It is an honour to be invited to deliver the 2014 Sir Maurice Byers Lecture. Sir Maurice was the outstanding appellate advocate of his generation. His unrivalled appellate practice was of the kind that rarely required him to trouble the court with the facts. So it may seem rather pedestrian to select the subject of 'Appellate Review of the Facts' in a lecture delivered in his honour.
When Sir Maurice reflected on the changes that he had witnessed over nearly 50 years of practice at the Bar, prominent among those changes was the increase in complexity and cost of litigation. He favoured radical procedural changes to reduce delays and cost1. The Hon A M Gleeson AC QC when delivering this Lecture last year identified the abolition of most forms of civil jury trial as an obvious cause of that increase in cost and complexity2. The loss of the practical finality that accompanies the jury's verdict opened seemingly limitless opportunities for appellate challenge to the trial court's findings of fact. These remarks were not new to readers of Gleeson CJ's judgments in which his Honour on more than one occasion deprecated the view of the trial as merely the first round in the forensic contest3. They are remarks that direct attention to the principles that govern appellate review of the trial court's factual decisions.
Any system that lays claim to administering civil justice must make provision for the correction of error. Appellate review under s 75A of the Supreme Court Act 1970 (NSW) (and the equivalent provisions in the other Australian jurisdictions4) is by way of re-hearing on law and fact. The difficulty faced by the appellate court in determining that a challenged finding of fact is a wrong finding is reflected in the principles of restraint that apply to the review of fact.
Appellate review of the kind provided in s 75A is traced to the Judicature Acts 1873-1875 (UK). The principles applied to an appeal by way of re-hearing were stated in 1898 by Lindley MR, delivering the judgment of the Court of Appeal (Lindley MR, Rigby and Collins LJJ) in Coghlan v Cumberland5. His Lordship's statement is in language that remains familiar. In summary, Lindley MR said that it is the duty of the appellate court: to re-hear the case; to reconsider the materials before the trial judge together with such material as the appellate court may have decided to admit; to make up its own mind, not disregarding the decision below, but carefully weighing and considering it; not to shrink from overruling the decision if it is wrong; to be sensible of the great advantage of the trial judge in seeing and hearing the witnesses and, when the decision turns on which witness is to be believed, the appellate court must be guided by the impression made on the trial judge; but circumstances quite apart from manner and demeanour may show whether a statement is credible and may warrant the appellate court differing from the trial judge6.
The principle of restraint is not without its critics. It is argued that the statute conferring the jurisdiction to determine appeals on law and fact provides no warrant to confine review of the latter by deference to the trial judge's findings. Considerations of finality and of the capacity of well-resourced litigants to exhaust the reserves of less well-resourced opponents on this analysis are misplaced. It is an approach that invokes Lord Atkin's statement "finality is a good thing, but justice is a better"7. That pithy statement was made in the context of determining the appeals of a number of men who had been convicted of murder and sentenced to death following a trial at which a juror did not understand English, which was the language in which the trial had been conducted. The demands of justice were not difficult to identify in that case.
The demands of justice may take on a different complexion when considering appellate review of an action that has been determined following a fair trial at which the parties have had a full opportunity to present their respective cases and in which the trial judge has decided disputed questions of fact in a reasoned judgment that is not evidently attended by error.
Sir Thomas Bingham, writing extra-curially in the mid-1980s at a time when he was Master of the Rolls, suggested that a respectable rule would allow that "every litigant should be entitled to a full contest on the facts at one level only and that the facts should be open to review thereafter only if some glaring and manifest error could be demonstrated"8. In the event, concern about the cost and complexity of civil litigation in England and Wales has led to a more radical curtailment of the right to appellate review.
It is conventional to justify the restraint applied to findings that are substantially dependent on the assessment of credibility by reference to the trial judge's advantage in having seen and heard the oral evidence. The assumption underpinning this understanding has been questioned for more than a quarter of a century in light of psychological research casting doubt on the ability to discern truthfulness from an individual's physical presentation9. Acknowledgment of the strength of this body of research has led some commentators to question the foundation for the application of differing standards of review of findings of fact.
Even if it were accepted that the trial judge enjoys no advantage in the assessment of the oral evidence, it would remain to consider whether the value of finality warrants restraint in any event. Sir Thomas Bingham suggested that his "respectable rule" be squarely sourced in finality and not in deference to the trial judge's supposed advantage10.
The principles stated by Lindley MR have been adopted and applied by the High Court in decisions commencing with McLaughlin v Daily Telegraph Newspaper Co Ltd (No 2)11. Although, as the joint reasons in Fox v Percy neutrally observed, in the circumstances of particular cases the principles have been given differing emphasis12. The force of that observation is illustrated by the separate reasons of McHugh and Callinan JJ in Fox v Percy.
In that case, it will be recalled, the New South Wales Court of Appeal overturned Herron DCJ's finding, based upon his acceptance of the evidence of Ms Fox and her witness, that Ms Percy's car was on her incorrect side of the road at the point of the collision13. The Court did so because skid marks on the road (about which there was no contest) incontrovertibly established the contrary14. The fact that 11 years after the collision the High Court should have been poring over the evidence of the skid marks, in Professor Luntz's view, is a "disgrace" to the administration of justice15. This is because, in Professor Luntz's analysis, intermediate appellate courts should not be subject to any principle of restraint in reviewing the trial court's factual findings16. Trial judges in his view are as likely to get the facts wrong as the law and restraint may occasion practical injustice.
Professor Luntz is not alone among distinguished commentators in considering that appellate courts should unshackle themselves from the restraints conventionally accepted as arising from the trial judge's advantage in seeing and hearing the evidence17. In an account of the work of the English Court of Appeal, Professor Drewry, Sir Louis Blom-Cooper QC and Charles Blake argue that the deference accorded the decision of the lower court's credibility-based findings should be understood as the product of Victorian cases decided before the invention of photocopying, word-processing and tape-recording. In the context of modern litigation, in which much evidence is documentary, they suggest that this long line of authority is in need of re-examination18.
Some colour is lent to Professor Luntz's criticism of the grant of leave in Fox v Percy by the circumstance that, on the hearing of the appeal, there was no challenge to the principles enunciated in the Victorian cases and affirmed in the trilogy of decisions culminating in Devries v Australian National Railways Commission19. The High Court was unanimous in upholding the decision of the Court of Appeal given that no deference to Herron DCJ's assessment of credibility could stand in the way of the skid marks.
Justice Callinan, while content to decide the appeal in the way it had been argued, took the opportunity to state his view that Devries imposes an "emphatically high test" that pays insufficient regard to the jurisdiction conferred by s 75A20. The same view had been earlier expressed by Kirby J in State Rail Authority (NSW) v Earthline Constructions Pty Ltd (in liq)21. Justice Kirby considered Lindley MR's statement of the principles as reflective of 19th century judicial disdain for the messy business of fact-finding22. He was particularly critical of Lord Sumner's restatement of the principles for introducing the concept of the "palpable misuse of the trial judge's advantage"23 into the discourse. The phrase, redolent of judicial misconduct, Kirby J saw as imposing an unduly demanding requirement for the demonstration of error; a requirement not justified by the text of s 75A or the concept of "appeal" itself24.
The belief in the oracular power of the judge to divine the truth has been out of vogue for as long as I have been a judge. In my experience, trial judges are alive to the importance of contemporary materials and are inclined to weigh the probabilities in light of those materials. Nonetheless, it still occurs that in some cases disputed facts fall to be resolved by the acceptance or rejection of oral evidence. In these cases, is the appellate court right to continue to be guided by the impression made on the judge who saw and heard the evidence?
The Hon David Ipp AO QC has argued that the principle of restraint should be relaxed: appellate courts should regard demeanour-based findings, which are contrary to the probabilities, as raising appellable error absent adequate reasons for them25. Such a rule, he suggests, would enhance the administration of justice by setting aside the "virtually untrammelled power of trial judges" to make what amount to final decisions based on the judge's assessment of the witness' physical reactions in testifying26. The restraint currently applied is, in his view, "an anachronism in a system of justice that prides itself on objectivity and rationality"27.
This view finds support in Callinan J's analysis in Fox v Percy. His Honour observed that few decisions can be said truly to turn on a mere "gesture, a tone or emphasis, a hesitation or an undue or unusual alacrity in giving evidence"28. No doubt most trial judges would agree that it is a rare case that turns on a mere gesture. But many might acknowledge that the impression formed by seeing and hearing the evidence plays an important part in the determination of some disputed questions of fact. David Ipp says that in his experience a judge "cannot help but develop antennae sensitive to deliberate untruths"29. The psychologists may tell us that this puts it too high. It remains that a judge, alive to his or her limitations in ascertaining truth, may nonetheless assess that no reliance could fairly be placed on a witness' account of events.
An impression that testimony is unworthy of belief will almost certainly be the subject of an express finding. However, not every impression formed by the trial judge in the course of seeing and hearing the evidence will form part of the reasons. Lord Hoffmann made the point in Biogen Inc v Medeva plc30:
"The need for appellate caution in reversing the judge's evaluation of the facts is based upon much more solid grounds than professional courtesy. It is because specific findings of fact, even by the most meticulous judge, are inherently an incomplete statement of the impression which was made upon him by the primary evidence. His expressed findings are always surrounded by a penumbra of imprecision as to emphasis, relative weight, minor qualification and nuance … of which time and language do not permit exact expression, but which may play an important part in the judge's overall evaluation."
The trial judge's conclusion as to the reliability of oral evidence based on his or her impression of the witnesses, may not be failsafe but it may not be irrational to prefer it to a conclusion based on an assessment of the probabilities disclosed in the record of the trial.
Mr Diprose's claim in equity to set aside his gift of the Tranmere property to Ms Louth succeeded notwithstanding that King CJ, the trial judge, rejected a critical aspect of Mr Diprose's evidence. Important to King CJ's conclusion, that Mr Diprose was subject to Ms Louth's influence, was his impression of Mr Diprose as a "strange romantic character"31. In the Full Court, Matheson J in dissent, considered that he was in as good a position as King CJ to draw inferences from the undisputed facts and that King CJ had wrongly concluded that Mr Diprose had been emotionally dependent on Ms Louth32. If one puts aside King CJ's impression of Mr Diprose's strange romantic character, it is easy to see the force of Matheson J's assessment of the probabilities. Mr Diprose was a 48 year-old solicitor of some years' standing. Applying the "Ipp rule", King CJ's conclusion, that Mr Diprose's professional qualifications and experience counted for nothing when he made the gift33, was against the probabilities and for that reason indicative of error. King CJ's assessment of Mr Diprose' character would not constitute an adequate reason supporting acceptance of his conclusion, since to find that it was sufficient would be to restore the trial judge's "untrammelled power", which it is the purpose of the rule to remove.
Adoption of the "Ipp rule" would provide a stimulus to appellate activity. Whether that activity would result in superior decisions is another matter. Chief Justice King's estimate of Mr Diprose' character may have been wrong. However, it is not self-evident that Matheson J's assessment based on the probabilities, without the benefit of seeing Mr Diprose and Ms Louth, should be thought more likely to be right.
In the High Court in Louth v Diprose, Dawson, Gaudron and McHugh JJ observed in their joint reasons that King CJ's finding turned not so much on the assessment of credibility as on the assessment of character34. Their Honours said that it is precisely because different people may come to different conclusions as to character, credit and disputed matters of fact that findings as to those matters are entrusted to the trial judge in accordance with rules that guarantee a considerable measure of finality35. It is a statement that recognises the element of judgment that is inherent in much fact-finding.
Courts find historical fact by acceptance that a disputed event occurred if the occurrence of the event is more probable than not. In theory, it may be said that there is a correct answer to the question of whether a fact has been proved. Fact-finding, however, is not a science and in the resolution of conflicting evidence there may be scope for legitimate differences of view about what facts have been proved36. Findings that are substantially dependent upon the assessment of the credibility of the witnesses are no longer, if they ever were, immunised from appellate challenge37. Nonetheless, the restraint applied before overturning them has not been shown to be misplaced either by the results of psychological research or today's enhanced means of recording evidence. The measure of finality to which Dawson, Gaudron and McHugh JJ adverted is not inconsistent with doing justice to the parties.
The duration and cost of litigation were the drivers behind the Woolf reforms in England and Wales38. The need for certainty, reasonable expense and proportionality are said to have informed the introduction of the requirement of permission to appeal to the Court of Appeal39. The decision of the "appeal court", whether a circuit judge or a High Court judge, is in most cases now final40. It is no longer possible to pursue an appeal to the Court of Appeal because the appeal is "properly arguable" or has a "real prospect of success"41. Where permission to appeal is granted the court must make its own assessment of the inferences. However, where an inference involves an element of judgment, the court will not interfere unless it is satisfied that the trial judge's conclusion lay outside the bounds within which reasonable disagreement is possible42. A more demanding standard, akin to that adopted in the United States and Canada, applies to the determination of Scottish appeals43.
The Federal Rules of Civil Procedure, which govern appellate review of facts in federal courts in the United States, provide that findings of fact, whether based on oral or other evidence, must not be set aside unless clearly erroneous44. A finding is "clearly erroneous" when, although there is evidence to support it, the reviewing court on the entire evidence is left with the definite and firm conviction that a mistake has been committed45. The Supreme Court of the United States has rejected the division of facts into categories and, in particular, the division of findings into those dealing with "ultimate" as distinct from "subsidiary" facts46. This reflects the text of the Rule and is not a rejection of the soundness of the distinction47.
The stringency of the Rule is illustrated by the statement of the Supreme Court of the United States in Anderson v City of Bessemer City, NC48:
"If the district court's account of the evidence is plausible in light of the record viewed in its entirety, the court of appeals may not reverse it even though convinced that had it been sitting as the trier of fact, it would have weighed the evidence differently."
White J, delivering the opinion of the Court, explained that the rationale for restraint is not limited to the trial judge's superior position in the determination of credibility. His Honour said49:
"The trial judge's major role is the determination of fact, and with experience in fulfilling that role comes expertise. Duplication of the trial judge's efforts in the court of appeals would very likely contribute only negligibly to the accuracy of fact determination at a huge cost in diversion of judicial resources. In addition, the parties to a case on appeal have already been forced to concentrate their energies and resources on persuading the trial judge that their account of the facts is the correct one: requiring them to persuade three more judges at the appellate level is requiring too much."
"The trial judge has sat through the entire case and his ultimate judgment reflects this total familiarity with the evidence. The insight gained by the trial judge who has lived with the case for several days, weeks or even months may be far deeper than that of the Court of Appeal whose view of the case is much more limited and narrow, often being shaped and distorted by the various orders or rulings being challenged."
The standard of "palpable and overriding error" is applied to appellate review of fact in Canada51. It is a standard that applies to all findings regardless of whether the finding depends upon the assessment of credibility, whether it is of primary or inferred fact or a global assessment of the evidence52. A conclusion that the judgment below contains "palpable and overriding error" it would seem might equally be expressed by a finding that it is "clearly wrong"53. Either formulation expresses the same idea, which is that the appellate court will not interfere with the trial judge's factual findings unless it can plainly identify the imputed error and that error is shown to have affected the result54.
In the leading Canadian decision on the topic, HL v Attorney General of Canada, Fish J, giving the majority reasons, cited with approval Professor Zuckerman's summary of the principles55:
"[I]f the appeal court cannot conclude that the lower court's inference from the primary facts was wrong, in the sense that it fell outside the range of inferences that a reasonable court could make, the appeal court should allow the lower court's decision to stand. The nature of the appellate evaluation of the lower court's decision will vary in accordance with the type of judgment that the lower court was called upon to make. But whatever the nature of the issues and however wide or narrow is the room for disagreement, the test remains the same: was the lower court's decision wrong. …
A decision will be wrong if … it was based on a plainly erroneous factual conclusion. … Put another way, as long as the lower court's conclusions represent a reasonable inference from the facts, the appeal court must not interfere with its decision."
The Canadian approach treating all findings of fact as subject to the same degree of restraint is one justified by finality expressed more particularly as the need to limit the cost of litigation and to value the autonomy of the trial process56.
The Canadian and American standards of fact review are reminiscent of the standard proposed by Barwick CJ and Windeyer J in decisions that culminated in Edwards v Noble57.In short, it was Barwick CJ's view that, even in cases in which the trial judge's finding did not depend upon the credibility of witnesses, that finding should only be disturbed if the appellate court was satisfied that it was wrong: even if the appellate court would have drawn a different inference, were it trying the matter itself, it should not overturn the inference drawn by the trial judge absent clear error58. In Da Costa v Cockburn Salvage and Trading Pty Ltd, Windeyer J proposed that the decision of the trial judge on the question of negligence should be treated by the appellate court as the equivalent of the verdict of the jury59. These views were controversial at the time. Hutley JA in an article published in the Sydney Law Review did not take a backward step: the Barwick/Windeyer test for appellate review was an aberration60.