Rider v Christou and Trimithia Enterprises Pty Ltd

[1994] QCA 182

IN THE COURT OF APPEAL	[1994] QCA 182
SUPREME COURT OF QUEENSLAND

Appeal No. 154 of 1993

Brisbane

[Rider v. Christou]

BETWEEN:

MARGUERITE ROSE RIDER

(Plaintiff)	Appellant
-and-

PETER CHRISTOU

(First Defendant)	Respondent
-and-

TRIMITHIA ENTERPRISES PTY LTD

(Second Defendant) Respondent

The Chief Justice Mr Justice Davies Mr Justice Thomas

Judgment delivered 06/06/1994

JOINT REASONS FOR JUDGMENT DELIVERED BY THE CHIEF JUSTICE AND MR JUSTICE DAVIES. MR JUSTICE THOMAS DELIVERING SEPARATE DISSENTING REASONS.

APPEAL ALLOWED. Orders made below set aside. In lieu, judgment should be entered for the plaintiff against the defendants for $228,775.64 with costs of the appeal and costs of the trial including reserved costs, if any, to be taxed. Liberty to the parties to apply reserved in respect of the form of the order with submissions, if any, in respect of it to be lodged with the Court in writing within seven days.

Appeal No. 154 of 1993

CATCHWORDS:	NEGLIGENCE - Motor vehicle accident - acceptance by trial judge of conflicting evidence - Appellate Court's function - whether trial judge misused advantage in assessing witnesses - interference by appellate court with findings of fact and credibility - whether new trial should be ordered or judgment entered.
Counsel:	Mr R. Hanson Q.C. with him Mr R. Myers for the appellant. Ms M. Wilson Q.C. with her Mr A. Innes for the respondents.
Solicitors:	Clayton Utz for the appellant Dillons for the respondents.
Hearing Date:	21/02/94

IN THE COURT OF APPEAL

SUPREME COURT OF QUEENSLAND

Appeal No. 154 of 1993

Brisbane

Before The Chief Justice

Mr Justice Davies

Mr Justice Thomas

[Rider v. Christou]

BETWEEN:

MARGUERITE ROSE RIDER

(Plaintiff)	Appellant
-and-

PETER CHRISTOU

(First Defendant)	Respondent
-and-

TRIMITHIA ENTERPRISES PTY LTD

(Second Defendant) Respondent

The first defendant, who has also been accepted by the trial judge as a credible and

REASONS FOR JUDGMENT - THE CHIEF JUSTICE and MR JUSTICE DAVIES

Judgment delivered 06/06/1994

We agree with the reasons expressed by Thomas J. for concluding that the judgment entered below should be set aside but we have arrived at a different conclusion as to the orders which should be made by this court in consequence.

Since error justifying interference has been demonstrated, this court on appeal has a full formal power to act to correct it and this includes power to enter the judgment which should have been entered below: see Order 70, rule 11. The question is how the court should proceed in the very particular circumstances of this case.

In other cases if it should happen that fundamentally opposed versions are accepted by a trial judge or that a judge has not derived a relevant advantage from seeing and hearing witnesses, it may be appropriate to make an order for retrial. However, an appeal court is justified in adopting a particular and undoubtedly less usual approach when it is presented with a problem which calls for a special solution: c.f. Chambers v. Jobling (1986) 7 N.S.W.L.R. 1, and Whitehouse v. Jordan (1980) 1 All E.R. 650 in the Court of Appeal and (1981) 1 W.L.R. 246 in the House of Lords.

It is already almost ten years since the accident in the present case occurred. The plaintiff was accepted as an honest and reliable witness by the trial judge and nothing emerged in the course of proceedings below to throw any doubt on this assessment. On a retrial she will be able to throw no further light upon the events of the last moments before impact at a time when the approaching truck was almost upon her since she has no recollection which relates to that period. She has been consistent in maintaining that claim of loss of recollection.

convincing witness, has given versions essentially inconsistent with that of the plaintiff. has from time to time provided. With the lapse of time that has occurred, there is every reason to think that he could not now give a more accurate version of events if provided with an opportunity to do so on a retrial, and since he has already been given a very full opportunity to set out his account definitively and to explain the inconsistencies within his earlier versions, there is no reason to think that in these respects he will be able to do better. Justice would not seem to demand that he should, ten years on, be given a further opportunity to attempt to achieve some advantage from a retrial.

Accordingly, on present indications the benefits which might be expected to result from a new trial, are negligible if not totally absent, yet the cost and further delay involved in such a course would be substantial. It would be an excessive requirement to oblige another trial judge to produce a decision upon material which cannot be expected to be more illuminating than that which is already before this court.

The divergences in the first defendant's different accounts are so large and so critical that, confronted as they are by the plaintiff's contrasting account, they should be rejected. The evidence upon which the court could come to its conclusion on the remainder of the existing material may not be extensive, but the task is manageable.

There are these features: the concentration of glass on the outbound side of the road, the final position of the plaintiff's car on that side, the concession of the first defendant that when he stopped his truck its front right hand side was over the centre line, his subsequent action in parking on the outbound verge of the roadway and the fact that the first point of impact with the car was made with the right hand front corner of the truck although the main impact was with the truck's rear wheels occurring at a time when the first defendant indicated he was steering his truck to the right. These are all individually slight indications but taken together have a certain force.

In combination with the plaintiff's accepted evidence that she was proceeding properly on her correct side of the road until three car lengths from the truck, they are sufficient to lead to the conclusion that the truck must have drifted over the centre line to make first contact with the car and that if driven and kept further to the left, it would have avoided contact with the car altogether. It is necessary only to come to a conclusion on the balance of probabilities.

It should be concluded that the accident was caused by the negligent driving of the first defendant and that the plaintiff has not been shown to have crossed the centre line at any relevant time or to have exhibited negligence otherwise in the control of her vehicle. It should be held that the first defendant was wholly to blame for the collision.

The assessment of damages with interest, made by the trial judge, has not been challenged and from it, the refunds which he specified totalling $68,221.40 have to be made. A judgment for the plaintiff for $228,775.64 should be entered.

The appeal should be allowed and the orders made below set aside. In lieu, judgment should be entered for the plaintiff against the defendants for $228,775.64 with costs of the appeal and costs of the trial including reserved costs, if any, to be taxed. There should be liberty granted to the parties to apply, reserved in respect of the form of the order, with submissions if any in respect of it, to be lodged with the Court in writing within seven days.

IN THE COURT OF APPEAL

SUPREME COURT OF

QUEENSLAND

BRISBANE	Appeal No. 154 of 1993
Before The Chief Justice

Mr Justice Davies

Mr Justice Thomas

[Rider v. Christou and Ors]

BETWEEN:

MARGUERITE ROSE RIDER

(Plaintiff) Appellant

- and -

PETER CHRISTOU

(First Defendant) Respondent

- and -

TRIMITHIA ENTERPRISES PTY LTD

(Second Defendant) Respondent

REASONS FOR JUDGMENT - THOMAS J.

Judgment delivered 6th June 1994

This is an appeal against the decision of a trial Judge dismissing a plaintiff's action for damages for personal injury arising out of a motor vehicle accident.

The incident in question involved two vehicles travelling in opposite directions on the Centenary Highway on 21 December 1984. Because of the passage of time and the conflicting evidence that was led, it is difficult to tell on what part of the road the accident occurred. In the result the learned trial Judge did not base his decision upon objective data such as marks or deposits on the roadway. The decision was based upon acceptance of both the plaintiff's and the defendant's evidence. However, the plaintiff, who was seriously injured, had no memory of the last moment before the collision whereas the defendant gave an account fully favourable to himself. In this way, although accepting both witnesses His Honour found in favour of the defendant.

The main problem is that the evidence of the plaintiff whom His Honour found to be "an honest and reliable witness" cannot be properly reconciled with the defendant's evidence which His Honour accepted. His Honour also accepted, apparently without reservation, the evidence of an independent witness Mr Kuiper, and there are some further respects in which there are serious difficulties in reconciling this evidence with that of the defendant. These difficulties were not adverted to in His Honour's reasons. The determination was in the end based upon an acceptance of what the defendant said in the witness box (in contrast to what he said on other occasions).

Courts of appeal are very reluctant to interfere with decisions based on findings of credibility, but in this instance the question arises whether error is disclosed in the determination, such that it is appropriate to set aside the ultimate finding. This may be done only in very limited instances, including cases where the trial Judge makes findings inconsistent with established facts or where the advantage of the trial Judge has not been used or has palpably been misused. (Devries v. A.N.R.C. (1993) 67 A.L.J.R. 528, 531; Brunskill v. Sovereign Marine and General Insurance Co. Ltd (1985) 59 A.L.J.R. 842, 844). In order to consider whether this is a case calling for such interference, it has been necessary to examine the evidence in some detail.

The highway consisted of two lanes divided by a broken white line. The plaintiff was travelling in a Mazda sedan in a southerly direction and the defendant was travelling in a tip truck in a northerly direction. The bitumen strip was about 22 feet wide, with about 11 feet available in each lane. The truck was about 8 feet wide.

The plaintiff's evidence was that she drove in a normal fashion along the highway that morning. She was not aware of any distraction or of the occurrence of anything unusual prior to the incident. She was on the correct side of the road travelling at about 60kph in an 80kph zone following a white van that was about one and a half car lengths ahead of her. She saw the defendant's truck approaching in the opposite direction also on its correct side. That was the position when the truck was about three or three and a half car lengths away from her. She has no recollection beyond that point. Her account was consistent with what she told the police officers who interviewed her in hospital. As already mentioned, His Honour found the plaintiff to be reliable as well as honest.

The position with respect to the defendant's evidence is rather more complex. He gave divergent accounts on different occasions. His Honour acknowledged this, but in the end, accepted the evidence given by him in the witness box, observing that he was "credible and convincing", that there was "no significant variance between his accounts given at various times, when taken together," that he "was not shaken by cross-examination", and that "to invent such an explanation would have required the exercise of a great deal of devious creativity."

At the scene, having described his own speed at about 55kph and guessing the plaintiff's speed to have been about the same, his essential response to the police is contained in the following question and answer.

"Q.	What happened?
A.	It was as though she turned into me. She came onto my side of the road. It happened so fast I couldn't believe it was happening. I tried to avoid her by turning to my left but she hit."

Later, on the same day at the police station, he gave and signed a statement to the

following effect:-

"As I was travelling along the straight section of road, I noticed a car travelling in the opposite direction, gradually veer onto the incorrect side of the road and was travelling towards my truck. ... I then realised that this other car was still travelling directly towards my truck. I then swerved to the left and accelerated and tried to then go to the right to encircle this car which was then totally on the incorrect side of the road. This other car collided with the driver's side front corner of my truck and then along the body of the driver's side of the truck. I can't remember actually braking at the time of the collision. Although when I first saw this car veering onto the incorrect side of the road, I slackened off thinking that the other car was going to travel back onto the correct side of the road.

I immediately stopped my truck in the middle of the Centenary Highway. I then shifted the truck off the highway to allow the traffic to flow freely. After stopping in the middle of the road, I raced back to assist the driver of the other car."

The initial description of a sudden event has changed to a gradual one with a description of his own reactions including a slackening of speed after observing the other car coming onto the incorrect side. It is still however, at this stage, a story of a single progressive deviation by the other driver onto the wrong side of the road.

The next account was in response to interrogatories, and it includes supplementary corrections of his first answers. It is unnecessary to say anything of the divergence between the two sets of answers to interrogatories. The change was based upon a reconsideration of the point at which he started to slacken speed, and it probably does not matter much whether he started to do so 200 metres before the point of impact (as his original answer suggests) or 100 metres (as the supplementary answer suggests). His final answers suggest that he was travelling at 50kph until about 100 metres before the point of the accident, that his speed had dropped to 20-30kph at a point at about 50 metres from the point of impact, and that at the time of impact his speed was "no more than 20kph and probably a little less." He described the plaintiff's vehicle at the time of impact as on the bitumen entirely on the plaintiff's incorrect side of the road.

At trial the account was rather more elaborate. It is difficult to set out a concise summary as the evidence shows a tendency to diffuseness. The essential story was to the following effect. The other car turned into his path maybe 70, 80 or 100 metres away. He

slowed right down and

"was ready to come to a complete stop because I expected the car to go into the Golf Course ... I was almost to a stop ... when it was coming at me, it was veering at a slight angle ... so I thought, well, if I move over to the left, right over as far as I can ... it wasn't at a sharp angle that the car was coming at - I thought I would give it more room to get past ... I moved right over as far as I could and then the car turned. It was as though it was out of control. The car turned and headed straight for me again. ... What I tried to do is still give it more room to get past. What I was trying to do was encircle the car, the oncoming car."

"Q.	Doing an arc to the left?
A.	That's right."

(The description of the "encircling" movement is reconcilable with an intention on his part to pass to the left of the other vehicle and then come back onto the roadway. Neither counsel submitted that his evidence should be interpreted as indicating that he attempted any evasive movement to his right before the accident.)

"... As it was, I managed to avoid a head on smash.
... When the vehicles collided, I felt a ... very faint mark on my truck, you know,
faint feeling on the truck on the front. It didn't affect the ride in the cab or
anything, you know, throw me around or anything like that and I thought I'd
made it. I thought I cleared - just minor things, but then next thing, bang, the
real jolt came on the back of the truck. So much so, as it hit the cab went up
and down like that, because Volvo trucks are well sprung, and I felt the impact
even in the cab."

He came initially to a halt, on his account, "at an angle of about 60 degrees to the road, but the front bumper bar - maybe the front wheel - the front of the steering wheel onto the oncoming lane. The rest of the truck was way over on the other side."

He estimates that he moved only five or ten metres forward after the impact before coming to an initial halt in that position, that is to say with the front driving side corner of the truck slightly over the centre-line.

He dismounted from his vehicle, went to the other car, saw the injured woman, ran back to his truck, radioed his depot and told them to send the ambulance and police. Within about five minutes a police motor cyclist arrived and started to guide the traffic. The defendant's account continues:

"I said to him 'Look, do you want me to leave the truck here or will I move it out of the way to allow the traffic to flow through?' He had a quick - he had a look around. He said 'no you're right. Take it over there.' I said 'do you want me to leave it on my side of the road or put it over there on the wider side where the turf farm is.' He said, 'put it over there', so that it is where I put it."

He described the initial movement of the Mazda as a "sharp turn as if you were going to do a u-turn." In cross-examination he gave further details including the fact that he was almost stopped before impact, that his movement was so slow that he had to put the truck into low gear to start off again, and that from almost a stop he accelerated to try to get out of the way. He described the alleged double movement of the oncoming vehicle as follows:-

"I took evasive action as best I could to avoid a head on smash and the car was
not happy with doing that, it had to turn in against me again."

It is difficult to agree with the learned trial Judge's view that there was "no significant variance between his accounts given at various times, when taken together". The more significant of the variances are those between a sudden incident and a gradual one in which he had time to perform various evasive manoeuvres; between a story of a single progressive deviation by the other driver and a double manoeuvre by her; and between a story of his own gradual deceleration to 20kph at time of impact and one of having virtually stopped prior to impact and having then accelerated to try to get out of the way. It is true that all versions lead to the accident happening on his correct side, but each of them achieves this result by means of different stories.

Putting to one side any reserve that might be felt about accepting an oral version eight and a half years after an accident inconsistent with earlier recorded versions, I consider His Honour was in error in thinking that there was no significant variance in the defendant's accounts. Further, if the plaintiff's evidence is to be taken as reliable, the defendant's account as accepted would be quite impossible. The plaintiff's story lacks only the last fraction of a second, remembering the combined approaching speeds. These things could not have occurred within the three and a half car lengths that separated the two on-coming vehicles. This is not based upon slavish mathematical reconstructions. On a broad level as well as upon any more precise analysis the two stories cannot both be right.

I now turn to the evidence of damage to the vehicles and of marks and deposits on the roadway. The damage suggests that the major impact occurred between the car and the rear rims of the wheels and perhaps some housing on the driving side of the tip truck. This seems to be common ground. The car was very extensively damaged, with particularly severe damage to the front driving side. The plaintiff's vehicle ended up mainly on the gravel on the eastern side of the road, that is to say on her correct side, with only a small part of the rear of her car on the bitumen.

After stopping his vehicle the defendant removed it and parked it on the same side as the plaintiff's vehicle, claiming that he did so with the permission of a policeman. There are reasons to doubt at least this part of his evidence, as it appears inconsistent with the evidence of Mr Kuiper, whose evidence was also accepted by the learned trial Judge.

In a case such as this physical indicia such as marks on the roadway, deposits of glass and other substances may be useful. Sometimes they show facts which tend to support one story or the other; sometimes they may afford prima facie evidence of the area in which the impact occurred; sometimes they may be virtually conclusive; and sometimes they are quite unsafe as a basis for any inference even of probability. In the present case the best evidence of such indicia was provided by Mr Kuiper, a householder who lived nearby. He heard the accident and it would seem was the first person upon the scene. He estimates that he arrived about three minutes after the accident. The learned trial Judge considered him to be a convincing witness and has accepted his evidence.

He described deposits of glass over an area which lay principally on the eastern side of

the roadway, that is to say, on the correct side of the road from the plaintiff's point of view. and the rear of the plaintiff's vehicle which had come to rest almost completely on the gravel shoulder on the eastern side of the road.

The police evidence and photographs do not provide much assistance. The relevant police officers who took the photographs and measurements did not arrive until an hour after the accident, by which time a heavy stream of traffic moving in both directions along the available roadway had undoubtedly disturbed any deposits and compromised any marks that may have been present. The police noted two skid marks on the western side of the roadway, but it is common ground that these cannot safely be related to the accident. There was by that time no observable concentration of glass in any particular spot and this seems confirmed by the photographs. One of the photographs shows some presumably metallic pieces slightly on the western side of the roadway, and also some white marks on the surface of the bitumen but it is difficult to characterise them. The police report (and Mr Gray in evidence) express the view that the point of impact was at a point slightly on the western side of the centre-line, but no factual justification or explanation was suggested for the opinion, and His Honour rejected that view. In the end, although there are the photographs taken more than an hour after the event which show some marks and deposits, they do not provide much assistance and it would seem that this was the view taken by the learned trial Judge. In the end the most significant evidence of deposits is that of Mr Kuiper who described a clear and substantial deposit of glass on the plaintiff's side of the roadway, with only minor amounts spread elsewhere.

All the glass came from the plaintiff's vehicle. No glass part of the truck was broken.
Evidence was received from an engineer, although with some reservations, which

expressed the opinion that the accident had happened close to the middle of the road. The learned trial Judge indicated in his reasons for judgment that he did not find the evidence helpful and that he did not rely on it. Matters such as the possible shielding effect of the spread of glass by reason of the side of the truck, and the natural rotational forces generated by an essentially straight-on collision between the corners of vehicles were discussed, but in the end the conclusions to be drawn from the objective facts, and from the combination of primary facts accepted from the witnesses, would seem to have been a matter for the court rather than one for expert evidence. It is not necessary to pursue this point further, as His Honour did not rely upon it as expert evidence. However, it might be thought that Mr Kuiper's observation of the glass deposits raises an inference in favour of the accident having occurred substantially on the plaintiff's side of the roadway. I shall return to this point in due course.

The defendant denied that there were deposits of glass as described by Mr Kuiper. His Honour did not advert to this conflict, but his acceptance of the fact that the deposits were present as described by Mr Kuiper necessarily involves rejection of this evidence by the defendant. Another instance of conflict between the defendant's evidence and that of Mr Kuiper would seem to be the defendant's explanation for moving his vehicle, namely that a policeman agreed he should do so. When Mr Kuiper arrived, only three minutes after the collision, the defendant's truck had already been removed and parked at the side of the roadway. The times given by the defendant and his description of his activity after coming to a halt do not reconcile with this. Apart from the defendant's own time estimate, it seems highly unlikely that he could have stopped, gone to the Mazda and checked it, come back to his vehicle, radioed his office for assistance, waited for the police to arrive (which he said was about five minutes later), observed the policeman guiding traffic, conversed with that policeman and obtained his leave to move his vehicle, and moved and parked it, all before the arrival of Mr Kuiper. Mr Kuiper ran by the shortest route from his home to the scene of the accident immediately upon hearing it. This particular unreconciled inconsistency between the defendant and Mr Kuiper might not mean very much standing alone, but is one of several matters which cause real concern in a determination that in the end proceeded entirely upon an assessment of credit.

In my view it is very difficult if not impossible to reconcile the evidence of the defendant with that of Mr Kuiper on two matters that are certainly more than peripheral; and it is logically impossible to have accepted the evidence of the plaintiff as truthful and reliable and also to have accepted the version of the defendant. In addition to these matters, the inconsistencies in versions given by the defendant at different times are by no means minor and would of themselves call for considerable caution.

For the above reasons I have concluded that error is demonstrated in the determinative process. The reasoning is flawed and the advantage of the trial Judge in choosing between witnesses has been lost. The present determination cannot stand, and the judgment must be set aside.

It is more difficult to decide upon the course that should be taken upon setting aside of the findings of the primary Judge. If the materials permitted the substitution of a correct finding for an incorrect one, or if the matter could be dealt with by drawing inferences which ought to have been drawn, I would not hesitate to make findings upon the evidence in the record. Appeal courts commonly, indeed normally, will reassess damages when an error is disclosed in the original assessment (Emery v. Wilson 48 A.L.J.R. 131); and whether the issue relates to liability or damages, courts of appeal are conscious of the advantage of avoiding the expense and delay of a retrial when it is necessary to set aside a determination by reason of error. When the primary facts are not in doubt, there is little difficulty in doing so, as the appeal court is in as good a position to draw inferences as the primary tribunal (Warren v. Coombes (1979) 142 C.L.R. 531). Equally, where the primary court has expressed a view upon credibility, the appeal court will respect that view in identifying the primary facts and inferences upon which it should act.

What however is the position in the comparatively rare case such as the present where the primary Judge's view on credit is found to be erroneous, so that the whole exercise of assessment of credit may be thought to have miscarried? In the present matter there is no safe starting point by which one or another witness may still be said to have the benefit of the primary Judge's acceptance. In holding that the primary Judge could not logically find that the plaintiff was honest and reliable at the same time as accepting the defendant's version of the accident, I cannot safely say that the error lay in the acceptance of the defendant or of the plaintiff. Perhaps the weight of the evidence might persuade many if not most tribunals to reject the evidence of the defendant in the present circumstances, but I cannot say with certainty that the defendant had to be rejected and the plaintiff accepted.

If the acceptance of the defendant could discretely be set aside thereby leaving intact the evidence of the plaintiff and Mr Kuiper and such other evidence as might be thought to be reasonably capable of acceptance, the task would be straightforward. The difficulty lies in the circumstance that the entire fact finding process has failed, and there remains a body of evidence, containing actual and potential conflicts, which would need to be analysed without the benefit of hearing and seeing the witnesses.

Many instances may be found of appeal courts finding error in determinations by trial Judges and substituting their own findings. Conversely many instances may also be found of the High Court holding that the substitution was erroneous, and that the findings of the trial Judge should be restored (see for example O'Neill v. Chisholm (1973) 47 A.L.J.R. 1; Chang v. Chang (1974) 48 A.L.J.R. 362; Hicks v. Roberts (1977) 16 A.L.R. 466; Abalos v. Australian Postal Commission (1990) 171 C.L.R. 167). These decisions however proceed on the basis of which findings and inferences were the correct ones, and they do not deal with the question whether, upon setting aside the findings of a trial Judge, the appeal court should substitute its own findings or order a retrial. A problem similar to this did arise in Pearlow v. Pearlow (1953) 90 C.L.R. 70. In that matrimonial case the trial Judge considered the plaintiff a person of low credibility and in the end exercised his discretion against granting a decree. The High Court was of the view that the discretion was exercised on inadequate materials and that the decision had miscarried. Dixon C.J. concluded:

"This is not a case where this Court ought without more to proceed to exercise a discretion of its own in lieu of that exercise by the learned Judge. The case is one for a fresh investigation carried further into the possible grounds for a discretionary refusal of a dissolution. The proper remedy is a new trial or rehearing of the action."

A robust example of substitution of findings of fact by an appeal court may be found in the majority decision (Kirby P. and Samuels J.A.) in Chambers v. Jobling (1986) 7 N.S.W.L.R. 1. The only real issue in the case was who was driving at the time. Both the plaintiff and the defendant were in the vehicle at the time of the single vehicle accident. There was one eye witness who gave oral evidence (six years after the accident) that the plaintiff had been driving. His contemporaneous statement to the police however did not say this and did not suggest that he had observed who had been driving. The trial Judge accepted his evidence, passing the somewhat negative observation that there was nothing in his demeanour which would enable him to disregard the evidence. There was a body of other circumstantial evidence which tended to suggest that the defendant had been the driver.

In this situation the majority took the view that the trial Judge had acted not so much upon an assessment of credibility as upon objective factors such as comparison, reconciliation and evaluation of the weight to be afforded to conflicting evidence. Assuming that that correctly characterises the approach of the trial Judge in that case, it could be seen as a case in which the appeal court was virtually in as good a position as the trial Judge, and there could be little objection in principle to the substituted findings. If however the elimination of the trial evidence of the eye witness, and the finding that the trial Judge should have preferred the inconclusive contemporaneous statement to the evidence actually given can be seen as usurping a function that should be performed only after seeing and hearing the witnesses, one would hesitate to act similarly.

Chambers v. Jobling should be interpreted similarly to Taylor v. Johnson (1982-1983) 151 C.L.R. 422, 426, where the trial Judge's remarks were taken as indicating that there was nothing to be gained from seeing or hearing the witnesses.

"The first question in the appeal is whether the New South Wales Court of Appeal was, in the circumstances, entitled to substitute its own conclusions as to the knowledge, state of mind and motivation of Mr Taylor for those of the trial judge. The answer to that question depends upon the meaning to be given to a comment of the trial judge that, subject to one presently immaterial qualification, 'there was nothing in the demeanour of any of the witnesses which would lead one to conclude that any of them was doing other than endeavouring, to the best of his or her ability and recollection, to tell the truth' and that, in the result, he was 'left to determine the matter upon the balance of probabilities'. In the view we take, the members of the Court of Appeal were correct in accepting his Honour's comments as being intended to convey that, in a position which he had derived no significant assistance from observing and hearing the witnesses give their evidence, he had resolved the matter by analysis of the evidence before him. That being so, we consider that the members of the Court of Appeal were correct in taking the approach that, subject to giving due weight to the conclusions of the learned trial judge, they were entitled to determine the matter for themselves."

In short when it is clear that the trial Judge's advantages are insignificant, the appeal court may proceed to act upon its own view, giving due weight to the conclusions of the trial Judge.

The High Court (on 13 February 1987) refused special leave to appeal from the decision in Chambers v. Jobling, being of the view that the principles of appellate review were not in doubt, and being unpersuaded that the Court of Appeal had failed to give proper weight to the advantages possessed by the trial Judge. It should not be inferred from this that there has been since 1987 a general encouragement to appeal courts to substitute findings of primary facts without the benefit of an assessment of credibility. To do so would be to intrude unduly into the trial process, and to underestimate its value.

Although the "trial Judge advantage" in the present case has so far miscarried, the Court of Appeal would trespass beyond its legitimate function were it to hold that the defendant is to be disbelieved and the action is to be determined on the other materials as though he had not given evidence.

Miss Wilson Q.C. for the defendant submitted that if it were concluded that the evidence of the defendant should be rejected, the court would be left in the position of not knowing how or where the accident happened, in which case the plaintiff must fail.

(Nesterczuk v. Mortimore (1965) 115 C.L.R. 140.) In my view there would remain a sufficient body of evidence to enable a court to infer that the collision occurred on the plaintiff's side of the roadway. In that event, in the absence of satisfactory explanation from the defendant, negligence could be inferred against the defendant. But the question remains whether this Court should now draw these inferences.

If the plaintiff's evidence is accepted, and the defendant rejected, the preponderance of other evidence in the case would support a finding in favour of the plaintiff. Her evidence, it will be remembered, has her on the correct side of the road up to what might be described as a fleeting moment before the collision. Whilst it is possible that the glass was deflected to that place by the side of the truck from the point of impact on the other side of the road, the absence of a trail or connection between any point on the other side of the road and the area of the deposit goes against this being a probability. The defence version fails to support the theory of deflection of glass generally to the right because that version has the truck being steered to its left at the moment of impact. However these considerations suggest no more than that the plaintiff will be in a strong position upon a retrial.

Our system contemplates that there is generally an advantage in resolving conflicts of evidence at trial level. It is true that the advantages of a further trial in the present matter are less than usual. More than nine years have elapsed since the accident. The plaintiff does not pretend to be able to remember the last vital moment. The defendant's statements on various occasions are all recorded. They contain inconsistencies on fairly important points in the narrative, though it is true that he has always claimed that the accident happened on his correct side of the roadway. Notwithstanding my view that the objective factors in assessment of credit favour the plaintiff, I cannot say that a primary Judge who heard and saw the witnesses must come to the same view.

Although such factors may tempt an appellate court to substitute its own findings of fact, this is a case where conflicts of evidence remain which cannot satisfactorily be resolved at appellate level. A determination by this Court of Appeal would deprive the defendant of an opportunity to have a direct conflict of evidence resolved in the usual way, that is to say by a primary Judge who hears and sees the witnesses. In my view this would be an unwarranted intrusion into the domain of the trial courts.

The appeal should be allowed with costs and there should be directions that there be a retrial before another Judge limited to the issue of liability. The judgment below including the order for costs should be set aside. It should be directed that the costs of the first trial be costs of each party in the cause.