dr Sew Hoy. Consequently, there is no direct evidence that the material recorded in their notes was in fact supplied by the [first respondent] rather than from the hospital documents which would have accompanied him. In the absence of any expert evidence on the topic, the question whether the [first respondent] dived or simply fell head first into the shallow water, striking his head on the bottom, would appear to be of no medical significance. The fact that Dr Sew Hoy, the orthopaedic registrar, who saw the [first respondent] approximately one and a half hours after his admission to the hospital, recorded that the [first respondent] had 'no recollection of having fallen? dived into the water' cannot, in my view, be treated as merely 'equivocal'. It is consistent with the [first respondent's] sworn testimony that he had no recollection of speaking to any doctors at the hospital. It also suggests that, at this very early stage, there was at least uncertainty as to whether he had dived or fallen. The note, clearly enough, indicates that, in a short space of time after being seen by Dr Trevithink [sic], the [first respondent] was unable to provide information as to the happening of the accident." 1. Foster AJA next turned his mind to the evidence of Dr Trevithick which the trial judge had held to be conclusive. One passage in particular in his Honour's analysis of the doctor's evidence is worth noting[30]: "The [first respondent's] alleged statement that he had dived into the water because he thought his friend was at risk of drowning is simply out of step with all the other evidence in the case. This purpose for the dangerous dive is not referred to in any other of the notes in the hospital and ambulance records. It appears only in Dr Trevithick's note. The evidence, including Dr Trevithick's, cannot support a suggestion that the [first respondent] was so affected by alcohol as to form a mistaken view that Mr Wilmot was in danger of drowning or to have acted on some inebriated impulse. This being so, the whole of the evidence renders nonsensical the suggested attempt at rescue. Mr Wilmot was never in any danger, nor did he appear to be so. According to the sworn testimony of the [first respondent] and the eye-witnesses, he was standing in shallow water, or even making his way out of it, at the time when the [first respondent] entered the water. There was no rational basis upon which, in these circumstances, the [first respondent] could have made the statement recorded in the doctor's notes. Nor, having regard to the evidence, was there any basis for anyone else making such a suggestion. In this regard, no evidence was called by the [appellant and the second and third respondents] from any persons also present at the scene to support the proposition that the [first respondent] was diving to attempt the rescue of Mr Wilmot. The absence of such evidence strongly suggests that none was available." 1. Having found that the trial judge placed too much weight on the evidence of Dr Trevithick, the question was then as to the orders that should be made for the disposition of the appeal. The first respondent had contended principally for either one of two results in the Court of Appeal, a retrial on all of the issues, or that the Court of Appeal could and should re-determine the matter finally, and give judgment for him. The first respondent did not however confine himself to those contentions. His submissions in reply sought relief in the alternative: "It is submitted his Honour's finding should be set aside and in lieu thereof the Court should find the [first respondent] sustained injury as a result of falling from the deck because he had been either jostled or pushed and remit the matter to the Court below for the determination of the issues left undetermined by the trial judge ... It is noted that the [second and third respondents] in their written submission have adduced no substantive arguments in support of the judgment entered by the trial judge." Counsel for the first respondent contemplated the possibility of an order of the kind that the Court of Appeal decided to make as appears from this exchange: "Foster AJA: I take it the only result if you are successful in this appeal would be a new trial in those circumstances. [Counsel for the first respondent]: There'd have to be a new trial. The only question might be if this Court was persuaded that on the facts his Honour's factual finding as to how the event occurred should be reversed and the [first respondent's] version should be accepted, but apart from that the matter would have to go back for a new trial, and it may be the Court would not be disposed to determine the factual issue but would send that back as well, assuming the appeal was successful." It is true that at a later stage of the hearing of the appeal the presiding judge indicated that if there were to be a retrial it should be on all issues but not only did the first respondent never depart from the position stated in his submissions in reply, but also he orally submitted that the essential issue was whether he had dived or otherwise entered the water voluntarily. In allowing the appeal, the Court ordered that: "there be a new trial of the action conducted on the basis that it is established in favour of the [first respondent] that, through being jostled or pushed, he lost his balance and fell from the jetty into the water". The Court also ordered that the appellant and the second and third respondents pay the first respondent's costs, not only of the appeal but also of the trial. The appeal to this Court 1. The appellant appeals on a number of bases, including that the Court of Appeal was wrong in its assessment of the evidence, and that its decisions and orders should be overturned in their entirety. The second and third respondents support the appellant's appeal and adopt its arguments as well as advance some of their own. 2. The appellant's second ground is that the order made by the Court of Appeal which I have set out above was one that could not and should not have been made. The appellant mounted a third argument that the intimation by the trial judge, and the appellant's reliance on it, amounted to a denial of natural justice. In addition to being a respondent to the appellant Authority's appeal, the Council sought special leave to appeal against the order for costs made against it by the Court of Appeal. 3. The first respondent also seeks to rely upon a notice of contention filed out of time the sole ground of which was: "Order 3 of the Court of Appeal's orders entered on 22 March 2004 was the order which should in any event have been made in the proper exercise of that Court's power pursuant to s 75A of the Supreme Court Act 1970 (NSW) and/or Part 51 Rule 23 of the Supreme Court Rules." 1. The first respondent should be allowed to rely upon that notice. On any view, the power, statutory or otherwise of the Court of Appeal to make the order that it did, was always going to be a central issue in this Court. Section 75A of the Act confers many extensive powers upon the Court of Appeal. Pursuant to that section appeals are to be conducted by way of rehearing[31]. The Court of Appeal has powers, inter alia, to draw inferences and make findings of fact[32], assess damages[33], receive further evidence[34], although only on "special grounds"[35], and to make any finding or assessment, and, give any judgment or make any order or direction which ought have been given or made[36] or the nature of the case requires, pursuant to sub-ss (10) which is as follows: "The Court may make any finding or assessment, give any judgment, make any order or give any direction which ought to have been given or made or which the nature of the case requires." In Fox v Percy I said this of the powers conferred by s 75A[37]: "Section 75A of the Act imposes positive duties upon the State appellate court, the performance of which is in no way conditioned by judge-made rules stated in very different language, and to a substantially different effect from the plain meaning of the section which, by sub-ss (6) and (10) imposes affirmative duties on the Court of Appeal, including to do what the nature of the case requires ... By the Act, the Court of Appeal was armed with all of the ample powers and duties of an appellate court under the Equity Act 1901 (NSW) (ss 81-89), and in particular the duty to rehear the case pursuant to s 82 which might even, for example, permit the Court to undertake a review in exceptional circumstances[38]." In the joint judgment of Gleeson CJ, Gummow and Kirby JJ their Honours said[39]: "The character and features of [an] appeal are governed by the Supreme Court Act 1970 (NSW) ... The nature of the 'rehearing' provided in these and like provisions has been described in many cases. To some extent, its character is indicated by the provisions of the sub-sections quoted. The 'rehearing' does not involve a completely fresh hearing by the appellate court of all the evidence. That court proceeds on the basis of the record and any fresh evidence that, exceptionally, it admits ... The foregoing procedure shapes the requirements, and limitations, of such an appeal. On the one hand, the appellate court is obliged to 'give the judgment which in its opinion ought to have been given in the first instance'[40]. On the other, it must, of necessity, observe the 'natural limitations' that exist in the case of any appellate court proceeding wholly or substantially on the record[41]. These limitations include the disadvantage that the appellate court has when compared with the trial judge in respect of the evaluation of witnesses' credibility and of the 'feeling' of a case which an appellate court, reading the transcript, cannot always fully share[42]. Furthermore, the appellate court does not typically get taken to, or read, all of the evidence taken at the trial. Commonly, the trial judge therefore has advantages that derive from the obligation at trial to receive and consider the entirety of the evidence and the opportunity, normally over a longer interval, to reflect upon that evidence and to draw conclusions from it, viewed as a whole[43]. Nevertheless, mistakes, including serious mistakes, can occur at trial in the comprehension, recollection and evaluation of evidence. In part, it was to prevent and cure the miscarriages of justice that can arise from such mistakes that, in the nineteenth century, the general facility of appeal was introduced in England, and later in its colonies[44]. Some time after this development came the gradual reduction in the number, and even the elimination, of civil trials by jury and the increase in trials by judge alone at the end of which the judge, who is subject to appeal, is obliged to give reasons for the decision[45]. Such reasons are, at once, necessitated by the right of appeal and enhance its utility. Care must be exercised in applying to appellate review of the reasoned decisions of judges, sitting without juries, all of the judicial remarks made concerning the proper approach of appellate courts to appeals against judgments giving effect to jury verdicts[46]. A jury gives no reasons and this necessitates assumptions that are not appropriate to, and need modification for, appellate review of a judge's detailed reasons. Within the constraints marked out by the nature of the appellate process, the appellate court is obliged to conduct a real review of the trial and, in cases where the trial was conducted before a judge sitting alone, of that judge's reasons. Appellate courts are not excused from the task of 'weighing conflicting evidence and drawing [their] own inferences and conclusions, though [they] should always bear in mind that [they have] neither seen nor heard the witnesses, and should make due allowance in this respect'"[47]. 1. It is the duty of courts of appeal therefore to ensure that appeals are fully and carefully considered, and that their consideration not be confined to the identification of errors of law. The great preponderance of cases turn on their facts. Judges at first instance, armed with all of the advantages, sometimes exaggerated, which appellate courts have from time to time attributed to them, of seeing and hearing the witnesses, are not infallible. 2. In this case, in my opinion, the Court of Appeal undertook precisely the sort of review of the facts that was entrusted to it by the legislature. In consequence, it decided to allow the appeal and to order a retrial upon the basis that one issue of fact be foreclosed by, what in substance was a finding that it was prepared to, and did make. As unusual as such a course may be, it was a course open to it, and specifically contemplated by s 75A(10) of the Act. I am unable to say therefore that the Court of Appeal should not have done that, either for the reasons advanced in argument by the appellant which I will shortly examine, or otherwise. 3. But first reference should be made to Pt 51, r 23 of the Supreme Court Rules (NSW) which provides a further source of power and is as follows: "51.23 New trial (1) The Court of Appeal shall not order a new trial: (a) on the ground of misdirection, non-direction or other error of law, (b) on the ground of the improper admission or rejection of evidence, (c) where there has been a trial before a jury, on the ground that the verdict of the jury was not taken upon a question which the trial judge was not asked to leave to the jury, or (d) on any other ground, unless it appears to the Court of Appeal that some substantial wrong or miscarriage has been thereby occasioned. (2) The Court of Appeal may order a new trial on any question without interfering with the decision on any other question. (3) Where it appears to the Court of Appeal that some ground for a new trial affects part only of the matter in controversy, or one or some only of the parties, the Court of Appeal may order a new trial as to that part only, or as to that party or those parties only. (4) Where the Court of Appeal makes an order under subrule (2) or subrule (3), the Court of Appeal may give such judgment or make such order as the nature of the case requires for the disposal of the remainder of the appeal. (5) Where the Court of Appeal orders a new trial, the Court of Appeal may: (a) impose conditions on any party for the purposes of the new trial, (b) direct admissions to be made by any party for the purpose of the new trial, (c) order that the testimony of any witness examined at the former trial may be read from the notes of the testimony, instead of the witness being again examined, and (d) for the purposes of subparagraphs (a) to (c) from time to time make such orders as the Court of Appeal thinks fit." 1. If there were any question about the amplitude and sufficiency of the powers of the Court of Appeal under s 75A of the Act to do what the Court of Appeal did here, which in my opinion there is not, there is no doubt that the rules which I have just quoted, especially sub-r 3, singly and collectively would resolve that question. In terms those sub-rules permit the Court of Appeal to do exactly what it did: effectively to order a new trial as to "part[s] only of the matter[s] in controversy", the former being every issue except as to the involuntariness of the first respondent's entry into the water. 2. I turn now to the arguments of the appellant and the second and third respondents. The first submission of the appellant was that the circumstances of the trial and the facts generally were not so exceptional as to justify the highly unusual course which was adopted by the Court of Appeal, even though what was done may have literally been within the power conferred by s 75A of the Act. The appellant developed this submission by contending that the order of the Court of Appeal presented a severe handicap to the appellant and the second and third respondents in mounting their case of causative and contributory negligence against the first respondent at the retrial. 3. It is right, as the appellant submits, that orders for retrials on a restricted basis are rare, and should be reserved for exceptional cases. Indeed, as I have said on other occasions[48], the perceived advantages of splitting trials are often illusory, and can create more problems than they resolve. The problems are likely to be greater and more numerous however in jury cases as the ones cited by the appellant were[49]. 4. The appellant made the submission that it had been denied natural justice. That submission was based upon the matters to which I have referred, namely the intimation given at one stage of the appeal by the presiding judge that if there were to be a retrial it should be on all issues. The balance of the appellant's submissions were directed to the intervention of the Court of Appeal on questions of fact generally, and the absence of any sufficient basis for their Honours to do so at all in this case. 5. For the second respondent it was submitted that the order made by the Court of Appeal would inhibit the appellant and the second and third respondents on a retrial in contesting the quantum of damages: that the division between that issue and the issue of liability can be misleading. Reference was also made to difficulties which would arise in establishing causation, or absence of causation, as well as contributory negligence. The second respondent also sought to make the point that it would be denied the opportunity of attacking the credibility of the witnesses called on liability by the first respondent if (and to the extent that) they were witnesses on quantum as well. In advancing these submissions the second respondent put matters in rather extravagant language, that the retrial would be "crippled" or "a travesty of a trial". A number of further submissions were made concerning the differences in expression between the first respondent's witnesses on liability. Those differences appear to me to be only minor and semantic, particularly having regard to the ambiguities latent in the word "dive". A further criticism was made of the Court of Appeal: that it gave weight, indeed undue weight to a piece of hearsay evidence, the statement by Mr Moon that he heard the first respondent say: "Who pushed me in?". 6. The principal submission of the third respondent was also that the inhibitions upon the conduct of the retrial were so great as to make it, in effect, practically impossible to be conducted fairly. The difficulties were identified as being, as to the objections which might be taken to evidence; the need for the parties to reflect on the meaning of the order of the Court of Appeal in advance of the retrial; and the need for the parties to make decisions as to the witnesses to be called, and the questions to be put to them. 7. It is convenient to deal with the third respondent's submission first. In my view, on examination there is very little substance in it, as, ultimately, the third respondent was bound to concede in argument. With respect to the asserted problem, as to how an objection could properly be made if a question were asked that was relevant to the pleading of contributory negligence that did not raise any issue of the voluntariness of the first respondent's entry into the water, the only response that counsel was able to make, was, that although no objection could legitimately be sustained, enterprising counsel might still attempt to persuade a trial judge that the question could not be asked. 8. With respect to the issue of causation, the third respondent initially adopted the stance that on the approach of the Court of Appeal, it would be unable to argue that the first respondent's entry into the water was caused or contributed to by anything that he had done. Subsequently however, the third respondent accepted that it would still be open for it to argue that the first respondent stood too close to the edge of the jetty, or put himself in a position to be jostled, thereby causing or contributing to his injuries. Similarly, the third respondent was unable to draw any real distinction between jostling or pushing in the circumstances of the case. And again, without being able to demonstrate why this would be so, the third respondent contended that a discrete admission under Pt 51 r 23 that it might have been required to make, would not have caused the same difficulties as the finding and order of the Court of Appeal. 9. The last matter to which the appellant and the second and third respondents referred was the order of the Court of Appeal that the first respondent should have his costs not only of the appeal but also of the trial, an argument which will require separate attention. 10. Even though the course adopted by the Court of Appeal was unusual, indeed perhaps even highly unusual, it was as I have said, open to it. In my opinion, the perceived difficulties to which the appellant and the second and third respondents have referred are largely illusory, as in substance the third respondent was bound to concede. A retrial will not be crippled and certainly will not be a travesty. The issue of causation remains open. It is not inconceivable that the first respondent could fail entirely on the issue of negligence even though there can no longer be any question about the voluntariness of his descent into the water. That the jetty lacked a railing was something that all could see. Similarly, there was no reason for anyone including the first respondent to make any assumptions about the depth of the water beside the jetty. The width of the jetty was a matter of ordinary observation, and that it could accommodate only a limited number of people in a huddle at any one time must have been obvious. It would be open for the appellant and the second and third respondents on any retrial to make use of these matters as they see fit, and to contend, if they are so minded, that nothing that they did or omitted to do in conducting the ball, in choosing that location for it, and in approving the structure of the jetty and the form that it took, involved negligence on the part of any of them. In litigious times, it is sometimes overlooked that accidents, even accidents with catastrophic consequences, can occur without fault on the part of anyone. So too, there are cases in which a plaintiff will fail simply because he or she is unable to establish on the balance of probabilities that anyone was negligent. 11. I turn now to the submission that the Court of Appeal should not have intervened in the case, whether by making the orders that it did, or at all. I would reject this submission. In my view the challenge must fail. The omission to refer to Mr Treharne's evidence was a serious one. That the first respondent would voluntarily dive into the water is improbable. And whilst no doubt Dr Trevithick was impartial, allowance should have been made for the fact that whether the first respondent entered the water voluntarily or involuntarily was completely irrelevant to the matter with which the doctor was concerned, the assessment and treatment of the first respondent which were being undertaken on an exceedingly busy night. As to the criticism, as hearsay, of the evidence of the first respondent's question "Who pushed me in?", it is sufficient to say that it was in evidence, and provided a basis at least for a contention that the doctor may have misunderstood, or not accurately recorded what a highly distressed person was saying very soon after suffering the massive injury that he did. The question mark which appears in written versions of the first respondent's account, associated as they were, with the word "dived" add nothing, except perhaps to highlight the ambiguity inherent in the use of the word "dived" in the circumstances of this case. 12. Having regard to the matters to which I have discussed intervention by the Court of Appeal was not only open, but almost inevitable. 13. I do not think that the appellant and the second and third respondents have been denied natural justice. It was at first instance in Stead v State Government Insurance Commission[50] that the judge intimated that a party need not address on a particular issue. In this case, there was some argument in the Court of Appeal on the issue which has loomed largest in this Court, the form of the Court of Appeal's order. And now the issue has been fully argued here. Everything that could be said on behalf of the parties has been said. Furthermore, a retrial on the restricted basis should be less costly than a trial on all issues. The restriction is plainly stated. The parties will be, following the full argument in this Court, well aware of what they will need to do in preparing and conducting their cases. 14. The Court of Appeal should not however have made the order that it did in respect of the costs of the trial in favour of the first respondent. There was no proper basis for such an order. It may turn out, if the first respondent's case were ultimately to fail, that the appellant and the second and third respondents should never have incurred any costs of a trial at all. The proper order is that the costs of the trial should abide the result of the retrial. 15. The orders that I would make are that that the applications for special leave to appeal to this Court should be granted, but that the appeals should be dismissed except the appeals with respect to the costs of the trial. I would therefore make the following orders: 1. The time for filing the appellant's notice of contention filed on 13 May 2005 be extended as necessary. 2. The appellant's appeal be dismissed with costs subject to the following orders. 3. That the appellant and the second and third respondents pay the first respondent's costs of the appeal. 4. Order 4 of the Court of Appeal made on 3 December 2003 be set aside and in lieu thereof order that the respondents pay the appellant's costs of the appeal to that Court. 5. That the time for filing the applications for special leave in S98/2005 and S131/2005 be extended as necessary. 6. That the applications for special leave be granted. 7. That the applications be treated as appeals and heard instanter. 8. That the appeals be allowed with respect to order 4 made by the Court of Appeal on 3 December 2003 insofar as it relates to the costs of the first trial. 9. That the appeals otherwise be dismissed. 10. That the costs of the first trial abide the result of the retrial. 1. Dakhyl v Labouchere [1908] 2 KB 325 (n) at 327 per Lord Loreburn LC. ↑ 2. Ketteman v Hansel Properties Ltd [1987] AC 189 at 220 per Lord Griffiths, approved in State Pollution Control Commission v Australian Iron & Steel Pty Ltd (1992) 29 NSWLR 487 at 494 per Gleeson CJ. ↑ 3. Fitzgibbon v The Waterways Authority [2003] NSWCA 294 at [106], adapting the words of Chilwell J in Connell v Auckland City Council [1977] 1 NZLR 630 at 634, as they had been by Kirby P in Soulemezis v Dudley (Holdings) Pty Ltd (1987) 10 NSWLR 247 at 259. ↑ 4. Abalos v Australian Postal Commission (1990) 171 CLR 167 at 178 per McHugh J, quoting Watt or Thomas v Thomas [1947] AC 484 at 488 per Lord Thankerton. ↑ 5. Fitzgibbon v The Waterways Authority [2003] NSWCA 294 at [69]. ↑ 6. Section 64(3) provides: "(3) If the person who made the representation has been or is to be called to give evidence, the hearsay rule does not apply to evidence of the representation that is given by: 1. that person; or 2. a person who saw, heard or otherwise perceived the representation being made; if, when the representation was made, the occurrence of the asserted fact was fresh in the memory of the person who made the representation." Section 67(1) provides: "(1) Subsections 63(2), 64(2) and 65(2), (3) and (8) do not apply to evidence adduced by a party unless that party has given reasonable notice in writing to each other party of the party's intention to adduce the evidence." ↑ 7. Fitzgibbon v The Waterways Authority [2003] NSWCA 294 at [72]. ↑ 8. Fitzgibbon v The Waterways Authority [2003] NSWCA 294 at [73]. ↑ 9. Fitzgibbon v The Waterways Authority [2003] NSWCA 294 at [75]. ↑ 10. (1979) 142 CLR 531. ↑ 11. (2003) 214 CLR 118. ↑ 12. Oxford English Dictionary, 2nd ed (1989), meaning I.1.a. The Macquarie Dictionary (Federation Edition), (2001) vol 1, gives as meaning 1: "to plunge, especially head first, as into water". ↑ 13. Reasons of Callinan J at [175]-[179]. ↑ 14. Reasons of Gleeson CJ at [15], McHugh J and Gummow J agreeing; reasons of Hayne J at [132]. ↑ 15. SS Hontestroom v SS Sagaporack [1927] AC 37 at 47. ↑ 16. (1957) 97 CLR 521 at 527, quoting Hutchinson v Piper (1812) 4 Taunt 555 at 556 [128 ER 447 at 448] per Gibbs J. ↑ 17. Carlson v King (1947) 64 WN (NSW) 65 at 66 per Jordan CJ cited in Soulemezis v Dudley (Holdings) Pty Ltd (1987) 10 NSWLR 247 at 257 per Kirby P. ↑ 18. Pettitt v Dunkley [1971] 1 NSWLR 376 at 382 per Asprey JA. ↑ 19. Supreme Court Act 1970 (NSW). ↑ 20. Supreme Court Rules 1970 (NSW), Pt 51. ↑ 21. (2003) 214 CLR 118. ↑ 22. (2003) 214 CLR 118 at 125‑126 [23] per Gleeson CJ, Gummow and Kirby JJ. ↑ 23. Fox v Percy (2003) 214 CLR 118 at 125‑126 [23] per Gleeson CJ, Gummow and Kirby JJ. ↑ 24. Fitzgibbon v The Waterways Authority [2002] NSWSC 1230 at [40]. ↑ 25. Fitzgibbon v The Waterways Authority [2002] NSWSC 1230 at [26]-[27], [34]-[35] and [41]. ↑ 26. (1959) 101 CLR 298. ↑ 27. Connell v Auckland City Council [1977] 1 NZLR 630 at 634. ↑ 28. Fitzgibbon v The Waterways Authority [2003] NSWCA 294 at [74]. ↑ 29. Fitzgibbon v The Waterways Authority [2003] NSWCA 294 at [75]. ↑ 30. Fitzgibbon v The Waterways Authority [2003] NSWCA 294 at [99]. ↑ 31. s 75A(5). ↑ 32. s 75A(6)(b). ↑ 33. s 75A(6)(c). ↑ 34. s 75A(7). ↑ 35. s 75A(8). ↑ 36. s 75A(10). ↑ 37. Fox v Percy (2003) 214 CLR 118 at 164-165 [146]-[147]. ↑ 38. Attorney-General (NSW) v Wheeler (1944) 45 SR(NSW) 321. ↑ 39. Fox v Percy (2003) 214 CLR 118 at 125-127 [21]-[25]. ↑ 40. Dearman v Dearman (1908) 7 CLR 549 at 561. The Court there was concerned with s 82 of the Matrimonial Causes Act 1899 (NSW) which provided that "on appeal every decree or order may be reversed or varied as the Full Court thinks proper": see Dearman v Dearman (1908) 7 CLR 549 at 558. ↑ 41. Dearman v Dearman (1908) 7 CLR 549 at 561. See also Scott v Pauly (1917) 24 CLR 274 at 278-281. ↑ 42. Maynard v West Midlands Regional Health Authority [1984] 1 WLR 634 at 637; [1985] 1 All ER 635 at 637 per Lord Scarman, with reference to Joyce v Yeomans [1981] 1 WLR 549 at 556; [1981] 2 All ER 21 at 26. See also Chambers v Jobling (1986) 7 NSWLR 1 at 25. ↑ 43. State Rail Authority (NSW) v Earthline Constructions Pty Ltd (In Liq) (1999) 73 ALJR 306 at 330 [89]-[91]; 160 ALR 588 at 619-620, citing Lend Lease Development Pty Ltd v Zemlicka (1985) 3 NSWLR 207 at 209-210; Jones v The Queen (1997) 191 CLR 439 at 466-467. ↑ 44. Builders Licensing Board v Sperway Constructions (Syd) Pty Ltd (1976) 135 CLR 616 at 619-620; State Rail Authority (NSW) v Earthline Constructions Pty Ltd (In Liq) (1999) 73 ALJR 306 at 322-325 [72]-[80]; 160 ALR 588 at 609-613. ↑ 45. Public Service Board of NSW v Osmond (1986) 159 CLR 656 at 666-667 citing Housing Commission (NSW) v Tatmar Pastoral Co Pty Ltd [1983] 3 NSWLR 378 at 386; Soulemezis v Dudley (Holdings) Pty Ltd (1987) 10 NSWLR 247 at 257-258, 268-273, 277-281. ↑ 46. eg, Hocking v Bell (1945) 71 CLR 430; Hocking v Bell (1947) 75 CLR 125 at 131-132; cf Naxakis v Western General Hospital (1999) 197 CLR 269 at 271-272 [2], 274-275 [16], 282-283 [41]-[42], 288-290 [57]-[58], 310-311 [119]-[123]. ↑ 47. Dearman v Dearman (1908) 7 CLR 549 at 564, citing The Glannibanta (1876) 1 PD 283 at 287. ↑ 48. for example, Perre v Apand Pty Ltd (1999) 198 CLR 180 at 332 [436]. ↑ 49. Pateman v Higgin (1957) 97 CLR 521; Balenzuela v De Gail (1959) 101 CLR 226. ↑ 50. (1986) 161 CLR 141. ↑