The plaintiff's arguments
28 The plaintiff's oral arguments to this Court opened by placing considerable emphasis on a matter which played a much more minor role in the Notice of Appeal and in the plaintiff's written submissions. This concerned the delay of about one year from the time when the trial ended to the time when judgment was delivered. The plaintiff drew the court's attention to the following authorities.
29 Goose v Wilson Sandford & Co (unreported, 13 February 1998) was a case in which the trial judge had reserved judgment on 13 July 1994, but did not deliver judgment until 1 April 1996 - about twenty-one months. In the trial, opening submissions lasted two days, the evidence lasted seventeen days, and closing submissions lasted eight days. There were daily transcripts. The documentary evidence was "huge". The English Court of Appeal (Peter Gibson, Brooke and Mummery LJJ) referred to Rolled Steel Ltd v British Steel Corporation [1986] Ch 246, in which Lawton LJ criticised reservation for eight months after a nineteen day trial, and Bishopsgate Investment Management Ltd v Maxwell [1993] BCC 120 in which the English Court of Appeal, speaking through Hoffmann LJ, said that a five month delay after a five day trial was excessive. Their Lordships then said:
"As the judge himself was the first to recognise, a delay of this magnitude was completely inexcusable. The Plaintiff, who was not a young man, was claiming that Mr Wilson's fraudulent conduct had been causative of his financial ruin. Mr Wilson for his part was a professional man charged with serious professional misconduct amounting to fraud. Both parties were entitled to expect to receive judgment before Christmas 1994 at the very latest. The fact that they were obliged to wait another year and a quarter, even allowing for the judge's illness, is wholly unacceptable.
The plaintiff's first ground of appeal was that the court should infer that the judge had forgotten large parts of the essential facts and evidence in the case, and that he had no clear recollection or impression of the demeanour of the witnesses of fact or their credibility by the time he came to deliver his judgment. Our attention was drawn to certain mistakes he made, to which reference is made in this judgment. We were also told that the judge mislaid a detailed chronology on which he had made manuscript notes of counsel's opening submissions on the facts and also the written closing submissions prepared by counsel for Mr Goose. Replacement documents were requested and supplied in November 1995, but the judge's own notes could not be replaced.
In these circumstances we were invited to find that the judge's misdirections had occasioned such a substantial wrong or miscarriage of justice that we should order a new trial (see Order 59 Rule II). Both parties had incurred very heavy costs in the original trial, and the Plaintiff was and is in receipt of legal aid. There are no provisions enabling a court to make an order out of central funds to compensate the parties in a situation like this. These considerations illustrate some of the very serious problems that are likely to arise if a judge delays giving judgment in this extraordinary way.
A judge's tardiness in completing his judicial task after a trial is over denies justice to the winning party during the period of the delay. It also undermines the loser's confidence in the correctness of the decision when it is eventually delivered. Litigation causes quite enough stress, as it is, for people to have to endure while a trial is going on. Compelling them to await judgment for an indefinitely extended period after the trial is over will only serve to prolong their anxiety, and may well increase it. Conduct like this weakens public confidence in the whole judicial process. Left unchecked it would be ultimately subversive of the rule of law. Delays on this scale cannot and will not be tolerated. A situation like this must never occur again.
Because of the delay in giving judgment, it has been incumbent on us to look with especial care at any finding of fact which is now challenged. In ordinary circumstances where there is a conflict of evidence a judge who has seen and heard the witnesses has an advantage, denied to an appellate court, which is likely to prove decisive on an appeal unless it can be shown that he failed to use, or misused, this advantage. We do not lose sight of the fact that the judge had transcripts of the evidence, as well as very extensive written submissions from counsel. But the very fact of the huge delay in itself weakened the judge's advantage, and this consideration had to be taken into account when we reviewed the material which was before the judge. In a case as complex as this, it is not uncommon for a judge to form an initial impression of the likely result at the end of the evidence, but when he has come to study the evidence (both oral and written) and the submissions he has received with greater care, he will then go back to consider the effect the witnesses made on him when they gave evidence about the matters that are now troubling him. At a distance of 20 months, Harman J denied himself the opportunity of making this further check in any meaningful way."
30 The plaintiff then referred to R v Maxwell (unreported, 23 December 1998), in which a judge of this Court was criticised for having reserved judgment for ten months in a murder case which apparently lasted about three weeks. The Court of Criminal Appeal (Spigelman CJ, Sperling and Hidden JJ) set out the second last and last paragraphs quoted above from Goose v Wilson Sandford & Co, and adopted them. They added, in language which the plaintiff submitted was equally applicable to civil cases:
"Public confidence in the judicial process is of particular significance in the administration of the criminal law. Indeed one of the most important aspects of the welfare and stability of Australian society is the hard earned, but widespread, belief that judges administering the criminal law do so with competence, fairness and impartiality. The administration of criminal justice is of great significance in the preservation of the liberty of Australian citizens. This Court must be rigorous in ensuring that judicial conduct in the course of criminal trial does nothing to disappoint the high expectations which the community as a whole has of the judiciary in this regard."
They also said that the considerations described by the English Court of Appeal:
"have informed this Court in its review of the reasons for judgment given by his Honour, specifically the statements made by his Honour in his judgment of a general assertive character, which in the normal course would be accepted as encompassing a detailed consideration of the evidence before him, have been treated by us with reserve. Indeed, a delay of the order of ten months is, of itself, such as to require a more comprehensive statement of the relevant evidence than would normally be required, in order to manifest, for the parties and the public, that the delay has not affected the decision. For the reasons we will give below, his Honour's judgment is unsatisfactory in this regard. His Honour failed to give reasons of a character which the law required him to give in all the circumstances of the case, including the circumstances of the delay between the trial and delivery of judgment."
31 Later they said, after setting out the statutory duty to give reasons in criminal cases tried without a jury created by the Criminal Procedure Act 1986 (NSW) s 33(2), the High Court's analysis of it in Fleming v R (1998) 197 CLR 250 at [22], and McHugh JA's analysis in Soulemezis v Dudley (Holdings) Pty Ltd (1987) 10 NSWLR 247 at 278 of the judicial duty to give reasons:
"These principles are applicable to the instant case. Indeed, for the reasons we have given above, the delay in delivery of judgment made it even more necessary than usual for his Honour to err on the side of thoroughness in the exposition of his reasoning process. The Appellant had a right to expect that the arguments put on his behalf would be dealt with in such a way that he could be satisfied that they had been understood and, either accepted, or, if rejected, that the rejection was based on a clear and rational process of reasoning.
Furthermore, the public interest in the administration of criminal justice required that his Honour should have put beyond question any suggestion that the substantial delay of ten months had affected his Honour's understanding of the scope and range of issues, and to demonstrate that all relevant facts had been properly and carefully assessed and all pertinent submissions dealt with. As we have said above, matters that could have been dealt with by means of a brief summary reference in a decision given at a time when all parties would have accepted that relevant evidence and submissions were fresh in his Honour's mind, could not be dealt with in the same way after a delay of the magnitude which occurred in this case."
32 The Court of Criminal Appeal then quoted the following passage from Mahoney JA's judgment in a case in this Court, New South Wales Medical Defence Union Ltd v Crawford (No 2) (unreported, 30 June 1994):
"His Honour accepted the Todd-Taylor evidence. He referred in terms to the conclusions which they expressed. Ordinarily I would regard myself as constrained to accept those conclusions. Each of the psychologists gave evidence before the judge. I would ordinarily accept that, absent indications to the contrary, the judge was influenced in his acceptance of witnesses by what he saw of them. … The trial judge's advantage operates, or may operate, in respect of expert witnesses. …
But an appellate court may depart from the conclusions of a trial judge where, inter alia, it appears that he has not made use, or proper use, of the advantage that he had. In the present case the trial judge did not refer in relevant respects to the Evan Davies' evidence. He did not indicate his understanding of that evidence, how he contrasted it with the Todd-Taylor evidence, or why he preferred the latter to the former. It is not necessary that a judge, in his judgment, refer to every aspect of the case or every step in the reasoning to his conclusion … nor should it be readily inferred that, because evidence is not referred to, it has been overlooked. … Such an inference is warranted only where a reference to the evidence is to be expected. Where the purpose of the judgment is to articulate conclusions rather than to detail reasoning or where, as here, the facts are complex and the judgment is necessarily long, such an inference will not ordinarily be drawn.
However, the contrast of the psychometric evidence in this case is of considerable significance. It was, in my opinion, to be expected that the judge would deal with it. As I shall indicate the acceptance of the Evan Davies' evidence would not be determinative of the existence of brain damage but it would be a matter of considerable importance in assessing the whole of the evidence on this issue. If he did not in fact deal with it, he would, I think, have failed properly to assess all of the evidence. … A judge must, in deciding an issue of fact, take into account all of the evidence that is relevant to it. But he may in some circumstances put aside evidence where that is simply unacceptable. But the Evan Davies' evidence was not evidence which could simply be put aside as a mere preliminary to the acceptance or rejection of the Todd-Taylor evidence. Unless it was rejected or at least significantly qualified, it was difficult for the judge to accept the Todd-Taylor evidence as the judge did.
In the circumstances I do not think that the Evan Davies' evidence can be put aside as the judge did. On the face of it, it was no less acceptable than the Todd-Taylor evidence. There was no reason why a person approaching the diagnosis of brain damage in this case would not give credence to it. The main psychometric witnesses did not, for example, merely put it aside as inherently flawed or for reasons of that kind. Therefore it would be rejected in the Todd-Taylor evidence referred if, essentially, Dr Evan Davies was to be rejected. … I am satisfied that that is not the reason why the judge put his evidence aside in this case. What a judge, including this judge, does is to be inferred from the judgment and its context. If the judge [concluded] that Dr Evan Davies' evidence was to be put aside for a reason of this kind, he would I think have said so. … The conflict of psychometric evidence was too important to be ignored or to be left to an Abalos inference."
33 The next case to which the plaintiff referred was a decision of this Court, Moylan v The Nutrasweet Co (unreported, 24 November 2000). In that case the evidence at trial was taken between 24 January and 13 September 1996. Oral submissions were heard between 27 October and 23 December 1997. The evidence was voluminous. Judgment was reserved for fourteen months. Sheller JA (Beazley and Giles JJA concurring) referred to R v Maxwell, set out some of the passages quoted above, and held that the trial judge in the case before them had breached the relevant duty.
34 The plaintiff then submitted that at some point the passing of time from the moment when judgments are reserved causes delay to arise which alters the normal approach of appellate courts. In this summary of the plaintiff's submissions, this type of delay is called "operative delay" for short. In cases not affected by operative delay, an appellate court tends not to assume that the mere failure to refer to evidence meant that it had been overlooked, or that other forms of error had occurred. But if there were operative delay, no favourable assumptions could be made, and it was up to the trial judge to put beyond question any suggestion that he or she had lost an understanding of the issues. Something should be said about how the possible effects of delay on the judicial process have been overcome. Some explanation should be given as to how the trial judge had recorded or recaptured impressions formed of witnesses at the time they testified. A judge might, for example, say "I have a perfect recollection of all the characters in the trial" or "I have contemporaneous notes of my impressions". The plaintiff submitted that the delay here went beyond any minimum period of "operative delay". The plaintiff submitted, indeed, that the delay was "beyond the pale". But the trial judge made no statements of the kind just indicated, and no assumption in her favour that she had retained any relevant impressions could be made.
35 Next, the plaintiff submitted that the trial judge was obliged to demonstrate that all the facts had been carefully assessed; yet some were admitted and there were mistakes in relation to others. It was not possible to make favourable assumptions that the omitted matters had been considered or that the mistakes were insignificant.
36 The plaintiff went on to submit that the delay had disabled the trial judge from effectively checking whatever impressions of demeanour she had against a reading of the evidence.
37 Finally, the plaintiff submitted that the trial judge was under a duty to be extremely thorough in setting out her reasoning processes, in showing that the plaintiff's arguments were understood, and in demonstrating that, if they were rejected, they were rejected for clear reasons.
38 The other oral submissions of the plaintiff, and almost all his written submissions, were dedicated to a detailed attack on many parts of the trial judge's reasoning. It must be said that some of the plaintiff's arguments were more forceful and significant than others. Because of the need to order a new trial at which arguments similar to those advanced to this Court may have to be considered, it is unnecessary to deal with all of the plaintiff's arguments, or with the defendant's responses to them. But before considering the reasons why a new trial should be ordered, it is convenient to say something about the arguments based on delay.