Whisprun Pty Ltd v Dixon

GLEESON CJ, McHUGH AND GUMMOW JJ. This case is concerned with facts and the forensic tactics of the parties at trial and an appellate court's perception of them. Special leave to appeal was granted to determine whether a miscarriage of justice had occurred in the particular circumstances of the case and not because the case concerned any important principle of law.

The trial judge had dismissed the respondent's action for damages on the ground that her credibility was such that he was not satisfied that she suffered from chronic fatigue syndrome, the condition from which she claimed to suffer. The Court of Appeal of New South Wales set aside the judgment because it appeared to that Court that the trial judge had not satisfactorily considered the respondent's case. So the question for this Court is whether the Court of Appeal erred in holding that the trial judge had not properly considered the respondent's case. That, of course, depends on what the respondent's case was at the trial, not on a hypothetical construction of what could or should have been the respondent's case.

Despite its lack of legal significance, the appeal is a fascinating one. It illustrates how an appellate court can perceive a case in a way dramatically different from the case that was run at the trial. And it demonstrates how, when it is so perceived, the trial judge's reasons may not deal adequately with the reconstructed case and appear to have done less than justice to the losing party's case.

In arguing the present respondent's case in the Court of Appeal, Mr D F Jackson QC - who had not appeared at the trial - contended that a substantial miscarriage of justice would have occurred if "the evidence was there" and the judge in writing his judgment "overlooked relevant parts of it". In substance, the Court of Appeal adopted this submission. But the evidence that that Court saw as "overlooked" was different from the evidence that Mr Jackson argued had been "overlooked" by the trial judge. He was speaking of evidence concerning infection . The Court of Appeal's reasons deal with "evidence" concerning chronic fatigue syndrome . At the trial, the parties treated this "overlooked" evidence concerning chronic fatigue syndrome as having no relevant evidentiary value unless the trial judge substantially accepted the respondent's evidence as to her condition. Because that is so, the trial judge did not overlook "relevant parts" of the evidence, and there was no miscarriage of justice. Moreover, even if the "overlooked" evidence had relevant evidentiary value independently of the respondent's credibility, we are not convinced that it was sufficient to overcome the effect of the various matters that the trial judge used to make his finding concerning her credibility. Accordingly, this appeal must be allowed. Statement of the case

Sonya Lea Dixon, the respondent, sued Whisprun Pty Ltd ("Whisprun") for damages in the Supreme Court of New South Wales claiming that it had breached the duty of care that it owed to her as her employer. She claimed that this breach had caused her to contract Q fever infection and that she now suffered from chronic fatigue syndrome as a consequence of the infection. Whisprun admitted that it had breached the duty of care owed to her and conceded that she had contracted Q fever infection. But it denied that she suffered from chronic fatigue syndrome. The case was tried by Newman J without a jury. His Honour rejected Ms Dixon's claim of chronic fatigue syndrome with the result that the damages he awarded to her for non-economic loss were lower than the minimum set by s 151G of the Workers Compensation Act 1987 (NSW). Because that was so, he was required to enter judgment in the action for Whisprun.

The Court of Appeal (Beazley and Heydon JJA and Davies AJA) allowed an appeal by Ms Dixon against that judgment. Acting under the power conferred by Pt 51 r 23 of the Supreme Court Rules 1970 (NSW), it ordered a new trial of the action. It did so on the ground that "it appears that some substantial wrong or miscarriage [of justice] has taken place in that the plaintiff's claim has not been properly considered." The Court of Appeal held that the trial judge had not given any weight to material in medical reports that were tendered in evidence. Issues

The appeal gives rise to a number of related issues: (1) Did the parties conduct the trial on the basis that, so far as chronic fatigue syndrome was concerned, the evidentiary value of the medical reports depended entirely on the judge accepting Ms Dixon's evidence? In other words, did the medical reports have evidentiary value on the chronic fatigue syndrome issue irrespective of what view the judge took of Ms Dixon's credibility? (2) If the parties conducted the trial on the basis that the truth or validity of the statements and opinions in the medical reports depended on the trial judge accepting Ms Dixon's evidence, did the judge err in not considering whether the material in the medical reports assisted Ms Dixon's case? (3) If the reports had independent evidentiary value, did they contain evidence of "objective symptoms" that should or could have induced the trial judge to find that Ms Dixon suffered from chronic fatigue syndrome? The material facts

In April 1994, Ms Dixon contracted Q fever in the course of her employment with Whisprun at an abattoir at Inverell in New South Wales. By July 1994, symptoms of the fever had developed and she was diagnosed in that month as having Q fever after undergoing pathological tests. Q fever is an infectious disease that is prevalent in animals and can be transmitted to humans. Its symptoms are similar to influenza. Infection usually has an acute phase that can evolve into a chronic infection that may be demonstrated serologically, that is, by blood tests. There is a 10-20 percent chance that a sufferer from Q fever may also develop chronic fatigue syndrome. The distinction between chronic Q fever infection and chronic fatigue syndrome that may exist after chronic Q fever infection no longer exists is of great importance in this case. Whether or not chronic fatigue syndrome exists cannot be proved or disproved by an objective test. A diagnosis of the syndrome must rely on statements by the patient as to subjective symptoms.

Serological tests carried out in February 1995 indicated that, although Ms Dixon had had an acute episode of Q fever, she did not have chronic Q fever infection at that time. Nevertheless, she claimed that, even if the acute period of infection had ended by February 1995, she suffered chronic fatigue syndrome, as a consequence of the infection, for an extended period (at least up until the trial in August 2000).

Between 1994 and 2000, Ms Dixon saw many medical practitioners. She gave a consistent history of suffering from headaches, aches and pains, nausea and fatigue. The approach of the trial judge

The trial judge accepted Whisprun's contention that the initial Q fever infection did not cause any symptoms after February 1995. His Honour held that Ms Dixon had recovered from Q fever by February 1995. He was not satisfied that thereafter she suffered from chronic fatigue syndrome as the result of her Q fever infection. Newman J said: "When examined in chief, the plaintiff gave evidence which was consistent with the history of continuing symptoms which she had given to medical practitioners. Because the continuing symptoms are entirely subjective the only issue which arose in the case was whether the plaintiff's account of her continuing symptoms was credible. In other words has the plaintiff established on [the] balance of probabilities that she has had the symptoms of which she has complained?"

In his judgment, Newman J set out many inconsistencies between Ms Dixon's account of what she could and did do and what other evidence showed that she did do. This evidence showed her walking briskly around Sydney, attending race meetings, watching her daughter ride horses, riding a horse herself, driving cars and horse floats, riding on a jet ski and drinking and dancing at a wedding. The learned trial judge also took into account Ms Dixon's evidence that her relationship with Mr Cross, the father of her daughter, had broken up owing to her mood swings and that she told some of the doctors they were still together. His Honour concluded that "the matters raised in cross-examination with the plaintiff and her responses to them effectively destroyed her credibility." Because the existence of her symptoms depended on Ms Dixon's credibility, his Honour was not satisfied that she suffered from post Q fever chronic fatigue syndrome.

Having read the whole of Ms Dixon's evidence, we are not surprised that his Honour made the findings that he did in respect of her credibility. Indeed, as is so often the case, his Honour's summary of her evidence and his findings do not convey the full picture of her lack of credibility, a picture that can only be obtained from reading the whole of her evidence. By the time her lengthy cross-examination had finished, the picture of her state of health to which she had deposed in evidence-in-chief had been destroyed. She may well be sick - but it is certain that her illness does not affect her to the extent that she claimed in evidence and recited to doctors. Later, it will be necessary to return in detail to this evidence. The Court of Appeal

The Court of Appeal held that Newman J had not properly considered Ms Dixon's case. The Court's reasons were prepared by Heydon JA, Beazley JA and Davies AJA simply agreeing with his judgment. The Court of Appeal held that Newman J had erred in: . concluding that Ms Dixon's condition rested on subjective symptoms; . failing to make any findings about the existence of symptoms which were observable by the medical experts for themselves, and which they had taken into account; . failing to assess the totality of the medical evidence in light of the symptoms observed by medical experts which did not depend on Ms Dixon's own history, and in the light of their professional experience and assessment of the probability that she was suffering from Q fever chronic fatigue syndrome.

In particular, the Court of Appeal held that the trial judge had erred in failing to consider whether Ms Dixon had symptoms that were observable to medical practitioners. If such symptoms existed, then her case did not depend entirely on her credibility. The Court also held that, in assessing Ms Dixon's credibility, Newman J had placed undue weight on trivial discrepancies and had not taken account of the fact that her condition might have impaired her memory.

The Court of Appeal said that the "essence of the trial judge's reasoning was to conclude that once the plaintiff's credit and her reliability was damaged in the ways he set out, no further inquiry was called for." He "made no reference to the large quantity of medical evidence, most of which pointed in one direction." The Court said that, having found the witness unsatisfactory in various respects that could be checked, Newman J inferred that she was unsatisfactory in all respects. Objective symptoms

In the forefront of the Court of Appeal's reasons for ordering a new trial was the view that the trial judge had erroneously thought that the existence of the symptoms of which Ms Dixon complained depended on her evidence. Newman J had relied on a statement by Dr McGuirk that "there is no way any objective test could either substantiate or disprove" whether Ms Dixon had the syndrome. But the Court of Appeal said that Dr McGuirk was not suggesting that there were no objective symptoms , and that there was objective evidence that the medical experts had taken into account in forming their opinions that Ms Dixon had the syndrome. The objective symptoms included weight loss, abdominal tenderness, pallor and depression and memory impairment. In addition, the trial judge had not referred to the opinions of various doctors that she had chronic fatigue syndrome, the statistical probability of her having the syndrome or considered an explanation for her lack of credibility. The conduct of the trial

The trial commenced on circuit at Newcastle on 29 August 2000 and was adjourned on 30 August. It resumed at Sydney on 3 October 2000. Ms Dixon's evidence-in-chief and much of her cross-examination was heard in Newcastle. The plaintiff and the defendant each was represented by senior counsel whose long and extensive experience in the conduct of trials of actions for damages for personal injury was reflected in an appreciation that the best advocacy is selective and economical. Given the medical evidence concerning Q fever infection and chronic fatigue syndrome arising from that infection, it is not surprising that counsel for Ms Dixon saw the real issue at the trial as whether Ms Dixon was credible when she claimed to be suffering from symptoms that were consistent with that syndrome. Given the mass of surveillance data concerning Ms Dixon, it is not surprising that Whisprun's counsel saw the issue of her credibility as its best chance of succeeding in the action.

Because of the way that the parties appear to have conducted their cases, unsurprisingly, Newman J said "the only issue which arose in the case was whether the plaintiff's account of her continuing symptoms was credible." (emphasis added) Because that was so, counsel for each side obviously took the view that going to the expense of calling medical witnesses was irrelevant. If Ms Dixon's evidence was substantially accepted, she suffered from chronic fatigue syndrome and was entitled to substantial damages. If her evidence was not credible, there was no halfway house that would enable the judge to find that she was sick, but only a little bit sick. Upon what evidentiary basis could the trial judge refuse to accept that she was suffering as she claimed and yet find that she suffered from a lesser condition that she did not identify? In some cases, a tribunal of fact may properly refuse to accept either party's case and work out for itself "a view of the case which did not exactly represent what either party said." This case did not fall into that category. Ms Dixon's case was conducted on the basis that what she said in her evidence-in-chief and what she had consistently told the doctors was true.

Moreover, behind the conduct of the case was the dark shadow cast by s 151G of the Workers Compensation Act . At the relevant time, that section provided that no damages for non-economic loss could be recovered unless Ms Dixon recovered more than a specified minimim amount for non-economic loss. This statutory threshold made it imperative from Ms Dixon's viewpoint that the trial judge accept that she had suffered and continued to suffer from a debilitating chronic fatigue syndrome. Ms Dixon's evidence

In her evidence-in-chief, Ms Dixon said that, after working in Sydney, she returned to Inverell where she worked at Coles stacking shelves and at the abattoirs as a casual worker packing meat. She had a relationship with Mr David Cross, became pregnant, and in 1992 had a daughter, Sarah. After the birth of her daughter, she worked at three jobs: at Coles, at the abattoir and at a fruit and vegetable shop. She and Mr Cross bought a house for $89,000 financed by a mortgage of about $80,000. At the abattoirs, she obtained a full-time job that involved sucking foetal blood into a pipette from unborn calves' hearts. It was doing this work that caused her to be infected with Q fever. She also worked up to four shifts each week at Coles from 3.30pm onwards to near midnight. She also worked on Saturday morning at the fruit and vegetable shop. She continued to work at these jobs until July 1994 when she became sick and was diagnosed as having Q fever. Before the infection, she had taken part in sporting activities - playing hockey, touch football and riding horses.

On contracting Q fever, she felt very sick and had headaches and nausea, similar to the flu. She went back to work after eight days but she felt very sick and had to get help at work. In September 1994, she gave a certificate of unfitness to her employer. Her employment was terminated on 21 September 1994. However, she later went back to work although she had difficulty working a full week and had time off because she was sick. She had nausea, vomiting, hot and cold sweats all the time and her head was "just aching". After a day's work, she felt terrible and "just went straight to bed." Her relationship with Mr Cross broke up because "of my being sick, he couldn't cope with my mood swings." She thought that occurred in 1995. She moved in with her parents; the house that she and Mr Cross had bought was sold. Later, she moved out of her parents' house into a shed about 200 metres away. The "shed" has a small kitchen, a bathroom, a laundry and toilet, a small lounge room and two bedrooms. Asked if the "shed" was "still on your parents' property", she replied, "Yep."

Since 1994, her condition "fluctuates up-and-down" but it was not as bad as when she first got Q fever. There has been only a little improvement and she has "some fluctuations from day to day and from week to week". She gets headaches a lot, probably three times a week. They could last a couple of hours or all day. When her headaches are really bad, she stays in bed. She gets pains in the joints in the knees, the back and her elbows. In the good weeks, headaches may occur "once or twice". She takes Panadeine Forte and Panadol. When she gets a really bad headache, she gets chest pains and hot and cold flushes and sweats a greal deal. The sweating leaves the bedclothes saturated and has a bad odour. She still vomits and has nausea. On some days, she does not vomit or have headaches but such days are rare. She gets very tired and loses energy. Although she tries, she does not always get the housework done. Even on her best days, she is not able to stand all day on her feet. She gets tired and runs out of energy and always sleeps for some period during the day for "an hour, maybe three hours." She sees Dr Thatcher once or twice a fortnight. He prescribes medication including anti-depressants that sometimes help her. Her appetite is not very good. She has not really had much social life in the last few years although she had been out to a dinner with her parents the previous June for her daughter's birthday. She doesn't feel like going out because afterwards she gets tired and it takes a couple of days for her to recover. In her mood swings, she is snappy, yells and screams and gets agitated. She also has difficulty in sleeping and has had problems with dizziness. She now has trouble with her memory.

Thus, in evidence-in-chief, Ms Dixon presented a picture of a woman who since July 1994 had been severely disabled, had had virtually no social life and had been forced to live with her daughter in a shed on her parents' property. She was unable to work in gainful employment, was in constant pain and discomfort, was regularly nauseated, was tired and lacked energy, was having difficulty in sleeping and was so moody that her unpleasantness had broken up her relationship with Mr Cross. This picture of her condition was entirely consistent with what she had told a number of doctors over several years.

The picture that emerged in cross-examination and from three video films and other evidence was different. Initially, when asked whether she owned any property at all, Ms Dixon said, "Yeah, I had. I had." This answer seemed to be a reference to the fact that she and Mr Cross had owned a house which had been sold. Asked whether she owned "any now", she replied, "Just a car and that, that's it." When asked whether she owned any real estate, she admitted that she owned land that she had purchased two years ago with $50,000 she had borrowed. There were four stables on this land and there were two horses there. Asked whether she owned any other land, she said that she did not. However, although the matter is not clear, further cross-examination appeared to reveal that this land contained the "shed" where she lived. Asked why she had not said in her evidence-in-chief that the shed was on her property, she said "I've got no idea." She denied that she did not want the Court to know that she owned property.

Ms Dixon said that the only time that she had been on a horse since leaving the abattoir was "when Sarah has had to go to the toilet or something like that, I have held the horse." She said that Sarah had asked her to show her how "I used to ride when I was younger, and I couldn't." Ms Dixon also said that she could not "remember when I last rode a horse." Pressed, she conceded that she had ridden a horse "this year" but could not remember how often. She said that she wasn't quite sure how many times. Cross-examined further, she said that it would rarely be the case that she would ride several times each week. She claimed that she rode only because her daughter hassled her.

Cross-examined as to her attendances at pony clubs where Sarah was riding, Ms Dixon said "mostly I sit" and that she could only stand for "maybe an hour". After seeing a video of Ms Dixon at an equestrian event, Newman J commented that she showed "no signs of fatigue on this occasion".

Ms Dixon said that she had to sleep every day and that her mother had to help her with housework because headaches forced her to lie down. But she admitted that she had been at a wedding from 3.00 pm until the bride and groom left. When it was put to her that she was there until 1.00 am the following morning, Ms Dixon said she was "not quite sure". She said that she had "a couple of dances" and "a couple of drinks". She admitted that a photograph of her with the bride showed that she did not look depressed and that she was not depressed that night.

Ms Dixon was shown a video of the wedding and other videos. After that, the case was adjourned. It did not resume for nearly five weeks. When the cross-examination resumed, Ms Dixon admitted that she was "not fatigued" at the wedding, that she was "dancing vigorously" and was "drinking beer out of a bottle". But in re-examination she sought to offer an explanation for her vibrant behaviour claiming that the bride had told her that she had put an amphetamine in her drink. When the bride, Michelle Mair, testified, she denied that she had done so or had a conversation to that effect. In his judgment, Newman J refers to this explanation and, although he did not expressly find that Ms Dixon had invented this explanation, his judgment makes it plain that he did not accept it or at all events was sceptical as to its truth. Newman J said that the video of the wedding "demonstrated her to be very active both in terms of dancing and socialising and drinking."

Ms Dixon was also cross-examined concerning her ability, while on a visit to see doctors, to walk from a hotel in George Street, Sydney to the Darling Harbour area. However, the cross-examination led nowhere when she said that she could not place where these areas were. She conceded that on one day - it was suggested to her the next day - after seeing a doctor she and her mother caught a taxi to a Wax Works and that they had gone into the Sydney Casino and an aquarium. Mr W K Dodd SC, counsel for Whisprun, suggested that these places were near Darling Harbour. When he suggested to Ms Dixon that she had been on her feet "several hours that day", she replied that she would not have been on her feet "all that day" and that she would have been "sitting and walking". After she had seen a video film of her walking on this visit to Sydney, she conceded that she was "able to walk briskly". When it was put to her that she had walked "three or more kilometres", she said, "I don't think it was that far." Newman J said that the video "showed her walking briskly an extensive distance between Railway Square, Darling Harbour and streets of Sydney." His Honour said that her "actions as depicted in that video were quite inconsistent with the picture she had painted of herself both in chief and in histories to doctors."

Asked whether she had gone to race meetings, Ms Dixon said she could not remember going to a race meeting since she left the abattoir. When shown newspaper photographs of herself at race meetings, she admitted that she had been at meetings at Inverell in October 1997, January 1998 and January 2000 with Mr Cross. She conceded that she was at the January 1998 meeting for four and a half hours but claimed that she would have been sitting down. After being shown a video of the January 1998 meeting, she admitted that she was "walking briskly" when walking around the course. She also agreed that on occasions she was standing while others were sitting, that she went to the bar area and purchased drinks and that she "jumped with exuberance" when she backed a winner. Asked whether she showed any tiredness in the video, she answered "I can't answer that question, I am sorry." Newman J noted in his judgment that Ms Dixon had conceded that she was present at this meeting "for no less that four hours and twenty minutes." His Honour said that the video demonstrated that Ms Dixon was "showing no signs whatsoever of fatigue and was demonstrative of the plaintiff enjoying herself at that meeting."

Ms Dixon was cross-examined as to whether she had been to Copeton Dam, a place used for watersports. Her initial answers indicated that she had been there only once at Christmas time with her parents and her daughter. She could not remember indulging in watersports like swimming "or other things". When asked whether she had been "towed behind a speedboat on an inner tube", her guarded answer was "I have been in Nick and Shelly's speedboat, when Sarah was on a tube." When it was put to her that she had been on the tube, she answered "[o]nly in the water, like, not towed." She also denied driving a jet ski. Pressed further about the tube, she said, "I sat on the tube but I couldn't handle it because I don't like swimming in water when I can't see." Asked whether she had been towed any distance, she said, "No, not that I can remember, because I don't like it. They were scaring me." Asked again whether she drove a jet ski, she said, "I got on, I said, with Shelly, and we only went from there to there and I said stop, I can't handle it."

Shelly was Michelle Mair. In evidence, Mrs Mair said that in November 1999 she and her husband and Ms Dixon, Mr Cross and Sarah went tubing which involved a boat towing one to three rubber tubes. She said that Ms Dixon was in a tube next to her and that they went for 20 minute runs on four or five occasions. A Mr Andrew Pelja gave evidence about Ms Dixon tubing and driving a jet ski. He said that in tubing "[i]t's up to the boat driver to put them into a whip to bounce over the wake and everything and have a lot of fun." He said that on one occasion - alleged to be the same one Mrs Mair spoke about - Ms Dixon "was on the tube for about 10 or 15 minutes." He was one of those driving the boat. Mr Pelja also testified that at Christmas 1999 Ms Dixon had driven a jet ski. She was "[j]ust riding around on it like we all do, just not as hard as we do." She had ridden as a pillion passenger at first "to show her the ropes and how to use the jet ski and then by herself." Newman J said that he "formed the view that Mr Pelja was a witness of truth." His Honour also said that "the evasive nature of the evidence she gave in relation to her activities with [the Mairs] at the Copeton Dam did not enhance her credibility."

Ms Dixon was also cross-examined about driving, shopping and doing other things in Inverell. She said that such trips would not "exceed more than four or five days" a week. She was taken through a week of car trips she had taken in August 2000. She agreed that on the Wednesday she had left home about 9.30 am and driven to a property where a woman was looking after one of her horses. She left the property at 10.40 am. The following day she left home about 12.25 pm and went into town to a bank where she used an ATM. She then went to Coles supermarket where she met her mother. They went to lunch at the Bridge Cafe. At 1.25 pm she left her mother and drove to a brake centre where she paid a bill. She then drove to a bowling club and paid her father's membership subscription. She then went to a convenience store before driving home. The following day she again drove into town, met her mother at Coles at 11.15 am and again went to lunch with her at the Bridge Cafe. When it was suggested to her that this was "hardly the lifestyle of someone who is stuck at home doing nothing all day", she replied "[w]e would have been doing things, and fixing up errands or something, that was it." After lunch, she and her mother walked to a bank and used an ATM. She then drove home. When it was suggested to her that she did not "dawdle in a very slow fashion as though you were tired or lethargic", she replied, "[j]ust normal, like, I didn't drag my feet." Newman J said that "a video showing the plaintiff moving around the town of Inverell at times in company with her mother, was demonstrative only of the fact that the plaintiff was exhibiting no fatigue."

Ms Dixon also denied that she and Mr Cross lived together and could not recall telling doctors in 1998 and 1999 that she lived in a de facto relationship with him. But she conceded that after Mr Cross left Inverell she had gone to the Gold Coast and lived with him for three months in 1995. She also conceded that sometimes he spends "night after night" at her place although she claimed he "comes and goes, because Sarah always wants him." She agreed that she and Mr Cross and two other persons had gone to the wedding to which we have referred "as a group". Asked whether she and Mr Cross were living together, she said, "Like I said David comes and goes." Evidence of them being together at the wedding, at the races and at Copeton Dam also suggested that the relationship was far from over. And if her statements to doctors as recorded in the medical reports in evidence have any probative value, they contain admissions that in 1998 and 1999 she and Mr Cross lived together in a de facto relationship. Mr Nicholas Mair also gave evidence that he regularly visited Ms Dixon's property and that Mr Cross "lives there."

Other evidence also suggested that as late as the year of the trial the relationship between Mr Cross and Ms Dixon was a close one. Mr Pelja gave evidence that, when a dispute arose between Ms Dixon and the Mairs during the year 2000, Mr Cross had gone with her to the Mair's house and that she made threats while carrying an iron bar. Ms Dixon denied that she had threatened Mr Mair and claimed that "David jumped out of the ute with this bar" and that she had taken the bar away from him. Mr Mair also claimed that Ms Dixon had threatened him with the iron bar. The evidence concerning this incident is not only inconsistent with Ms Dixon's claimed condition but it suggests that her account of her relationship with Mr Cross if not false was much closer than she had claimed in her evidence-in-chief. Newman J thought that her evidence concerning her relationship with Mr Cross went to her credibility.

The video films, the admissions and evasions of Ms Dixon in cross-examination and the evidence of Mr Pelja made an overpowering case for rejecting the account that Ms Dixon gave in evidence-in-chief and to the doctors concerning her condition. And when that account was rejected, there was no alternative account that could take its place. The Court of Appeal proceedings

Despite the many factors that Newman J referred to in rejecting Ms Dixon's case, the Court of Appeal held that he had not properly considered her case. The Court concluded that Newman J had erred in: . concluding that Ms Dixon's condition rested entirely on subjective symptoms; . failing to make any findings about the existence of symptoms which were observable by the medical experts for themselves, and which they had taken into account; . failing to assess the totality of the medical evidence in light of the symptoms observed by medical experts which did not depend on Ms Dixon's own history, and in the light of their professional experience and assessment of the probability that she was suffering from Q fever chronic fatigue syndrome.

The transcript of the argument in the Court of Appeal shows that the point of no proper consideration of the medical evidence concerning chronic fatigue syndrome came from the Court itself. Mr D F Jackson QC adopted a suggestion from the Court that "there wasn't a proper trial because appropriate attention wasn't given to all the factors that bore on the issues in [the credibility finding]". But Mr Jackson's argument did not rely on the reasons on which the Court of Appeal later relied. Instead, his argument was directed to establishing that the trial judge had erred in considering evidence as to whether Ms Dixon had Q fever infection after February 1995. As Mr Jackson said in argument: "The ... medical reports to which I've gone don't really suggest that it's right to say that the medical evidence showed no relevant serology after February 1995. That's really essentially what we would seek to say in relation to the appeal proper ...".

The Notice of Appeal on which the Court of Appeal acted was the third Amended Notice of Appeal filed in the appeal. It was filed on the morning of the appeal hearing. Yet it contained no ground that the trial judge had failed to consider the contents of the medical reports as evidence of chronic fatigue syndrome . Instead, the grounds of appeal were directed to his Honour's finding that Ms Dixon had recovered from the infection by February 1995.

Ground 1 asserted that his Honour misunderstood and misconstrued the serological data in evidence and the medical opinion thereon. Ground 2 asserted that the finding that Ms Dixon had recovered from an acute episode of Q fever by February 1995 was not medically justifiable. Ground 3 asserted that his Honour's findings concerning non-economic loss based on a closed period of Q fever infection to February 1995 was not sustainable on the evidence. Grounds 4 and 5 dealt with the effect of fresh evidence upon which Ms Dixon sought to rely. Ground 6 asserted that the trial judge had failed to take into account the serological tests done on 4 December 1995 and commented upon by Professor Boughton in his report. Ground 7 asserted that his Honour failed to take into account the opinions of three doctors in respect of his finding that the appellant had recovered from Q fever infection in February 1995. Ground 8 asserted that his Honour failed to take into account Professor Boughton's opinion that impairment of mental concentration and memory made the eliciting of a reliable history difficult in reaching his assessment of Ms Dixon's credibility. Ground 9 asserted that his Honour failed to take into account Professor Boughton's opinion in finding that Ms Dixon's acute Q fever infection ceased in February 1995 and the consequential assessment of damage under s 151G of the Workers Compensation Act .

Only Ground 8 had the remotest bearing on the issue of credibility or chronic fatigue syndrome . None of the grounds suggested, as the Court of Appeal found, that a miscarriage of justice had occurred because the trial judge had concluded that Ms Dixon's condition rested on subjective symptoms. No ground suggested that it had occurred because he had failed to make findings about the existence of symptoms that were observable by the medical experts. No ground suggested that it had occurred because he had failed to assess the totality of the medical evidence in light of the symptoms observed by medical experts which did not depend on Ms Dixon's own history. No ground suggested it had occurred because he had failed to consider the observations in the light of their professional experience and assessment of the probability that she was suffering from Q fever chronic fatigue syndrome.

Indeed, at the appeal a question arose as to whether even the matters raised in the grounds of appeal were run at the trial. During the hearing of the appeal, Heydon JA pointed out that Mr R J Burbidge QC, who appeared for Whisprun, objected to Mr Jackson's arguments concerning the failure to consider the infection issue because of Coulton v Holcombe - a leading case on the right of parties to run a new case on appeal. Heydon JA asked Mr Jackson whether he had "any instructions or material which would give a rebuttal to Mr Burbidge's point?" Mr Jackson replied: "I can't point to material that either supports what he is for or supports absolutely what we say ... However, all this material, all the medical evidence was before the judge and the situation in our submission was this ... and the matter with which we wouldn't disagree is this: What he said was because the continuing symptoms were entirely subjective the only issue which arose in the case was whether the plaintiff's account of her continuing symptoms was credible. When we're speaking of the only issue we're speaking about the matters to which the oral evidence was directed and we're speaking there of the continuing symptoms, the issue was whether her account was credible. In determining that issue he had to take into account - and there was nothing to suggest that we said otherwise - all the medical evidence and the consequence of the condition if she had it on the mental processes. He is entitled to find that she's overstating, it's a different thing."

Even if Mr Jackson's reply is seen as including a reference to the medical evidence concerning chronic fatigue syndrome - and the context is against such a construction - two matters are clear. First, he had no instructions and was unable to point to any evidence to the effect that Ms Dixon's case at the trial was that the medical evidence had probative value on the fatigue issue independently of the judge accepting her evidence. Instead, reversing the onus, he contended that nothing in the case pointed to the opposite conclusion. Second, at no stage did he argue - apart from in relation to memory lapse - that the trial judge had overlooked the matters that the Court of Appeal held that Newman J had overlooked.

The Court of Appeal seems to have assumed that the entire contents of the medical reports - including the histories and complaints - constituted independent evidence that was to be treated as if it had been given in the witness box. It seems to have assumed that, independently of Ms Dixon's credibility, the opinions in the reports were in themselves evidence of chronic fatigue syndrome . Indeed, because the medical experts were not called and their reports were not challenged, the Court of Appeal appears to have treated the contents of the reports as important benchmarks against which the parties' cases were to be tested. If the parties had so conducted their cases as to treat the reports as not depending on Ms Dixon's credibility, the failure of the judge to treat the reports in this way or to mention it would be an extraordinary omission. This is particularly so after the judge referred to the evidence of Ms Dixon's mother and Mr Cross's mother as providing some corroboration of Ms Dixon's evidence. It is in the highest degree improbable that Newman J would have mentioned this evidence as corroboration and omitted reference to the medical reports if counsel for Ms Dixon had placed any reliance on them as corroborative evidence of chronic fatigue syndrome which was not dependent upon Ms Dixon's evidence. The record does not disclose any support for the proposition that the case for Ms Dixon ran at trial was that later formulated by the Court of Appeal.

Furthermore, it is inherently unlikely that counsel for Whisprun would have accepted that the opinions of the doctors and the complaints to them had independent evidentiary value on the issues respecting chronic fatigue syndrome without insisting on cross-examination of the doctors. It is hard to credit that counsel would have failed to seek admissions from these witnesses that, if the evidence adduced by Whisprun were accepted, Ms Dixon did not have chronic fatigue syndrome or, at all events, that there was a strong probability that she did not.

The course taken by Mr W K Dodd SC, counsel for Whisprun, concerning the tender of the medical reports - and for that matter by Mr P Webb QC who appeared for Ms Dixon at the trial - is consistent with counsel accepting that, so far as chronic fatigue syndrome was concerned, everything in the reports depended on the trial judge accepting Ms Dixon as truthful. It is, or at all events was, a common enough practice in New South Wales in cases concerned with psychiatric conditions for psychiatric reports to be tendered on the assumption that the opinion expressed in them depended on the histories in them being proved in the witness box. That is the course that appears to have been taken in this case where no objective test existed to determine whether Ms Dixon had the symptoms of which she complained. Purists may decry this approach to expert evidence, but it saves time and expense. The days are long gone in civil trials in New South Wales when expert witnesses would sit through the whole of a party's or witness's evidence before expressing an opinion on that evidence. Here it seems plain that counsel proceeded on the basis that, if the credibility of Ms Dixon - the person who gave the histories recorded in the reports - was destroyed, the opinion of each doctor "may have little or no value, for part of the basis of it has gone" .

Apart from the fact that the medical reports were admitted in evidence, there is really no evidence that the case formulated by the learned judges of the Court of Appeal was ever the case that Ms Dixon ran at trial. And the notice of appeal - the third notice of appeal - did not suggest it.

Moreover, Ms Dixon's written submissions tell strongly against the way that the Court of Appeal saw the case. Those submissions were filed on 2 May 2001, over four months before the appeal was heard in September 2001. They were signed by Mr Jackson and contained no inkling that the trial judge had erred in the manner that the Court of Appeal subsequently found. The thrust of those submissions was the same as those found in the Third Amended Notice of Appeal.

In our opinion, the case formulated by the Court of Appeal was never run at the trial. Several matters point to this being so: . The statement by the trial judge as to the only issue in the case. The reasons of the trial judge are usually the best indication of what matters were in issue at the trial. . The omission of the parties to call the medical witnesses. . The failure of the Third Amended Notice of Appeal to suggest that the trial judge had not properly considered the case of chronic fatigue syndrome. . The focus in the Third Amended Notice of Appeal, the written submissions and the argument in the Court of Appeal on the issue of persisting Q fever infection. . The lack of any evidence - apart from the existence of the medical reports - to suggest that Ms Dixon's case at trial was as the Court of Appeal's reasons imply that it was.

Accordingly, this appeal must be allowed. It would be inimical to the due administration of justice if, on appeal, a party could raise a point that was not taken at the trial unless it could not possibly have been met by further evidence at the trial . Nothing is more likely to give rise to a sense of injustice in a litigant than to have a verdict taken away on a point that was not taken at the trial and could or might possibly have been met by rebutting evidence or cross-examination. Even when no question of further evidence is admissible, it may not be in the interests of justice to allow a new point to be raised on appeal, particularly if it will require a further trial of the action . Not only is the successful party put to expense that may not be recoverable on a party and party taxation but a new trial inevitably inflicts on the parties worry, inconvenience and an interference with their personal and business affairs.

As Water Board v Moustakas makes clear, a point may be a new point even though it is within the pleadings or particulars. The pleadings and particulars are frequently decisive in determining whether a party is seeking to raise a new point on appeal. But they are not conclusive. To determine whether a party is raising a new point on appeal, it is "necessary to look to the actual conduct of the proceedings" . Thus in Water Board, the plaintiff's case at trial had been that his employer was negligent in failing to prevent traffic from crossing in to the lane in which he was working. On appeal, the Court of Appeal of New South Wales allowed the plaintiff to raise a case that the employer was negligent in failing to provide a barrier to prevent the plaintiff from straying into the adjoining lane. This Court held that, although this alternative case was within the particulars, it had not been the plaintiff's case at the trial and the Court of Appeal had erred in allowing it to be raised on appeal.

As in Water Board , the case for Ms Dixon as formulated in the Court of Appeal is not the case that she ran at the trial. Moreover, it is a virtual certainty that, if such a case had been run at the trial, Whisprun would have wished to cross-examine the doctors. Because that is so, the Court of Appeal erred in considering this alternative case. No miscarriage of justice occurred because the trial judge considered the case that Ms Dixon ran at the trial. It follows that the Court of Appeal should have dismissed the appeal by Ms Dixon. The case formulated by the Court of Appeal

In any event, we are not convinced that the matters upon which the Court of Appeal relied would have affected the conclusion that Newman J reached concerning Ms Dixon's credibility or her failure to persuade him that she had established that she was suffering from chronic fatigue syndrome. The Court of Appeal held that the trial judge had erred in failing to consider whether Ms Dixon had symptoms that were observable to medical practitioners. The Court said that if such symptoms existed, then her case did not depend entirely on her credibility. However, with great respect to their Honours, none of these symptoms was "objective". They depended on what Ms Dixon told the doctors. Nor can it be said that the only possible cause of these symptoms was chronic fatigue syndrome. Weight loss

The Court of Appeal regarded Ms Dixon's weight loss as significant. Ms Dixon reported that she had originally weighed 12-13 stone, and that this dropped to around eight and a half stone. This evidence depends of course on Ms Dixon telling the truth about her earlier weight. But accepting that she did lose three and a half stone in weight, it says very little, if anything, concerning whether she was suffering from chronic fatigue syndrome after February 1995. No one disputes that she had Q fever infection and that it persisted for at least ten months. The infection itself is sufficient to explain the weight loss. What is more significant is evidence that subsequently she put on four and a half kilograms. A weight of nine stone or 57 kilograms (having regard to her height etc) hardly indicates ill health. A photograph of her at Inverell races does not suggest that Ms Dixon was underweight. Weight loss in and of itself does not necessarily point to Ms Dixon having chronic fatigue syndrome. Abdominal tenderness

The Court of Appeal also considered that Ms Dixon's complaints of abdominal tenderness particularly around the liver supported her case. Abdominal tenderness is consistent with chronic fatigue syndrome. But it is not conclusive of it. There are many other explanations that could account for this tenderness. And the existence of the abdominal tenderness itself appears to depend on the truth of Ms Dixon's responses. The medical reports are silent as to whether a doctor can actually tell by touch or experience whether a patient has such tenderness. If a diagnosis of abdominal tenderness depends upon the truth of what the patient says, her abdominal "tenderness" is not objective evidence. A patient may be untruthful or may falsely believe that she has the tenderness. The possibility of a psychosomatic illness of some sort also cannot be ignored. The evidence indicates that, by reason of the poor safety practices of Whisprun, a significant number of its employees have suffered from Q fever infection and chronic fatigue syndrome. Indeed Ms Dixon's father had so suffered. Nor should one ignore the fact that the symptoms of chronic fatigue syndrome largely coincide with the symptoms of Q fever infection from which Ms Dixon undoubtedly suffered for at least ten months. She was undoubtedly aware of the indicia of chronic fatigue syndrome. Her claim of abdominal tenderness, if accepted, supports her case but it falls well short of establishing that she suffered from chronic fatigue syndrome or that her condition was anywhere near as bad as her evidence-in-chief painted. Pale and depressed

The Court of Appeal also saw an objective symptom in the observations of doctors who thought that Ms Dixon looked pale and depressed. No doubt her paleness is more "objective" than some of the other symptoms to which the Court of Appeal referred. Nevertheless, Ms Dixon could have been pale and depressed for reasons other than the syndrome. And more importantly, the trial judge saw Ms Dixon in the witness box over a lengthy period undergoing an extensive cross-examination. Compared to his Honour's opportunities to observe Ms Dixon under the stress of cross-examination, the observations of doctors - particularly at times when she was undoubtedly suffering from Q fever infection - seem of little significance. Memory impairment

The Court of Appeal also classified as an objective symptom the fact that a number of doctors suggested some impairment of Ms Dixon's memory. But Newman J heard and observed Ms Dixon give evidence under the stress of cross-examination. He was in a better position than the doctors to form an opinion as to whether there was any impairment of her memory. No doubt some of her answers, as they appear in the transcript, suggest a lack of recollection. But his Honour was in the best position to judge whether this vagueness of recollection represented evasions or genuine lapses of recollection. Moreover, at different points of the cross-examination Ms Dixon's recollection seemed remarkably clear. She seemed to have no difficulty recalling the details of her visits to Inverell on which she was cross-examined. Significantly, at one point she corrected Mr Dodd when he suggested to her that she had taken an envelope into a forklift hire building. She said "you don't mean a forklift - it was - my mum gave me a bill to go and pay for her [at] a brake centre".

In any event, even if she was forgetful it does not necessarily mean that she had chronic fatigue syndrome. She had other worries during the relevant periods. She had had a pregnancy terminated which she concealed from her parents. This was obviously a matter of some concern to her: she claimed that after she had fallen out with her friend, Mrs Mair, she had threatened to tell Ms Dixon's mother about the termination. Failure to consider objective symptoms

Contrary to the stance adopted by the Court of Appeal, there was no reason why Newman J was required to give any weight to the doctors' opinions that Ms Dixon was suffering from chronic fatigue syndrome. The opinions themselves seem premised on an acceptance of Ms Dixon's account of her history and complaints. His Honour was entitled to take the view - which we would take ourselves - that their opinions were contingent on Ms Dixon suffering from the problems of which she complained. If she did not, there was nothing in the reports, with the exception of the abdominal tenderness and her paleness, that gave any support for the opinions expressed.

The Court of Appeal also considered that the trial judge should have considered that there was a statistically significant chance of Ms Dixon having chronic fatigue syndrome. But Newman J was entitled to take the view that the statistical probability of her having the condition was of no significance after what he had seen and heard from the cross-examination, the videos, the photographs and the evidence called by Whisprun.

It is not readily to be supposed that the various matters upon which the Court of Appeal placed so much weight would have induced Newman J to accept that Ms Dixon's condition was as she testified in her evidence-in-chief and as she reported to the doctors. Further, it should not be accepted that Newman J failed, at least in a general way, to consider the matters to which the Court of Appeal referred. The fact that his Honour did not refer to these matters in his judgment is not decisive. A judge's reasons are not required to mention every fact or argument relied on by the losing party as relevant to an issue. Judgments of trial judges would soon become longer than they already are if a judge's failure to mention such facts and arguments would be evidence that he or she had not properly considered the losing party's case.

However, it is unnecessary to determine whether the matters to which the Court of Appeal referred were or should have been considered by the trial judge. For the reasons that we have given, they were not part of Ms Dixon's case. To suggest that a trial judge has not properly considered a party's case is a serious charge. Such a suggestion should be accepted only when the record of the trial or other evidence persuasively suggests that the judge failed to discharge that paramount judicial duty. With great respect to the learned judges of the Court of Appeal, we do not think that the evidence met that standard. Order

The appeal should be allowed with costs. The orders of the Court of Appeal should be set aside. In their place should be substituted orders that the appeal to that Court be dismissed with costs.

KIRBY J. This appeal concerns the boundaries of appellate power. Specifically, it relates to the powers of an appellate court, conducting an appeal by way of rehearing, to reverse factual conclusions reached by a primary judge who states or implies that such conclusions were based on an assessment of the credibility of a party.

Considerations of this kind are not necessarily fatal to appellate reversal of the primary judge's conclusions. It always remains for the appellate court to undertake the appeal by way of rehearing. Such an appeal "should be a reality, not an illusion; if the judges of an appellate court hold the decision of the trial judge to be wrong, they should correct it" .

However, normally, to secure reversal of a primary judge's credibility-based conclusions, it is necessary for the challenger to demonstrate that such conclusions are flawed by reference to incontrovertible facts or uncontested testimony: showing that the primary judge's decision was erroneous, notwithstanding that it appears to be (or is stated to be) based on credibility findings . Such was the case in State Rail Authority (NSW) v Earthline Constructions Pty Ltd (In Liq) (" SRA "). There a witness, disbelieved by the primary judge, was supported by contemporaneous documentary evidence and unchallenged testimony of other witnesses that had not been considered and that evidence combined to demonstrate the fragility of the judge's conclusion. Such was also the case in Fox v Percy where the decision of the primary judge, although based on a credibility assessment, could not be reconciled with other testimony that the primary judge accepted. In particular, it did not accord with a contemporary record that contradicted the judge's conclusion.

Although, in the present case, the appellate court pointed to certain evidence of uncontested or undisputed facts to help establish the proposition that an incompatibility could be shown (similar to the foregoing cases), that evidence fell short of an objective demonstration that the primary judge "got it wrong". The question therefore arising, in the circumstances of the trial and of the primary judge's reasons in this case, is whether the appellate court's conclusion "that some substantial wrong or miscarriage has taken place in that the plaintiff's claim has not been properly considered" was correct and can stand compatibly with the grounds of appeal under consideration, the issues that were addressed in the appeal, the applicable law and, in particular, the rule of restraint upon the reversal of a primary judge's conclusions based, wholly or in part, upon findings about the credibility of critical evidence. The facts

The basic facts are set out in the reasons of Gleeson CJ, McHugh and Gummow JJ ("the joint reasons") and in the reasons of Callinan J . As is normal in a case of this kind, questions were asked, and evidence was tendered at trial, designed to probe the true extent of the disability of Ms Sonya Dixon (the respondent) in respect of the condition for which she claimed damages for negligence against her former employer, Whisprun Pty Ltd (the appellant).

The essential ingredients giving rise to the respondent's claim were not disputed. Before she was exposed to infection with Q fever at the appellant's abattoir in 1994, the respondent was "very healthy" . This was not simply her own assessment. It was confirmed by her family and friends and by two medical practitioners (Drs Hall and Thatcher). They had seen her professionally before and following the infection. Moreover, it was confirmed by the fact that the respondent, at the time she was exposed to Q fever, held down three jobs - in the appellant's abattoir, at a local supermarket for three days a week and at a fruit market on Saturdays. She was described thus: "She was working to pay off a home mortgage. She was an enthusiastic, reliable and responsible worker. She had had a daughter. Her relationship with her daughter, her de facto husband, his mother and other persons was happy. Until at least 1993 she played hockey, basketball and tip football. She swam. She rode horses. She weighed 12 stone. There was no existing medical condition before July 1994 which could have detrimentally affected her health thereafter."

Into this seemingly stable and healthy condition intruded the infection that occurred when the respondent's duties for the appellant obliged her to remove foetal blood from the hearts of unborn calves by a technique involving sucking the blood out with a pipette. The respondent ingested some of the blood. She had not first been administered a vaccine that could have protected her from the agents that cause Q fever. Professor Kendall said that he could think of "no method of ensuring infection with Q fever more certain than the work practice imposed on her". He found "the whole matter really astounding" .

After her exposure, the respondent suffered an acute episode of Q fever commencing in July 1994. Since then, she has been observed and treated by a series of medical practitioners, some of them nominated by the appellant. She suffered symptoms that were initially compatible with both the acute phase of Q fever infection and later with a chronic fatigue syndrome which often follows such exposure.

All but one of the medical practitioners who treated the respondent or reviewed her for the purposes of her claim against the appellant found that she was manifesting signs or symptoms that were compatible with the chronic fatigue syndrome, a flu-like condition, of which she complained. The only witness out of step in this regard was Dr Sutherland. He was qualified to give evidence for the appellant. However, it was demonstrated that he had received an incorrect history and was unaware of a bout with Q fever that the respondent had suffered late in 1994. This error, influencing his awareness of the progress of the respondent's condition, falsified Dr Sutherland's conclusion. It left the medical evidence otherwise speaking with a single voice, favourable to the respondent's condition and its aetiology, namely that it was caused by her exposure to Q fever.

Nor did the medical opinions favourable to the respondent describe a progress of the condition that was unique or even exceptional. Scientific materials received at trial indicated that at least 10-20% of those exposed to Q fever go on to suffer chronic fatigue syndrome. Whilst in some reported cases the condition clears up quickly after the acute phase, in others the disability endures for "2 or more years" . In still others : "6 years after the outbreak, anecdotal reports suggested that some patients involved in the original outbreak had long-term symptoms, notably fatigue and lethargy. With reports from Australia also suggesting chronic symptoms after acute Q fever, we assessed the prevalence of chronic symptoms in patients originally involved in the acute outbreak in the UK. 114 patients for whom we had current addresses were circulated with a questionnaire based on that used by Marmion and colleagues (personal communication). 102 (90%) responses were received and after removing those who had died or moved away, 83 (70 male) were left for analysis ... There was a high prevalence of fatigue (66%), joint aches (69%), sleep disturbance (65%), cough (89%), and sweats (53%) in the Q fever group ... There were a number of statistically significant differences between cases ... These findings support the view that a chronic, post-acute Q fever syndrome exists which is similar in many ways to chronic fatigue syndrome. Our group was not occupationally exposed and the presence of symptoms in these subjects is, therefore, not coloured by impending litigation."

The foregoing scientific evidence, which was not contradicted or qualified, as well as the opinions of treating medical practitioners, was received, in report form, without any objection being raised by the appellant. The appellant did not require, as it might have done, that any of the medical witnesses whose reports were tendered attend court for cross-examination on their reports.

The respondent's oral evidence recounted symptoms similar to those described in the scientific studies. Her histories, given to the medical practitioners over the years between her original exposure to Q fever and the trial, were also consistent. They indicated some variation in the severity of her condition from time to time. But they were always compatible with the diagnosis of chronic fatigue syndrome as described in the literature and as recognised by the medical experts who gave evidence. The respondent's complaints were also supported by testimony from her family and others. They were compatible with the signs presenting to the medical experts, qualified to give evidence, who examined her after her exposure to infection for the purposes of evaluating her claim.

The respondent's complaints were compatible with the reports of muscle tenderness and abdomen tenderness; with descriptions of her behaviour and appearance pre- and post-exposure; and, most importantly, with the severe weight loss that followed her infection with Q fever and continued up to the trial. The respondent's pre-infection weight of 12-13 stone (approximately 76-83 kg) fell, by May 1995, to 57.3 kg. This weight was maintained to May 1998. By March 1999 it had crept up to 61 kg. However, this was still markedly below her pre-infection weight. The respondent was never challenged about the weight loss, although she repeated reference to it during her testimony. It was an objective sign that was consistent with her complaint of lethargy, depression and unwellness, interrupted sleep, unreasonable irritability and interference in short-term memory and concentration - all of which (according to the medical evidence) were symptoms common to the chronic fatigue syndrome following Q fever with which the respondent had been diagnosed.

Given the appellant's general acceptance of the circumstances in which the respondent was infected; the immediate pre-infection state of the respondent; the resulting acute phase of Q fever; and the uncontested fact that this can, in 10-20% of cases, lead to prolonged chronic fatigue syndrome, what hypotheses could be propounded to support the termination by February 1995 of the respondent's entitlement to damages, as found by the primary judge (Newman J) ? On the face of things it seems arbitrary. But was it?

The primary judge's decision, in effect, confined the respondent's entitlement to damages to the period of acute infection beginning with the pathology tests reported in July 1994 by Dr Hall. Serological tests conducted in February 1995 showed that, at that time, the respondent was not suffering from acute or chronic Q fever infection . However, by this stage, her claim was for "chronic fatigue syndrome ", the not uncommon sequel to Q fever exposure. For that syndrome there was no objective pathology test that could incontrovertibly confirm or negate the condition. To ask for, or demand, such positive proof would be to require the impossible. It would be unreasonable and thus unlikely to be the requirement of the common law . It would risk imposing on the respondent a heavier burden of proof of the continuity of her condition than is required in a civil proceeding claiming damages for negligence. The competing hypotheses

As Heydon JA pointed out in the Court of Appeal, there were at least four possibilities to explain the respondent's complaints: (1) That she was suffering from "some form of chronic state" wholly coincidental and not connected with her earlier exposure to Q fever; (2) That, although she had initially suffered the symptoms complained of as a complication of Q fever, "psychological components and contributants" had taken over to explain her ongoing condition; (3) That her complaints were genuine and attributable to an aftermath of the Q fever exposure; or (4) That she had decided, around February 1995, "to live a complete lie and maintain a consistent but wholly or largely mendacious story to be told to her family, her friends, her doctors, her lawyers and ultimately the court over a period of six years".

A fifth possibility, suggested during argument before this Court, was: (5) That she might indeed have been suffering from "chronic fatigue syndrome" consequent upon her exposure to Q fever but, by reason of untruthful evidence and exaggerations at the trial, she had left the primary judge in a state of uncertainty about the extent of the disability attributable to the appellant's negligence. Upon this footing, the rejection of the respondent's claim after February 1995 rested upon her failure to prove that claim in accordance with the burden and standard of proof resting upon her as the plaintiff in the action. The foundations for the credibility conclusion

The untruthful evidence and exaggerations upon which the appellant fought its case at trial turned out to be critical for the conclusion of the primary judge. He noted that "the cross-examination [took] the form of a direct attack upon the plaintiff's evidence as to the symptoms she suffered but also as to a number of peripheral matters" . One such "peripheral" matter (mentioned by the primary judge) gives the flavour of the way her claim was met during the trial. In her evidence-in-chief, the respondent agreed that she lived on her parents' property in Inverell. It transpired that she was housed in a shed which stands on land that she herself owns. She was asked why she had not corrected the questioner and stated that the shed was on her own property. Her answer was that she had "no idea" . Perhaps, like not a few lawyers before and since, she was not quite sure what "property" meant . Perhaps she did not think the ownership of the land on which her shed stood really mattered enough to make a fuss so as to correct the questioner's assumption.

Much of the cross-examination of the respondent, however, was directed to demonstrating that she had attended events and participated in activities incompatible with her complaints of fatigue. At first, the respondent said that she could not remember that she had attended a race meeting organised by the appellant's abattoir after she left the appellant's employment. However, she was then shown photographs in the local newspaper and a video film proving that she was present at a race meeting on New Year's Day 1998. The primary judge found that the video film showed "no signs whatsoever of fatigue and was demonstrative of the plaintiff enjoying herself at that meeting" where she was present for "no less than four hours and twenty minutes". Further video film showed her "walking briskly an extensive distance" in Sydney streets . The judge said that her "actions as depicted in that video were quite inconsistent with the picture she had painted of herself both in chief and in histories to doctors" .

There followed questions about a Christmas barbecue at the Copeton Dam. The judge described the respondent's evidence in that respect as "evasive" and as such "did not enhance her credibility" . He detected a difference between the evidence of a witness, Mr Pelja, concerning certain limited watersport activities on the Dam in which the respondent participated and the respondent's recollection about her activities on the given day. However, even Mr Pelja's testimony about the events was imperfect.

The respondent was tackled about statements to a psychiatrist that she did not smoke or drink. She explained that she took those questions to refer to regular smoking and drinking. A video film lasting eight minutes at a wedding "demonstrated her to be very active both in terms of dancing and socialising and drinking" .

The judge also viewed a video film showing an equestrian event which the respondent attended and in which her daughter was involved and also showing the respondent moving around the town of Inverell. Neither of these, he felt, showed the respondent exhibiting the fatigue of which she had complained in evidence .

Although, objectively, one might consider the foregoing cross-examination, photographs and video film as having only limited significance when stacked up against the entirety of the evidence called by the respondent and her medical reports and lay witnesses - and her own concession that her condition of fatigue varied over time - this was not the way the primary judge saw it. Despite the evidentiary support that the respondent received from her mother and from the mother of her former de facto husband (to the effect that she had good and bad weeks) the primary judge said : "I have concluded that the matters raised in cross-examination with the plaintiff and her responses to them effectively destroyed her credibility. ... I am not satisfied, on a balance of probabilities, that the plaintiff has in fact suffered from the symptoms which she recounted to medical practitioners and in chief to this Court."

Having treated the question whether the respondent's account of her continuing symptoms was credible as determinative of the claim, the primary judge confined the claim to the short period for which there was objective confirmatory evidence in the form of the serological tests proving (in effect beyond doubt) the presence of the acute phase of Q fever infection. Because of the applicable law, this conclusion led the primary judge to dismiss the respondent's action. The Court of Appeal reversed that conclusion. It ordered a retrial of the respondent's action. By special leave, the appellant has appealed to this Court.

The appellant complained that the Court of Appeal had erred in disturbing the primary judge's conclusion that rested on the foregoing reasoning. In particular, the appellant complained that no error had been demonstrated to justify the order allowing the appeal. So far as that order depended on a conclusion that the respondent's claim had "not been properly considered", it involved procedural unfairness to the appellant, in that such a ground had not been specifically relied upon before the Court of Appeal which had thus taken upon itself a function of identifying a flaw in the conduct of the trial of which the respondent had not expressly complained and which the appellant had not been afforded a proper opportunity to answer. Appellate disturbance of credibility conclusions

Overseas authorities : The rule of restraint, limiting the power of appellate judges to disturb fact-findings at trial that have been influenced by assessments of the credibility of witnesses, is universal. Appellate courts in Australia , as in other common law countries, are protective of such findings. The formulas used vary as between different jurisdictions. The principle of restraint is common.

Thus, the United States Supreme Court has adopted a rule obliging appellate deference to trial decisions based on credibility assessments unless such decisions are "clearly-erroneous" . In the New Zealand Court of Appeal, the rule has been expressed that such factual findings may not be disturbed unless "compelling grounds are shown for doing so", sufficient to demonstrate that the primary decision-maker has erred . In New Zealand, it is necessary for the appellant to show that the impugned fact-finding was "wrong" .

In England, from where the rules observed in Australia are originally derived, new Rules of Court have recently been adopted for application to appeal courts in the form of Pt 52 of the Civil Procedure Rules . These Rules came into force on 2 May 2000 following consideration between 1994 and 2000 of the reform of civil appeals . Part 52, r 11(3) of the new Rules provides that: "the appeal court will allow an appeal where the decision of the lower court was - (a) wrong; or (b) unjust because of a serious procedural or other irregularity in the proceedings in the lower court".

The issue of reconsideration of trial fact-finding has been recently considered by the Supreme Court of Canada in Housen v Nikolaisen . To authorise appellate disturbance of findings of fact made at trial in Canada, the challenger must demonstrate "palpable and overriding error" . The metaphorical language may be slightly different. But its purport is the same. It is not enough that, on the evidence, the appellate judges would themselves have reached a conclusion different from that reached by the primary decision-maker. Such an approach would involve usurping the functions of the decision-maker at trial and confusing them with the appellate function in fact-finding which is real, but limited and different.

The common features of the United States, New Zealand, English and Canadian formulas is therefore that they insist upon appellate restraint; require a demonstration of error; and oblige that such error be "clearly", "compellingly", "palpably" or "plainly" shown. The rule of restraint is not confined to cases where the decision at trial rests, wholly or in part, on conclusions about the credibility of a party or of an important witness. It applies generally to delineate the respective functions of trial and appellate tribunals.

Reasons for restraint : Several considerations have been identified as underpinning the rule of restraint . Such considerations include: (1) The relative scarcity of resources available to the courts to resolve the disputes of litigants and the need to deploy them efficiently; (2) The normal expectation that such disputes will be resolved finally at trial and should not be regarded as merely provisional ; (3) The need for appellate respect for the autonomy and integrity of primary judges in the discharge of their powers within the scope of their functions ; (4) The acknowledgment by appellate judges of the expertise of many primary decision-makers and the advantages which they often enjoy in hearing and seeing witnesses and in being exposed to the evidence in a case in its totality ; (5) The recognition that appellate procedures are not always well suited to the review of large masses of transcript and are often selective and "telescopic" in their examination of such materials ; and (6) The acknowledgment that the reasons for a primary decision can only ever represent a summary of the main elements that have led the decision-maker to his or her conclusions .

In every case, the appellate court must reconcile the rule of restraint (enhanced where a decision rests on a credibility assessment that was reasonably open to the primary decision-maker) and the need to protect parties against clearly flawed primary decisions, illogicality in reaching them and injustice that demands that the appellate court exercise the functions of review conferred on it by the legislature .

Authority for intervention : In SRA , I pointed out that findings as to the credibility of a particular witness do not "represent the end of analysis by the appellate court" . Such findings cannot justify the abandonment or repudiation of a jurisdiction and power conferred on such a court by legislation . In a lawful discharge of appellate judicial power it would be impermissible to close the appeal books simply because the credibility of a party or witness was raised and decided by the court below. In terms of legal principle it could not be so, given the broad language in which the appellate court's functions are typically stated; the judicial character of the repository of such functions; and the great variety of factual circumstances presented by appeals and by the judicial reasons that give rise to them.

In SRA , I listed a number of cases, illustrated by decisions of this and other courts, in which, despite credibility findings, appellate intervention had occurred and been upheld. As I pointed out in that case, the instances cited were "by no means exhaustive" . They included cases (1) where the primary judge's conclusion, although expressed in terms of credibility, was "plainly wrong" as demonstrated by incontrovertible facts or uncontested testimony ; (2) where the conclusion was based on evidence wrongly admitted, occasioning a substantial miscarriage of the trial ; (3) where the reasons, going beyond credibility, indicated a consideration at trial of irrelevant matters or a failure to weigh all relevant issues ; (4) where the circumstances in which evidence was given, relevant to credibility, was unsatisfactory ; or (5) where the primary judge had made it plain that credibility considerations or impressions were not determinative for the judgment in question .

There were two further categories that I mentioned in SRA . They are relevant to the present appeal. They were: (6) where the credibility determination "leaves untouched other evidence which requires separate evaluation with no obstacle of a credibility finding" and (7) where, notwithstanding the credibility finding, the "extreme and overwhelming pressure" of the rest of the evidence at the trial is such as to render the conclusion expressed at first instance so "glaringly improbable" or "contrary to compelling inferences" of the case that it justifies and authorises appellate disturbance of the conclusion reached at trial and the judgment giving it effect. As Callinan J points out in his reasons , the categories of appellate intervention listed in SRA are not closed.

The common law usually recoils from absolute rules of mechanical or inflexible application. It does so because its long experience illustrates too often the need to retain elements of flexibility to cover an exceptional case. Unyielding rules (such as would forbid any appellate disturbance of a judgment in any case, in any circumstance in which the primary decision-maker had rested his or her opinion in whole or part on an assessment of credibility) could, in a particular matter, become an instrument of serious injustice. That is why common law rules normally reserve a place for the exceptional case . The general rule of appellate restraint with respect to disturbing decisions depending on the impression of credibility of a party or witness carries with it exceptions that acknowledge the need, very occasionally, to intervene to correct demonstrated error and thereby to prevent a clear injustice. The appellate intervention was warranted

In the appeal to the Court of Appeal, the respondent relied on objective evidence (such as the undisputed initial infection; the significant, established and persisting weight loss and other signs compatible with chronic fatigue syndrome noted by the medical witnesses in their reports). She also relied on substantially uncontested and apparently reliable observations (including those recorded by the medical witnesses and those that the family and other lay observers related concerning the appearance and conduct of the respondent). She did this in order to build a case of the common kind that would authorise a conclusion by the Court of Appeal that the primary judge's credibility-based findings were contradicted by "incontrovertible facts".

The strongest point in this aspect of the respondent's case was the significant weight loss. Obviously, it was an objective sign, carefully recorded over an extended period, including long after February 1996 when the primary judge found that the effects of the exposure to Q fever had no further proved consequences for her health. Certainly, that evidentiary fact was significant. It was given pride of place in the collection of matters to which the Court of Appeal enumerated in justifying its intervention. But as that Court recognised, it was theoretically possible that the weight loss might have had another cause; that the complaints to the doctors were recounted falsely; and that the respondent, after February 1996, set about, effectively, a fraudulent course of claiming continuing disability based upon her general knowledge of the effects of Q fever derived from her own exposure and that of her father at an earlier time. Perhaps over six years she had maintained a diet, resisting all culinary temptations to keep up the litigious pretence so as to bolster the proof of the claim.

Against such possibilities there was no knock-out blow in the form of contemporary documents (as in SRA and Fox v Percy ) nor an objective serological test (as had earlier proved beyond doubt the initial exposure). However, the lack of the latter form of evidence to clinch the claim of disability was not unusual in the respondent's condition. For such a chronic fatigue syndrome there was no objective test available either to the appellant or the respondent. Accordingly, the lack of an affirmative serology was neutral so far as the acceptability of the respondent's claims and the appellant's dispute about them were concerned.

It was this analysis that led the Court of Appeal to the residual categories for appellate review of credibility-based decisions mentioned in SRA and like cases. The question was this: had the primary judge failed to give separate evaluation to the evidence untouched by the testimony of the respondent which he felt to be damaged by his credibility findings so that the evidence would be viewed as a whole and credibility impressions tested against the entirety of the evidence? Looking back over the trial, were the judge's findings in any case (viewed in the context of the rest of the evidence) "glaringly improbable" or "contrary to the compelling inferences of the case"? Did they suggest that the judge had not considered all of the evidence properly?

The Court of Appeal concluded that these questions should be answered in the affirmative. In my opinion that conclusion was open to that Court. Unlike the majority in this Court , I do not consider that it was open to the primary judge to disregard the medical evidence on the basis that it was undermined by the credibility findings. With respect, this was not a case where the opinion of a medical expert was entirely dependent upon factual assumptions alone, provided by a patient's history so that such an opinion would only be as acceptable as the history on which it was based . The medical evidence recorded objective signs. At least to that extent it had potential evidentiary value. The scientific evidence was received without objection. It should have been evaluated, as such, by the primary judge, not discarded .

This Court, whose function is to correct error on the part of the intermediate court, should not retry the case. Its sole function is to decide whether, in intervening as it did in the particular circumstances of the case (correctly described as unusual), the Court of Appeal erred. In my view, only a mechanical and unthinking application of the rule of restraint based on the credibility findings of the primary judge would warrant a conclusion of appellate error which this Court would be warranted to correct. There was no relevant procedural unfairness

Grounds and argument of appeal : There is no substance in the complaint that the appellant was denied procedural fairness in the manner in which the Court of Appeal decided the appeal in favour of the respondent. The correctness of the primary judge's finding on the credibility of the respondent at trial was squarely raised in the Court of Appeal by the grounds of appeal to that Court (grounds 1, 2, 3 and 8); by the submissions advanced orally before that Court; by the supplementary submissions filed there on the respondent's behalf; and by the issues raised with the appellant's counsel by the Court of Appeal itself, recorded in the transcript.

The Court of Appeal squarely presented the question whether the features of the appellant's evidence that the primary judge found to be unsatisfactory, could itself have been caused, in part at least, by the very disabilities of which she complained. There was medical evidence to support a conclusion that post Q fever chronic fatigue syndrome can result in impairment of mental concentration and memory. This fact had been the subject of submissions in the Court of Appeal. There is no indication that the primary judge gave proper consideration to that possibility. Nor, as Heydon JA pointed out , was there any indication that allowance had been made by the primary judge for the fact that the respondent was "not extensively educated and apparently not very articulate", was probably nervous and alleging a condition which, if present as she claimed, could have affected her testimony because it affected her memory.

Instead of analysing the weight and significance of the copious medical and scientific evidence, received without objection on the part of the appellant (including that recording the respondent's consistent record of complaints) and instead of giving attention to the recorded manifestation of signs and symptoms noted by the medical experts (none of whom raised doubts as to the respondent's veracity), the primary judge gave what appears, with respect, to have been disproportionate weight to a relatively small number of photographs and video films that showed, on particular occasions, that the respondent was able to walk and stand and engage in comparatively modest physical activities for comparatively limited intervals of time.

It is true that the primary judge did not have the benefit of the oral evidence of medical witnesses concerning the significance of what was seen in the photographs and video films. However, on the face of the evidence of, and concerning, the respondent, there was no such contradiction as justified the very serious conclusion that, in effect, the respondent had persisted for year after year in a false claim of disability or had failed to prove the continuation of her disability according to the civil standard.

Both the respondent, and her family and other witnesses, confirmed that she had ups and downs, good weeks and bad weeks, indeed "mood swings" which were compatible with the recorded descriptions of chronic fatigue syndrome following Q fever. In such circumstances, to dismiss her claim, and without proper analysis of the weight, consistency, duration and effective unanimity of the uncontested medical evidence, produced an outcome that demanded appellate correction. It was a disposition of the case in which evidence that required separate evaluation with no obstacle of a credibility finding had been neglected or ignored by the primary judge. It was also one which, viewing the evidence as a whole, had resulted in a judgment that was contrary to the overwhelming pressure of the entirety of the evidence and so was "glaringly improbable" or "contrary to the compelling inferences of the case".

I do not accept that the Court of Appeal reconstructed the respondent's complaints concerning the primary judge's reasons in a way that was unfair to the appellant. The conclusion that the respondent's claim had "not been properly considered" was inherent in the respondent's attack on the credibility-based conclusions recorded by the primary judge. As the cases concerned with complaints of such a kind collected in SRA demonstrate, it is not uncommon (and is often essential) for a party mounting an appeal in which a judge has stated reasons in terms of conclusions about credibility, to attempt to overcome such conclusions in precisely the way that the respondent sought here .

Reference to Supreme Court Rules : The power of the Court of Appeal, in disposing of the appeal is, as Callinan J points out, contained in s 75A(10) of the Supreme Court Act 1970 (NSW) . That power could not be stated in larger terms. That Court is empowered, relevantly, to "give any judgment ... which ought to have been given or made or which the nature of the case requires". The reference by Heydon JA to the language of Pt 51 r 23(1) of the Supreme Court Rules 1970 (NSW) was a reference to a rule of court that purports to confine the power of the Court of Appeal to order "a new trial". In another case, a question might arise as to whether that sub-rule is within the rule-making power granted by the Act , in so far as it purports to cut back the generality of the power conferred by the New South Wales Parliament in s 75A(10) of the Supreme Court Act . A question might arise as to whether such confinement constitutes a rule "for regulating and prescribing the procedure ... and the practice to be followed in the Court" . Such questions do not need to be decided in this appeal, given that the reference to the sub-rule was no more than an indication by Heydon JA of the fact that he had considered the rule that an order for a new trial should only be made if a "substantial wrong or miscarriage" had taken place . In his view it had.

I do not read Heydon JA's reasons for the Court of Appeal as suggesting that the limitation expressed in Pt 51 r 23(1) of the Supreme Court Rules is, in some undefined way, elevated into a positive ground for intervention in every case where a "substantial wrong or miscarriage" is shown. The Court of Appeal's reasons negative such an interpretation.

Clearly, the Court of Appeal considered that, when the whole of the evidence at the trial was taken into account, the rejection of the respondent's case on credibility grounds was unpersuasive. Whether expressed in terms of an affirmative conclusion that the respondent had pursued a false claim after February 1996 or in the negative terms that she had not proved her disability after that date, the Court of Appeal considered that the primary judge's reasons were unconvincing. Stacked up against the objective evidence that was available having regard to the nature of her condition, the consistent record and recorded opinions of the medical experts received without qualification or objection and the limited considerations on which the adverse credibility finding was based, I have no doubt that the Court of Appeal reached a conclusion that was open to it. Rationality of the trial and appellate process

Rational process v tournaments : Lying at the heart of the difference between the majority and the minority in this Court (and between the approach of the primary judge and of Heydon JA in the Court of Appeal) is a different notion of the function of a trial judge in a case such as the present.

With respect, the joint reasons in this Court, and the reasons of the primary judge, appear to approach that function as if the judge were the successor to the adjudicator of the combat of knights of old - in a kind of public tournament between parties. In my view, we have travelled some distance since those times. The modern civil trial process is a more rational undertaking. It is based upon a close analysis of the relevant evidence, evaluated by a competent decision-maker who is obliged, if a judge, to give reasons which explain the decision arrived at.

In this case, Heydon JA (for the Court of Appeal) and Callinan J and I in this Court, do not accept that the primary judge gave to the evidence - and all of it - the consideration that should have been given . If the primary judge simply viewed the case as a contest mainly about whether the respondent, in her evidence, was a credible witness, that would have been a departure from the judge's function. The law has advanced since the days when truth was distinguished from falsehood at trial by battle and ordeal or by their modern equivalent - conclusive judicial assessment based on impression and on necessarily limited evidence .

Lies and civil proceedings : Some judges in the past regarded untruthful evidence - even about peripheral or irrelevant matters - as fatal to a litigant. Most judges today understand that the evaluation of evidence involves a more complex function, requiring a more sophisticated analysis. Courts, after all, are not venues for the trial of the parties' morality or credibility, as such. As judges often explain to juries in criminal trials, people sometimes tell lies in court and elsewhere for extraneous and irrelevant reasons, having nothing to do with the legal issues in the trial . If this is true in criminal trials, it is equally true in civil trials. What is important is not the proof of untruthfulness, as such, but the significance (if any) of any demonstrated falsehoods for the issues at trial. That significance can only be judged when measured against the entirety of the relevant testimony. By its logical force, that testimony may well require that falsehoods be ignored as irrelevant or immaterial to the decision-maker's ultimate conclusion. In particular cases, it may require the decision-maker, within the pleadings, to consider and decide a case different from - or even contrary to - that advanced by the party, because such is the legal entitlement of the person concerned .

Obligations of this kind recognise the ultimate duty of the decision-maker in an Australian court to decide a case according to law and the substantial justice of the matter proved in the evidence, not as some kind of sport or contest wholly reliant on the way the case was presented by a party. Litigants are represented in our courts by advocates of differing skills. Litigants are sometimes people of limited knowledge and perception. Occasionally, they mistakenly attach excessive importance to considerations of no real importance. In consequence, they may sometimes tell lies, or withhold the entire truth, out of a feeling that they need to do so or that the matter is unimportant or of no business to the court. This is not to condone such conduct. It is simply to insist that, where it is found to have occurred, it should not deflect the decision-maker from the substance of the function assigned to a court by law.

Reading through the transcript of the trial in the present case (described at unusual length in the joint reasons ) I am left with a reminder of the features of jury trials in days gone by. The advocate tackled the respondent with a view to damaging her credibility. She was a soft target - a former abattoir worker, of limited education, inarticulate, living in a country town, with an alleged medical condition one feature of which was its possible impact upon her powers of concentration and memory. In the old days of civil jury trials, parties would fight such cases with as much bluster and prejudice as they could respectively get away with. In this trial, considerable latitude was allowed. But, in the end, objectively at least, what did the attack prove? That the respondent, on isolated occasions, when cameras were present, was able to partake in some recreational activities? That she was able to walk briskly over three kilometres on a kind of outing while in Sydney attending a medical appointment? That she was not entirely truthful or clear about her property holdings or about her relationship with a de facto partner who was the father of her child? That she had horses which she rode more often than she admitted? And that she joined one night in the exuberance of a country wedding and drank alcohol there like most of the other guests?

Consideration of all the evidence : A rational approach to the respondent's entitlements would make full allowance for such evidence - and the dents that it made into her assertions of general fatigue. But it would also make sure that it had taken into full account the objective and consistent evidence in the case. These included the respondent's clear initial infection with Q fever, her protracted weight loss and other consistent recorded signs, and the extensive evidence about the well documented features of this particular infection (including extended fatigue during the phase following acute infection and ups and downs of symptoms of exhaustion thereafter). To reduce the case to a simple contest of credibility was to fall into the trap of irrationality and the risk of prejudice - turning the court's process into one of punishing the respondent for her forensic performance instead of evaluating the objective testimony in the context of all of the evidence called at the trial.

Such evidence included powerful, indeed uncontradicted, medical and scientific reports about the course which an infection with Q fever sometimes takes. Lawyers naturally like the objective proof provided by the results of serological tests. However, the problem with Q fever and its aftermath, according to the evidence, is that serological tests are not always conclusive except during the initial acute phase. To do justice to both parties, the decision-maker was therefore obliged to enter upon a more detailed analysis: not to treat the case as a kind of sport, where a player whose credibility is damaged is inevitably judged the loser. Engaging in this kind of tournament can be comparatively easy for skilled, repeat players. The obligation to assess the claim justly, according to the entirety of the evidence and according to law, is rather more difficult, tedious and time consuming. When our system of civil trials substantially replaced the intuitive decisions of lay jurors with the reasoned opinions of professional judges, a higher standard of manifest rationality was required. The elements of the game were replaced by a more serious evaluation of the evidence and the clear demonstration of just and lawful outcomes. Courts of Appeal reinforced this change by their statutory duty to conduct appeals by way of rehearing - precisely as the Court of Appeal undertook in this case. It is a mistake to apply to the appellate reconsideration of judicial reasons the extreme restraint that was voiced in earlier times in appeals from judgments based on jury verdicts .

I find it impossible to believe that the respondent's counsel at trial and on appeal, addressing the flaws in his client's own evidence, as demonstrated by the appellant - did not in the end rely on the strengths of the case. These were the initial infection, the objective recorded signs and the medical and scientific evidence about chronic fatigue following infection with Q fever. I do not accept that, at trial or on appeal, the respondent rested her case entirely upon the claim of acute Q fever infection , to the exclusion of her post exposure fatigue . If counsel at trial were as experienced as the joint reasons suggest , it would have been irrational and irresponsible in the extreme to have taken such a course. The serological tests did not support it. Plenty of other evidence which had been tendered, showed the pattern of the Q fever condition as sometimes following the precise course recorded in the respondent's case - years of chronic fatigue syndrome, with ups and downs in the severity of the symptoms.

The difference between the approach that I favour and that contained in the joint reasons turns on a possible misunderstanding in those reasons about the meaning of the word "infection". An "infection" may be acute. But, in its later stages, it may be post acute or chronic. It is still, in a sense, an "infection". The "chronic fatigue syndrome" is not an entirely separate and different medical condition just because it is given a different label. It remains a consequence of the initial "infection". The sources of some infections remain in the body although they may be such as to render their presence undetectable to currently available serological tests. According to the evidence, this is the case with Q fever. Yet, instead of dealing with the respondent's condition on this footing, a simple way out was taken. She told some lies; so she was disbelieved and lost her case. The hard task of evaluating all of the medical and scientific evidence and weighing it as against the lies, was thereby avoided. A judge's assessment of credibility (as if that were a universal phenomenon) was allowed to trump rationality , involving the meticulous examination of all of the evidence viewed as a whole.

My approach may make me a "purist" in the eyes of those who hold different views . If that be so, I must live with the epithet. I gladly do so because I am not alone in my opinion and because purity in the stream of law and justice is the responsibility of this Court, not something to be despised.

Conclusion - no error shown : It follows that Heydon JA was substantially correct in his analysis and correct in the conclusions that he and the Court of Appeal arrived at. That Court was right to dismiss the superficial approach accepted at trial and urged on appeal by the appellant. Correctly, Heydon JA assumed the burden of analysing the whole of the evidence objectively, paying proper regard to the primary judge's findings on credibility. No error has been shown to authorise disturbance by this Court of the orders of the Court of Appeal made on the basis of Heydon JA's painstaking analysis. Order

The appeal should be dismissed with costs. CALLINAN J. The facts

The respondent contracted Q fever in the course of her employment by the appellant at its Inverell abattoir in 1994. Q fever is an infectious disease which, as a causative organism known as Coxiella burneti, is prevalent in cattle. It is transmissible to human beings and causes symptoms similar to influenza. The respondent's work exposed her to a serious risk of contracting the infection. The appellant negligently failed to eliminate or reduce that risk by arranging for the respondent to be vaccinated before requiring her to do the tasks that caused her to be infected. The appellant conceded negligence shortly before the trial began.

Q fever typically involves an acute stage that may in turn evolve into a chronic infection demonstrable serologically. In 10-20% of cases, Q fever may persist as a post Q fever chronic fatigue syndrome. This is a clinical diagnosis, based very largely, but perhaps not entirely exclusively on subjective symptoms, which cannot be readily disproved by any objective test.

The respondent was first seen by Dr Hall of Inverell on 27 July 1994. She was diagnosed as suffering from Q fever after pathological tests had been performed. Subsequent serological tests carried out in February 1995 revealed that whilst the respondent had an acute episode of Q fever, she did not have chronic Q fever infection. The respondent was seen by a large number of medical practitioners and presented to all of them, with one exception, a consistent history of continuing headaches, nausea and fatigue. If the respondent's claims of symptoms were true it followed that she suffered from post Q fever chronic fatigue syndrome and was significantly disabled. The procedure adopted by the parties

It is convenient to deal with other relevant factual matters in discussing the trial and the appeal to the Court of Appeal. The problem in this appeal arises, because, with no doubt, the best of intentions, but quite mistakenly, the parties chose not to call, or require for cross-examination any doctors who could speak to the respondent's true condition, the controversial matter of substance that the Court had to decide. How anyone could possibly imagine that that controversy could satisfactorily be resolved without testing the conflicting opinions in cross-examination eludes me. There will be cases in which it will be possible and convenient to rely simply on written materials, but that reliance will not only be misplaced but also will produce much less expeditious and efficient outcomes when a material controversy has to be resolved. This is so despite the enthusiasm in some quarters for radical change to longstanding orthodox procedures. The appellant's submission that the way in which the trial was run was a demonstration of pragmatic efficiency is certainly not made out. The presence of the case in this Court is sufficient to make that point. A further problem that frequently, and here graphically presents itself, when written medical reports only are relied on, is the status of the plaintiff's statements to the doctors: are the reports, because of an apparently unqualified acquiescence in their tender, evidence of the truth of the plaintiff's statements reproduced in them; are the statements only admissible as admissions against interest to the extent that they are prejudicial; to what extent do the relevant provisions of the Evidence Act 1995 (NSW) particularly Pt 3.3 thereof bear upon these statements? No attempt was apparently made by anyone to address these questions before or during the trial. As will appear, Heydon JA in the Court of Appeal was alive to the problem which the course adopted presented. Absence of reference by the parties to the Evidence Act or to the status of the contents of the reports at the trial means that the questions that I have posed simply remain unanswered. The trial

The respondent did give oral evidence however and was extensively cross-examined at the trial. To demonstrate the division between the experts, and therefore the nature and magnitude of the serious controversy at trial, it is sufficient to refer to extracts from reports made by only two of the medical experts. The respondent relied on several doctors including Professor Boughton. On 23 February 2000 he wrote a report generally consistent with other reports that he made from time to time, and in which, among other things he said: "The sequence of events according to the history given to me, is that of a previously fit girl active in many sports, positive and hardworking, who held two jobs simultaneously, was entrusted with a responsible task at the abattoir because she was conscientious; she acquired preventable Q fever in a highly dangerous situation, as a result of which her health has been severely affected since 1994. To become depressed, anxious, upset, angry, and demoralised, is a perfectly normal reaction to such marked incapacity. I see no abnormal or inappropriate response to this personal disaster by this girl. From what can be determined, this girl was psychologically normal prior to the Q illness; the latter would be wholly responsible for her current psychological and physical status. ... Finally there is no getting away from the fact that the severe disability this girl has suffered for the past six years was completely preventable by immunisation against Q fever using the then available Q vaccine. In view of the extremely high risk of infection by Q to which she was exposed in the course of her work, she should have been vaccinated before ever being allowed into the slink room."

One of the doctors upon whom the appellant relied was Dr Sutherland. He gave this opinion in writing on 6 September 1999: "Mrs Dixon developed an occupationally acquired Q fever infection in mid to late July of 1994. There is no history of an incapacitating illness at that time or shortly thereafter, and so it must be assumed that she suffered a relatively minor acute illness, or perhaps even an asymptomatic infection, from which she recovered rapidly. Mrs Dixon conceived in late September or early October of 1994, and this event, and the subsequent need to deal with the pregnancy, coincided with the onset of her current complaints. The temporal association is compelling, and if any attempt is made to attribute Mrs Dixon's complaints to some adverse life event, then the link would be with events surrounding the pregnancy, and not with the seemingly unremarkable acute Q fever infection some three months earlier. A consultant psychiatrist has deemed Mrs Dixon to show evidence of abnormal illness behaviour, in the form of a somatoform disorder, and this would appear to be the most reasonable explanation for her symptoms. On the basis of all the evidence available, there is no reason to link Mrs Dixon's current complaints and claimed incapacity with the acute Q fever infection she sustained in the course of her work in about July of 1994, nor with any other aspect of the nature and conditions of her work at Northwest Exports abattoir in Inverell."

The respondent submitted that Dr Sutherland's conclusions were shown to be wrong because they failed to take account of a recurrent disabling bout of illness after the termination of her pregnancy. Even so, the fact remains that the widely divergent pre-trial opinions provided a clear indication of what the battle ground at trial would be.

The respondent's action was commenced in Newcastle in August 2000 by Newman J and adjourned to Sydney in October 2000. In addition to the plaintiff's oral evidence, the trial judge heard evidence from a number of lay witnesses called on her side. His Honour was also shown video films one of which depicted the respondent "dancing and socialising and drinking" at a wedding at a time when she claimed to be suffering disabling symptoms. Another video film showed the respondent at an equestrian competition in which her daughter was competing, and a third showed the respondent moving actively around Sydney.

The trial judge formulated the issue before him in these terms: "When examined in chief, the plaintiff gave evidence which was consistent with the history of continuing symptoms which she had given to medical practitioners. Because the continuing symptoms are entirely subjective the only issue which arose in the case was whether the plaintiff's account of her continuing symptoms was credible. In other words has the plaintiff established on a balance of probabilities that she has had the symptoms of which she has complained."

His Honour, after summarizing the medical evidence, made a number of findings adverse to the respondent's credibility and to her claims of fatigue and lethargy. There was an evidentiary basis for findings of that kind. His Honour concluded that matters raised in cross-examination of the respondent and her responses to them effectively destroyed her credibility. He said this: "While senior counsel for the plaintiff submitted that the matters going to credibility were not put to any doctor in cross-examination (all doctors' evidence was received by way of reports) the fact is as I have earlier mentioned, that the on going symptoms from February 1995 are matters for which no objective tests can be performed by medical practitioners to ascertain whether the symptoms in fact exist. I have formed the view that from the time when the plaintiff had recovered from the acute episode of Q fever, which the serological evidence indicates was in February 1995, I am not satisfied, on a balance of probabilities, that the plaintiff has in fact suffered from the symptoms which she recounted to medical practitioners and in chief to this Court."

Because the damages which would otherwise be awarded fell short of the relevant threshold for which the Workers Compensation Act 1987 (NSW) provided, the trial judge dismissed the respondent's action with costs. The appeal to the Court of Appeal

The respondent appealed to the Court of Appeal of New South Wales (Beazley and Heydon JJA and Davies AJA). Beazley JA and Davies AJA concurred in the judgment of Heydon JA, that the appeal be allowed and that there be an order for a new trial. It was accordingly unnecessary for the Court of Appeal to deal with an application that the respondent had made to adduce fresh evidence in the appeal.

In the Court of Appeal the respondent relied in particular on three matters involving the interpretation of passages in the respondent's doctors' reports. First, it was contended that the trial judge had overlooked a reservation in one report (Professor Boughton, 12 December 1995): "These results ... should be repeated ... in (say) six months time to confirm ..." and had thus erred in finding: "The serology testing which established that the plaintiff did not have a chronic infection was carried out in February 1995. In a report dated 8 May 1995, Dr Thatcher stated that the results of the serology carried out showed that she has had Q fever but does not have a chronic infection. Again there is no evidence before me to suggest that Dr Thatcher's diagnosis in this regard was other than correct. Indeed, the evidence of all medical practitioners on this topic coincides [with] the views expressed by Dr Thatcher."

The second was a criticism of the conclusion of the trial judge that I have quoted by reference to various passages in several of the medical reports said to be inconsistent with the trial judge's finding that there was no serological evidence of chronic infection. These passages, it was contended, were at variance with his Honour's finding that the respondent had recovered from Q fever infection by February 1995.

The third proposition was that in reaching his assessment of the respondent's credibility, the trial judge failed to take into account the opinion of Professor Boughton that impairment of concentration and memory made the eliciting of a reliable history difficult.

The respondent's grounds of appeal involved no express complaint about the trial judge's adverse findings as to her credibility. As Heydon JA said, "[t]he plaintiff did not in terms assault [the credibility findings] by saying that they should never have been made". His Honour added: "Nor did the plaintiff's counsel deny that she had been shown in cross-examination to have been wrong in some respects, and to have gilded the lily. The plaintiff's submissions were largely not explicit. Often they went no further than to offer half-hints at implications which the court ought to make for itself. On occasion the submissions became subliminal. Taken by themselves, the arguments just described, so far as they were explicit, do not assist the plaintiff sufficiently to justify ordering a new trial, though if there were a new trial, their full development at that trial might well assist the plaintiff's cause. But those arguments are material, taken with other difficulties, in identifying certain weaknesses in the trial judge's reasoning now to be discussed."

Heydon JA was of the opinion that the trial judge had erred in these respects: "First, the trial judge's construction of the evidence which led him to the view that a conclusion about the plaintiff's condition could only rest entirely on her subjective symptoms was wrong. Secondly, he failed to make any findings about the existence of symptoms which were observable by the medical experts for themselves and which they took into account. Thirdly, he failed to assess the totality of the medical evidence in the light of the symptoms observed by medical experts which did not depend on the plaintiff's own history and in the light of their professional experience and assessment of the probability that she was suffering from Q fever chronic fatigue syndrome."

His Honour was also of the opinion that the trial judge had been overly critical of the respondent's evidence and that, on his analysis, several of the discrepancies and inconsistencies in it were no more than trivial or readily open to much less unfavourable inferences. Some of these Heydon JA thought, were explicable by a matter to which the trial judge did not advert, that the respondent's condition fluctuated: she had good days as well as bad ones; and that under the impetus of impulse, or necessity, or a desire for distraction or pleasure, she might exert herself in ways for which she might pay afterwards. Perhaps the strongest criticism of the primary judge's approach was that he disregarded a number of objective pieces of evidence confirmatory of the respondent's complaints. It is convenient to deal with these in discussing the appellant's submissions in this Court which were equally critical of the approach of the Court of Appeal to them.

The matters to which Heydon JA referred, were sufficient in his opinion to cast serious doubt upon the correctness of the result of the trial and the process by which it was reached. The most important of these, his Honour clearly thought was the primary judge's failure to deal adequately and satisfactorily with all of the medical evidence. In the circumstances, there were sufficient doubts and questions to make it just to order a new trial on the basis that, in the language of Pt 51 r 23(1) of the Supreme Court Rules 1970 (NSW), it appeared that some substantial wrong or miscarriage had taken place in that the respondent's claim had not been properly considered . The appeal to this Court

Very much at the heart of the appellant's submissions in this Court was the contention that this, being a case of subjective symptoms, was one in which it was for the respondent to convince the Court that the symptoms were both genuine and caused by the appellant's negligence: and, in consequence, the persuasiveness of the respondent's evidence and demeanour in the witness box, particularly in the absence of oral evidence from doctors, were critical and matters in respect of which the primary judge had an overwhelming advantage over any appellate court.

The appellant submitted in this Court that it is settled that the respondent should be bound by the conduct of her trial , and that the Court of Appeal erred in ordering a new trial on the basis of its own detailed analysis of the transcript and exhibits, and identification of errors in the approach of the trial judge, undertaken without reference to the parties, or by reference to an appropriate ground of appeal. It is possible to deal with the latter part of this submission immediately. The respondent's grounds of appeal to the Court of Appeal necessarily required that Court to look at all of the evidence and to compare the respondent's oral evidence with the medical evidence in the reports at length.

The third submission of the appellant is that the Court of Appeal effectively treated the history narrated by the respondent to various doctors as evidence of the fact, or at least, when coupled with the absence of comment on it by the doctors concerned, as a circumstance from which it could draw inferences of fact. The appellant pointed out that no reliance was placed by the Court of Appeal, or by counsel there on s 72 of the Evidence Act , the effect of which could be to make admissible the respondent's out of court representations to prove the truth of assertions in them - but only if, and to the extent that, the representations were "contemporaneous". Very little, if any of the material, it was submitted, could seriously be regarded as answering this description, the word used by the doctors, "history", being a better description of its true nature. Nor, it was pointed out, was s 60 of that Act relied on by the respondent at the trial.

The appellant makes the point that even if the doctors had been called to give evidence they would not have been permitted to express any opinions about the severity and genuineness of the respondent's complaints, or indeed about her honesty generally.

A further submission was that the respondent's appeal to the Court of Appeal was governed by the decisions of this Court in Warren v Coombes and Abalos v Australian Postal Commission , and that State Rail Authority of New South Wales v Earthline Constructions Pty Ltd (In liq) which, because it was a very special case, had nothing to say about this one.

Much of the appellant's time in oral submissions was taken up with an analysis and criticism of the reliance that Heydon JA placed upon a number of the objective symptoms and history of the plaintiff which in his Honour's opinion the primary judge had disregarded. With respect to this, and as Heydon JA pointed out, before identifying the relevant objective factors, difficulties do arise whenever, and as here, the experts did not always differentiate between the assumed facts and the opinions advanced with the result that the trier of fact (and, it follows an appellate court considering factual findings) will be confronted with problems in deciding how far the opinions can stand in the light of the particular facts found.

The first objective piece of evidence to which Heydon JA pointed was the plaintiff's claimed loss of weight of some three or four stone. She was not cross-examined about this. The appellant submitted that this was not an objective piece of evidence as no doctor had weighed the respondent before she became ill. In my opinion, it was a relevant matter of some importance and deserving of some consideration by the trial judge.

The second objective factor was muscle and abdominal tenderness as to which, the appellant submitted, the courts had only the respondent's word. I do not think that this is so. As Heydon JA pointed out, the presence of tenderness was recorded by a doctor who palpitated various parts of the respondent's body and who also found some slight epigastric tenderness. Other doctors on other occasions made similar observations following physical examination. The respondent was examined by experienced doctors who no doubt made their examinations in such a way as to distinguish between real and actual tenderness, and any feigned or psychosomatic conditions. Heydon JA was right to regard the tenderness as supportive to some extent at least of the respondent's claims, and again a matter worthy of consideration.

The third matter was the respondent's appearance and mood. Her paleness, apparent tiredness and depression were observed and commented upon by lay persons as well as by three doctors. Accordingly, Heydon JA was not wrong to regard these matters as relevant and confirmatory.

The fourth matter was the respondent's inability to recollect and concentrate. I would accept that there must be a large element of subjectivity about this matter but the fact is that one doctor, Robertson, formed an impression of at least some minor impairment of the respondent's memory. That does provide some slight support for the respondent's assertions. If that matter stood alone, it would not justify intervention by the Court of Appeal.

I am also in some doubt as to the objective quality of the next matter upon which Heydon JA relied, an impression of sincerity on the part of the respondent formed by a number of doctors. That matter I would place in the same category as the one that I have just discussed.

The sixth matter however, is of some real relevance, and that is the fact that one doctor prescribed a course of antibiotics for the improvement of the appellant's condition, and she took the antibiotics as well as generally acted on such advice as the other medical advisers gave her from time to time.

Heydon JA pointed out that there could be no question that the respondent did contract acute Q fever. There was good statistical and experiential evidence that about 10-20% of sufferers from Q fever, particularly acute Q fever, do go on to develop chronic fatigue syndrome which can last for years. Professor Lloyd thought the respondent's case, one of "a typical post infective fatigue syndrome following serologically documented Q fever". This last matter is a significant one. Taken with the other matters, the typical nature of the complaint does amount to a real matter of substance to which the trial judge ought at least have given some consideration.

A complete reading of the reasons for judgment of Heydon JA indicates that in addition to the several matters to which his Honour referred, he was left with an unmistakable impression that the respondent's case had not been given the consideration that it should have been: that in effect the respondent had been denied a full and balanced appraisal of all of the evidence in the case, and that had it been properly appraised, apparent discrepancies which Heydon JA thought minor ones only, would have been better understood and readily explicable. Not enough attention was given, Heydon JA thought, for example, to the unlikelihood that the respondent had resolved "to live a complete lie and maintain a consistent but wholly or largely mendacious story to be told to her family, her friends, her doctors, her lawyers and ultimately the court over a period of six years."

Before stating my opinion there is one further matter to which the respondent drew attention which I should mention. It is that some of the medical opinion advanced by the appellant at the trial was not fully informed by the fact that the respondent had suffered a recurrence of severe acute febrile illness at about the time that the appellant's doctors were saying that other events explained the respondent's subsequent condition of fatigue and general debilitation.

In my opinion, this appeal should be dismissed. It is unnecessary for me, in so holding, to decide whether the Court of Appeal was right to uphold the appeal under Pt 51 r 23(1) of the Supreme Court Rules made under s 124 of the Supreme Court Act 1970 (NSW) or whether the Rule was regularly and validly made under it, as, in my view, the errors at first instance required appellate intervention pursuant to s 75A of that Act .

As to the operation of that section, I adhere to what I said in Fox v Percy , and in particular to this : "To impose the test stated in Devries is, I think, to do what was said to be impermissible as long ago as 1920 , to elevate as a practical matter, the decision of a judge sitting alone to the level of a verdict of a jury. This is so even though judges are bound to give reasons and those reasons are required to be able to withstand scrutiny. The value of that scrutiny will be much reduced if a statement in the reasons that the demeanour of a witness has been determinative of the first instance decision, is effectively taken to be conclusive of the outcome of an appeal by way of rehearing. The vast majority of the cases tried in this country are tried by judges sitting alone and depend upon their facts rather than upon the application of complex legal principles. To impose an unduly high barrier, and not one sanctioned by the enactment conferring the right of appeal would be to deny recourse by litigants to what the Parliament of the State has said they should have. Judges are fallible on issues of fact as well as of law; sometimes they are obliged to work under a great deal of pressure, and sometimes they are denied a timely transcript. In the days when rights of appeal were first enacted, notes and transcripts were much less complete and reliable than they now are. And today courts of first instance, in some jurisdictions at least, rely heavily on written statements, certainly of the evidence in chief, the oral adducing of which might on occasions have been as, or even more revealing than, evidence adduced from an honest but inarticulate or nervous witness in cross-examination. Occasional errors of fact are bound to be made. No litigant should be expected to accept with equanimity that his or her right of appeal to an intermediate court is of much less utility because it goes to a factual error that can be explained away by a judge-made rule, than an appeal on a question of law: or that although the trial judge was wrong on the facts, there was no incontrovertible fact against which the judge's error could be measured."

It is true that in many respects, Earthline upon which the respondent relied, was a special case because there was documentary evidence on affidavit which went completely unchallenged, and which, inexplicably the trial judge rejected. But what their Honours (Gaudron, Gummow and Hayne JJ) said in and of that case, is of relevance and may be applied here : "It is true that the trial judge, in determining whether to accept the evidence of Mrs Page, was heavily swayed by his impression of her whilst giving oral evidence. However, this circumstance does not preclude a court of appeal from concluding that, in light of other evidence, a primary judge had too fragile a base to support a finding that a witness was unreliable ."

Furthermore, as the judgment of Kirby J explains, there are numerous cases in which, despite a strong impression which a trial judge may have formed, circumstances justifying interference by an appellate court may exist. I did not take his Honour to be saying in that case that the categories of cases for appellate intervention with respect to factual matters were closed.

Nor do I think that the observations that I made in that case are irrelevant to this one, that a failure to give the significance to particular evidence that it plainly deserves may, of itself, particularly when other indications of error, even only minor ones are present, will also warrant intervention by an appellate court .

Nothing that I have said should be taken to mean that deference, indeed a great deal of it, should not be accorded to findings of fact at first instance, particularly those based upon demeanour and impression. Nor should it be taken to mean that plaintiffs are not obliged to discharge the onus of proof of their damages. There will be very few cases in which the complaints and disabilities are not provable objectively, and in which the plaintiff has not by his or her own evidence persuaded the court of their existence, that an appellate court will be entitled to intervene.

This case is however one such case. The reasons why in summary I would dismiss this appeal are that, first, there were some quite significant objective pieces of evidence identified by Heydon JA in the Court of Appeal which should have been, but were not considered and weighed by the primary judge. Secondly, the Court of Appeal did not err in holding, as in effect it did, that the strongly adverse finding of the primary judge, that the respondent's evidence was unacceptable, had too fragile a basis to support it, particularly in the absence of a complete and careful examination of all of the evidence. Such an examination would have revealed, among other things the serious question concerning the aetiology of the respondent's illness to which the subsequent recurrence of a severe acute febrile illness was highly relevant. A further reason why the appeal should be dismissed is that, in my opinion the Court of Appeal properly carried out its function and duty, without discernible error, that s 75A of the Supreme Court Act requires it to perform.

An appellate court in a civil case is not bound in terms to review the evidence at first instance to see whether the result there is "unsafe or unsatisfactory" as it is sometimes obliged to do in a criminal appeal. But there may be cases, and this the Court of Appeal not erroneously took to be one, in which it is apparent that the whole of the evidence, and the relationships between the respective parts of it have not been adequately considered and analysed at first instance, with the result that the outcome of the trial is unsatisfactory, and requires that, pursuant to s 75A of the Supreme Court Act an appellate court intervene.

I would dismiss the appeal with costs. I would however make one further order with respect to costs and that is, whether the respondent succeed or not on the retrial which has been ordered by the Court of Appeal, the respondent should not have her costs of the first trial. Each party, by joining in conducting the trial in the way in which each did, contributed to its unsatisfactory outcome.