//
© 2026 Zoe. All rights reserved.
Zoe is a legal information platform. Always consult the Federal Register of Legislation for authoritative text.
[1986] HCA 54
appellant. Appeal allowed with costs. Order that the appeal to the Full Court of the Supreme Court of South Australia be allowed with costs. Set aside the order of the trial judge and in lieu thereof order...
No linked legislation citations have been extracted yet.
Mr Stead was injured in a car accident and said it triggered a serious psychological condition that stopped him working entirely. The insurer said his new marriage stresses were the real cause. At trial, after hearing all the doctors including the insurer's psychiatrist Dr Scanlon, the judge told the insurer's lawyer he thought the accident was the trigger. When Mr Stead's lawyer started to criticise Dr Scanlon's evidence the judge said he did not accept Dr Scanlon so the lawyer stopped arguing the point. Months later the written judgment accepted Dr Scanlon completely, rejected the psychological claim and awarded only modest damages for the neck injury. The High Court held this was unfair because the judge had shut down argument on the central factual dispute and then decided the opposite way. Because the evidence was evenly balanced and the trial judge had seen the witnesses, the appeal judges could not safely say that full argument would definitely have changed nothing. The case was sent back for a fresh hearing on damages only. The decision underscores that procedural fairness is not a technicality; if a party is deprived of a real chance to persuade the judge on a key fact, a new trial will usually follow unless the court can be certain the result would have been identical.
AI-generated legal information, not legal advice. Zoe can make mistakes. Check the cited source, and for advice about your situation consult a qualified Australian lawyer.
2,462 words · generated 24/04/2026
The appellant, Mr Stead, was involved in a motor vehicle accident on 9 June 1981. Liability was admitted by the respondent, the State Government Insurance Commission. The trial therefore concerned only the quantum of damages. Mr Stead had sustained a physical neck injury for which the trial judge awarded $10,000 for pain and suffering and loss of amenity, together with out-of-pocket expenses and interest, producing a total judgment of $11,300.
Mr Stead sought very substantial additional damages on the basis that the accident had triggered a severe neurotic condition. This condition led to a complete psychological and physical breakdown, beliefs that he suffered from muscular dystrophy and heart disease, deterioration of his marriage (he had married his third wife only one month before the accident), and total incapacity for work from mid-1982. The respondent contended that the neurotic condition was not caused or materially contributed to by the accident but resulted from stresses associated with the third marriage.
A large body of evidence was called on the causation issue. Mr Stead, his wife and father gave evidence. He called a fellow worker, his general practitioner, a psychiatrist, a dentist and a physiotherapist, and tendered reports from a surgeon, neurosurgeon, physician/rheumatologist and another psychiatrist. The respondent called two psychiatrists, one of whom was Dr Scanlon, and tendered reports from a neurosurgeon and two orthopaedic surgeons. The critical conflict lay between the psychiatric opinions. Dr Scanlon gave evidence that there was “no relation between the subject accident and his behaviour” and that the accident had not operated as a “trigger”.
After the close of evidence, counsel for the respondent, Miss Mack, addressed and relied heavily on Dr Scanlon. In the course of discussion with the primary judge about whether the accident could have operated as a trigger, the judge indicated a preliminary view that, on the balance of probabilities, the accident had been the trigger. He invited counsel to address on the alternative basis of the strength of any risk that the same result would have occurred anyway. When counsel for the appellant began his address and attacked Dr Scanlon’s evidence, the judge immediately stated: “Alright. I don’t accept Dr Scanlon on that. You needn’t go on as to that.” Understandably, appellant’s counsel did not develop the submission further.
In a reserved judgment the primary judge accepted Dr Scanlon’s evidence in full, found that the neck pains from the accident were not a material cause of the subsequent breakdown, and held on the balance of probabilities that the third marriage stresses were the cause. He expressly rejected the proposition that the accident caused or materially contributed to the decompensation.
Mr Stead appealed to the Full Court of the Supreme Court of South Australia. One ground was that the primary judge had deprived him of the opportunity to present argument on the vital issue of Dr Scanlon’s evidence and therefore denied him natural justice. The Full Court dismissed the appeal. Bollen J (King CJ and Prior J agreeing) held that the judge had been “perfectly entitled to accept the evidence” of Dr Scanlon. He framed the practical question as whether further argument “possibly” could have made any difference and answered it in the negative, noting that the arguments advanced on appeal must have been considered by the trial judge and that the trial judge had had the advantage of seeing the witnesses. The High Court granted special leave and heard the appeal.
The High Court (Mason, Wilson, Brennan, Deane and Dawson JJ) began from the fundamental proposition, drawn from the English Court of Appeal’s decision in Jones v National Coal Board [1957] 2 QB 55 at 67, that every litigant is entitled to a fair trial at which he can put his case properly before the judge. No cause is lost until the judge has found it so after a fair trial. That principle is, however, qualified by the rule that an appellate court will not order a new trial if it would inevitably produce the same result; such an order would be futile.
The Court emphasised that the qualification is applied with great caution when the denied opportunity concerns submissions on an issue of fact, especially one turning on the acceptance or rejection of a particular witness’s testimony. In such cases it is “no easy task” for an appellate court, which has not seen the witnesses, to conclude that further submissions could have made no possible difference. The difficulty is magnified when the trial judge has enjoyed the “inestimable advantage” of seeing and hearing the witnesses. The Court noted that the primary judge’s own change of mind—from an initial disposition to reject Dr Scanlon to ultimate acceptance of his evidence—showed the issue was finely balanced. Bollen J’s own description of the appellant’s arguments as “forcefully and attractively presented” and “fit for offering to the trial judge but not on appeal” implicitly recognised that the matter was one for the trial judge’s evaluation.
The High Court held that the Full Court had asked itself the wrong question. It had effectively asked whether further argument would probably have made no difference rather than whether a properly conducted trial could not possibly have produced a different result. Because the issue was one of fact involving competing expert opinions whose weight depended on credibility and demeanour, the Full Court was disabled from reaching a sound conclusion that the denial of the opportunity to address could have had no bearing on the outcome. The appeal was therefore allowed and a new trial ordered, limited (as the first trial had been) to the issue of damages.
Before Stead, the law on miscarriages of justice arising from procedural unfairness was informed by cases such as Balenzuela v De Gail (1959) 101 CLR 226 and Jones v National Coal Board. Balenzuela established that a new trial will be granted where evidence has been wrongly rejected unless the rejected evidence could not have affected the verdict. The High Court in Stead expressly stated that nothing in Balenzuela compelled a different conclusion and that the earlier decision in fact reinforced the approach taken. The English Court of Appeal’s statement in Jones supplied the basal principle of entitlement to a fair trial.
The law before Stead therefore already contained the idea that not every procedural irregularity produces a new trial; futility was a recognised limitation. However, the precise calibration of that limitation when submissions on a factual issue had been curtailed was not as clearly articulated. Stead clarified that, on factual questions involving witness assessment, an appellate court must be affirmatively satisfied that the result could not possibly have been different. It is not enough to conclude that the trial judge was entitled to reach the view he did or that the result would probably have been the same. The decision also underlined the special difficulty faced by appellate courts when they lack the trial judge’s advantage in evaluating oral testimony.
After Stead the law is that a denial of the opportunity to be heard on a live factual issue will ordinarily result in a new trial unless the appellate court can exclude, to the point of certainty, the possibility that submissions could have altered the outcome. The case has become the authoritative statement of the test to be applied where a judge forecloses argument on a critical factual matter and then decides that matter against the party who was interrupted. The limitation to damages-only retrials where liability or other discrete issues have been fairly determined is also now standard.
The judgment contains several passages that have become canonical. The first is the quotation from Jones v National Coal Board:
“There is one thing to which everyone in this country is entitled, and that is a fair trial at which he can put his case properly before the judge. No cause is lost until the judge has found it so; and he cannot find it without a fair trial, nor can we affirm it.”
In plain English: you do not lose your case until a judge, after listening to everything both sides properly want to say, decides against you. An unfair hearing is not a real hearing at all.
Later the Court stated:
“Where, however, the denial of natural justice affects the entitlement of a party to make submissions on an issue of fact, especially when the issue is whether the evidence of a particular witness should be accepted, it is more difficult for a court of appeal to conclude that compliance with the requirements of natural justice could have made no difference.”
Translation: when the dispute is about whose evidence to believe, appeal judges should be very reluctant to say that letting the lawyer speak would have changed nothing. They were not in the courtroom and cannot easily second-guess the dynamics of live testimony.
The Court’s statement of the operative test is equally important:
“All that the appellant needed to show was that the denial of natural justice deprived him of the possibility of a successful outcome. In order to negate that possibility, it was … necessary for the Full Court to find that a properly conducted trial could not possibly have produced a different result.”
Plain English: the appellant wins the appeal if there was any realistic chance that full argument might have changed the judge’s mind. The respondent can defeat that only by proving that the same result was inevitable no matter what was said. That is a high hurdle.
Finally, the Court observed that Bollen J’s own language showed he regarded the issue as one for the trial judge:
“they were arguments fit for offering to the trial judge but not on appeal. There was evidence which could have led to the rejection of Dr Scanlon’s opinion. But it was for the trial judge to say. He had the inestimable advantage of seeing and hearing the witnesses.”
Translation: the appeal judges themselves recognised that the arguments were strong enough that only the judge who saw the witnesses could properly weigh them. That recognition itself demonstrated why a new trial was required.
Stead is triggered whenever a judicial officer, during oral argument, indicates a preliminary view on a contested factual issue, cuts off counsel who wishes to address that issue, and then later decides the issue against the interrupted party on a basis that counsel was prevented from fully contesting. The paradigm case involves expert evidence whose acceptance or rejection depends on demeanour, consistency, or impression formed in the witness box. It is especially apt where the judge’s ultimate conclusion differs from the impression conveyed during argument, as occurred when the primary judge moved from rejecting Dr Scanlon to accepting him.
The precedent applies with full force where the factual issue is “finely balanced” or where the appellate court would have to speculate about the weight a trial judge might have given to particular submissions had they been fully developed. It does not apply, or applies only weakly, where the denied opportunity relates solely to a pure question of law that the appellate court can see must be answered against the aggrieved party. It is also less likely to result in a new trial where the procedural unfairness is minor or where the party had a reasonable opportunity to address the point by other means (for example, in written submissions tendered before judgment).
Because the remedy is ordinarily a new trial limited to the affected issue, the precedent is most commonly engaged in damages assessments or causation disputes after liability has been resolved. It is not limited to psychiatric or medical evidence; any factual contest turning on oral testimony or competing expert reports can engage the principle.
The judgment itself demonstrates how Stead treats its own predecessors. It applies the Jones v National Coal Board principle as the starting point and treats Balenzuela v De Gail as reinforcing rather than contradicting the approach. The Court carefully distinguishes the situation in which evidence is wrongly rejected (where the test is whether the evidence could have affected the verdict) from the situation in which submissions are wrongly curtailed, yet insists that the same underlying concern for the possibility of a different outcome governs both.
The decision insists on strict scrutiny of any claim that a procedural breach was harmless. It rejects the Full Court’s looser formulation (“would further argument possibly have made any difference?”) in favour of the stricter requirement that the appellate court must be satisfied that the result could not possibly have been different. That sharpening of the test has become the authoritative formulation. The judgment also makes clear that an appellate court’s acknowledgment that it lacks the trial judge’s advantage in assessing witnesses disables it from confidently declaring a breach harmless on factual matters.
The Court’s emphasis that the issue was “pre-eminently suitable for determination by the primary judge” has become a standard caution against appellate over-confidence in witness-based cases. The limitation of the retrial to the issue of damages illustrates the Court’s willingness to mould relief so that only the tainted part of the proceeding is reopened, thereby preserving judicial resources where possible.
The judgment leaves open precisely how an appellate court is to determine that a different result was “impossible” without effectively retrying the case on the papers. In borderline cases the distinction between “could not possibly” and “would probably not” may still prove elusive. The decision does not explore the position where the interrupted submissions would have been repetitive of matters already fully canvassed in earlier stages or in written outlines. Nor does it address the situation in which the judge’s indication during argument is itself ambiguous or where counsel fails to seek clarification or to reserve the right to make further submissions.
Questions also remain about the threshold at which an issue is sufficiently “factual” to attract the heightened caution. Mixed questions of fact and law, or cases where the factual finding is largely dictated by legal principle, are not squarely covered. The judgment assumes the trial was otherwise fair; it does not address cumulative procedural irregularities or the interaction between curtailed oral argument and restricted cross-examination.
Finally, the decision leaves unresolved the precise content of the opportunity to be heard that must be afforded. Must counsel be allowed to address for a minimum time, or is it enough that counsel had a reasonable opportunity that was not taken? These matters continue to require careful case-by-case analysis, but Stead supplies the governing framework: the touchstone is whether the party was deprived of the possibility of a successful outcome on the affected issue.
High Court of Australia Mason, Wilson, Brennan, Deane and Dawson JJ. Stead v State Government Insurance Commission [1986] HCA 54
ORDER Appeal allowed with costs. Order that the appeal to the Full Court of the Supreme Court of South Australia be allowed with costs. Set aside the order of the trial judge and in lieu thereof order that there be new trial limited to the issue of damages.
The Court delivered the following written judgment: -
Sept. 23 Mason, Wilson, Brennan, Deane and Dawson JJ.
In an action in negligence arising out of a motor accident on 9 June 1981, in which liability was admitted, judgment was given for the appellant in the sum of $11,300, consisting of $10,000 for pain and suffering and loss of amenity of life, $1,000 for out-of-pocket expenses and $300 for interest. The amount of $10,000 was awarded in respect of a physical injury sustained by the appellant to his neck or cervical spine.
The appellant had sought additional damages in respect of a neurotic condition from which the primary judge found that he was suffering. The additional damages sought were substantial because the condition led to a complete breakdown and rendered the appellant totally incapacitated for work. The primary judge found that the respondent's negligence did not cause or materially contribute to the condition.
Whether the accident which resulted from the respondent's negligence was the cause or a material factor contributing to the condition was a major issue at the trial. Before the accident the appellant had always been a regular and industrious worker. Some time shortly after the accident he began to experience pain and discomfort in the region of his neck and cramping pains all over his body during sexual intercourse. This lead to his consulting his doctor about the accident for the first time thirteen months after the accident and to a deterioration in what had been a loving relationship with his third wife whom he had married one month before the accident. He came to believe that he was suffering from muscular dystrophy and, later, heart disease. He virtually ceased to work from the time when he first consulted his doctor, Dr. Hardy, in July 1982.
The respondent's case on the issue of causation was that the cause of the condition was not the accident, but the stresses and strains arising from the appellant's third marriage. A substantial amount of evidence was tendered on the issue. The appellant gave evidence, as did his wife and father. He called a fellow worker, his doctor, a psychiatrist, a dentist, and a physiotherapist. He tendered reports by a surgeon, a neurosurgeon, a physician and rheumatologist and another psychiatrist. The respondent called two psychiatrists, one of whom was Dr. Scanlon, and tendered the reports of a neurosurgeon and two orthopaedic surgeons.
After the close of the evidence the respondent's counsel, Miss Mack, addressed. She relied strongly on Dr. Scanlon who had given cogent evidence against the existence of a causal connexion between the accident and the appellant's neurotic condition. Following some discussion between Miss Mack and the primary judge concerning Dr. Scanlon's evidence, particularly with respect to the accident operating as a "trigger" to set off the appellant's neurotic condition, Miss Mack submitted that the psychiatrists were in harmony as to the appellant's unusual personality make-up. The learned judge then said:
His Honour: Yes, the main difference is the triggering or non-triggering. I think as presently advised and having heard you on that, and very recently heard Dr. Scanlon and from my legal position I am not moved so far, from my present view that on the balance of probabilities it was the trigger You should argue on the second basis, as to how strong the risk was of the contingency of the same result, what effect it should have.
When the appellant's counsel came to address, he submitted that his Honour should not accept Dr. Scanlon's evidence that the accident had nothing to do with the neurotic condition. This elicited from his Honour the comment, "Alright. I don't accept Dr. Scanlon on that. You needn't go on as to that." The appellant's counsel understandably did not develop the submission and addressed no further argument on the acceptability of Dr. Scanlon's evidence on the issue of causation.
His Honour, having reserved his decision, ultimately accepted Dr. Scanlon's evidence on that issue, holding that the neck pains originating in the accident were not a material cause of the cramps and of the appellant's subsequent psychological and physical breakdown. His Honour specifically accepted Dr. Scanlon's testimony that there was "no relation between the subject accident and his behaviour" and that the accident did not operate as a "trigger". His Honour went on to find:
on the balance of probabilities, that the third marriage stresses were the cause of the plaintiff's decompensation in late 1981 and following years. The lower position I take is that the plaintiff has not proved, on the balance of probabilities, that the accident caused his decompensation or materially contributed to it.
The appellant appealed to the Full Court of the Supreme Court of South Australia on various grounds. One ground was that by stopping the appellant's counsel from addressing on the topic of Dr. Scanlon's evidence with respect to causation, the primary judge had deprived the appellant of an opportunity of presenting argument on a vital issue in the case. The Full Court dismissed the appeal, holding that the primary judge was "perfectly entitled to accept the evidence" of Dr. Scanlon. Bollen J. (with whom King C.J. and Prior J. agreed) said that "the inquiry whether a miscarriage of justice has occurred involves asking whether the event said to cause a miscarriage could have made any difference to the result". To support this proposition, his Honour called in aid decisions, such as Balenzuela v. De Gail [1] , which established that a new trial will be granted on the ground that evidence has been wrongly rejected unless it appears that the evidence rejected could not have affected the jury's verdict. His Honour, referring to the decisions, said:
They emphasize that we must inquire if there has been a miscarriage of justice. I think that they show that the practical question is - would further argument possibly have made any difference? He answered this question in the negative on the footing that the arguments advanced for the appellant in the Full Court must have been considered by the primary judge. His Honour concluded by saying: In my opinion, therefore, the further arguments and elaboration of argument which the appellant's counsel would have put could not have made any difference to the result. 1. (1959) 101 C.L.R. 226.
The general principle applicable in the present circumstances was well expressed by the English Court of Appeal (Denning, Romer and Parker L.JJ.) in Jones v. National Coal Board [2] , in these terms:
There is one thing to which everyone in this country is entitled, and that is a fair trial at which he can put his case properly before the judge. No cause is lost until the judge has found it so; and he cannot find it without a fair trial, nor can we affirm it. That general principle is, however, subject to an important qualification which Bollen J. plainly had in mind in identifying the practical question as being: Would further information possibly have made any difference? That qualification is that an appellate court will not order a new trial if it would inevitably result in the making of the same order as that made by the primary judge at the first trial. An order for a new trial in such a case would be a futility. 1. [1957] 2 Q.B. 55, at p. 67.
For this reason not every departure from the rules of natural justice at a trial will entitle the aggrieved party to a new trial. By way of illustration, if all that happened at a trial was that a party was denied the opportunity of making submissions on a question of law, when, in the opinion of the appellate court, the question of law must clearly be answered unfavourably to the aggrieved party, it would be futile to order a new trial.
Where, however, the denial of natural justice affects the entitlement of a party to make submissions on an issue of fact, especially when the issue is whether the evidence of a particular witness should be accepted, it is more difficult for a court of appeal to conclude that compliance with the requirements of natural justice could have made no difference. True it is that an appeal to the Full Court from a judgment or order of a judge is by way of rehearing and that on hearing such an appeal the Full Court has all the powers and duties of the primary judge, including the power to draw inferences of fact: Supreme Court Rules, O. 58, rr. 6 and 14. However, when the Full Court is invited by a respondent to exercise these powers in order to arrive at a conclusion that a new trial, sought to remedy a denial of natural justice relevant to a finding of fact, could make no difference to the result already reached, it should proceed with caution. It is no easy task for a court of appeal to satisfy itself that what appears on its face to have been a denial of natural justice could have had no bearing on the outcome of the trial of an issue of fact. And this difficulty is magnified when the issue concerns the acceptance or rejection of the testimony of a witness at the trial.
This is just such a case. At the trial the critical question on the issue of causation was whether Dr. Scanlon's testimony should have been accepted in preference to the appellant's expert witnesses, notably Dr. Donsworth. It was an issue pre-eminently suitable for determination by the primary judge who had an advantage over the Full Court in seeing and assessing the witnesses. We do not see how the Full Court, denied the important advantage of seeing and assessing the witnesses, could satisfactorily conclude that had the appellant's counsel been given a reasonable opportunity to present submissions on the issue, it could have made no possible difference to the result.
It is significant that Bollen J., after referring to "the forcefully and attractively presented arguments" of Mr Anderson for the appellant, made this comment:
they were arguments fit for offering to the trial judge but not on appeal. There was evidence which could have led to the rejection of Dr. Scanlon's opinion. But it was for the trial judge to say. He had the inestimable advantage of seeing and hearing the witnesses. Later Bollen J. said: In my opinion, the learned trial judge was perfectly entitled to accept the evidence of Dr. Scanlon. It was his evidence which the learned trial judge found acceptable. It was entirely open for him to so find.
These statements do not suggest that his Honour considered that the primary judge was bound to find the issue of fact in favour of the respondent or that the finding of fact was the only finding reasonably open on the evidence. Instead they imply that the issue was rather finely balanced, an assessment which accords with the primary judge's reaction. Initially he had been disposed to reject Dr. Scanlon, but on mature reflection he had come to the opposite conclusion.
It is natural that Bollen J. expressed himself as he did in the passages which we have quoted. He was conscious that, not having seen the witnesses, he could not evaluate their evidence in the way in which a trial judge can. It is for this very reason that, in our view, the Full Court was disabled in the circumstances of this case from reaching a sound conclusion that a new trial in which the applicant's counsel would have an adequate opportunity of presenting submissions on the issue of causation could make no difference to the result.
Alternatively, if the Full Court is properly to be understood as saying no more than that a new trial would probably make no difference to the result, their Honours failed to apply the correct criterion. All that the appellant needed to show was that the denial of natural justice deprived him of the possibility of a successful outcome. In order to negate that possibility, it was, as we have said, necessary for the Full Court to find that a properly conducted trial could not possibly have produced a different result.
There is nothing in Balenzuela v. De Gail [3] which compels us to a different conclusion. Indeed there is much in that case that reinforces what we have said: see the judgment of Dixon C.J. [4] . There the Court ordered a new trial because material evidence was wrongly rejected. It would have been otherwise had the respondent been able to demonstrate that the rejected evidence could have made no difference to the result. Then it would have emerged that the appellant had not lost a possible chance of obtaining a verdict. However, that was something which the respondent in Balenzuela v. De Gail was unable to demonstrate. So it is here.
Accordingly, the Full Court's refusal to order a new trial was, in the circumstances, mistaken.
The appeal should be allowed, and a new trial ordered limited, as was the first trial, to the issue of damages.
Stead
State Government Insurance Commission
Appeal allowed with costs. Order that the appeal to the Full Court of the Supreme Court of South Australia be allowed with costs. Set aside the order of the trial judge and in lieu thereof order that there be a new trial limited to the issue of damages.