Balenzuela v De Gail

He is a young man who by this time has attained full age but at the time when he sustained the injury of which he complains he was not eighteen years of age and when the writ in the action was issued on 15th June 1954 he was still an infant and therefore sued by his next friend. The defendants are ice vendors, father and son. On 15th September 1953, the day when the plaintiff sustained his injuries, they were conveying two tons of ice or thereabouts in a Dodge van or truck up a somewhat unsatisfactory and not very wide road called Rickard Road at Elanora near Narrabeen. The son who was in his father's employment was driving the vehicle, which was moving slowly. Coming down the hill a motor-cycle and side-car appeared ridden by the plaintiff. The latter's attempt to swing to his left failed to avoid a collision. The truck and the right-hand side of the motor-cycle collided and the motor-cycle and the boy were thrown down into the bush at the side of the road. The injuries he sustained proved serious.

His story was that the ice van was travelling over the crown of the road with its off side wheels on the wrong side and left him no room to pass. He was riding in his third gear, not his highest, and travelling at a moderate speed, fifteen miles per hour according to his evidence, about twenty-five miles per hour according to the evidence of the defendants. Their story was that they were over to their left of the bitumen as far as their vehicle could go. The bitumen was only twelve feet wide. On their left-hand side at the edge of the bitumen there was what amounted to a watercourse, a water "causeway" the elder defendant called it. On the left of the road he said, "there is a hill that comes down sheer like that" (referring to the cutting); "it is clay and rock formation; then there is the water causeway where the water runs the bitumen just goes sheer down, there is no shoulder above it at all." Asked about the depth of the watercourse "at the site of the accident" he replied, "Approximately it would be one foot to two feet, with variations, the holes." The plaintiff's case took by no means this view of the features of the locus in quo as they existed at the time of the accident. The place was identified by three photographs which were put in evidence for the plaintiff and the defendants appeared to acquiesce in the identification. A witness who had been at the time a police constable stationed at Collaroy was called for the plaintiff. He had gone to the scene of the accident immediately after it took place. In giving his evidence he looked at the photographs, stated the position of the vehicles and gave one or two details about the road at the curve depicted. He said that the width of the shoulder on the left or southern side, that is what at the accident would have been the defendants' left, was three feet and it was what he called "trafficable" and that on the opposite side there was no trafficable portion beyond the bitumen, it fell away downhill. At the trial, which lasted over one day, much attention was given to this contested question. On the second day the defendants called two witnesses familiar with Rickard Road who had been interviewed overnight, both friends of the younger defendant. The first, Barry Godden by name, said that he had gone up with him to the scene of the accident on the day following its occurrence. That defendant had shown him where it took place. He described the features he attributed to the bitumen strip, the "natural watercourse" alongside it, six inches to two feet six inches deep and two feet wide, and on the other side of the road (the plaintiff's side in the accident) a shoulder that was in his language "drivable for vehicles". The second such witness was named Adams. According to him before he came to the Court to give his evidence he had been taken that morning to the scene of the accident which was shown to him by Godden. He had, of course, known of the accident at the time it happened; and he had used the road so that he was familiar with its features. But clearly enough the purpose of taking him there on his way to the Court, which he said was his idea, was to enable him to speak of the particular place where the collision occurred. When he came to give his evidence he confirmed the description, as at the time of the occurrence, of the roadway and of the watercourse beside it to which Godden and the defendants had deposed. But he assigned the place which he had been shown as the site of the accident to a point in the road one hundred yards or possibly one hundred and fifty yards above that where the locus of the collision had been fixed up to that stage of the trial. On being asked whether he could fix the spot he said that there was a house on the left-hand side going up where they sell dogs and things and the spot was one hundred and fifty yards past that going up. He said that there was a bend to the left going up the hill and that Godden showed him that point. In cross-examination he said that it could only be during the last couple of years that the people had been selling dogs: it had a sign up "Boxer pups for sale"; but, he continued, he had made a mistake, the house was on the right-hand side not the left. However, he went on to say that the spot he was shown would be one hundred and fifty yards from the house he had in mind on the left which was where an old stonemason had a quarry. It would be one hundred yards up from where the pups were sold. One would have thought that this evidence made it important to fix the site of the house where the boxer pups are or were sold. If that point were fixed you could identify the place with more or less accuracy which Adams and he had been shown and which he purported to describe. You might conclude that Godden had described the same place. Was it in fact the place where the accident happened? Was it the same place as the defendants spoke of? As evidence in reply a witness named Barnes was called for the plaintiff. He said without objection that he had been present at Rickard Road when the photographs put in evidence had been taken. He was handed one of them and asked to identify a house shown at the bend where, or whereabouts, it will be recalled, the site of the accident had been placed by common consent before Godden gave evidence. The witness was then asked two questions both of which were objected to for the defendants and were rejected. They were directed to ascertaining whether that was not the house exhibiting the sign advertising the sale of boxer dogs. We were told that the identification of the position of the house was treated by the judge as not relevant but as going only to the credit of Godden and Adams and that it was for that reason that the evidence was rejected. Although it appears to have escaped notice at the trial, on the motion for a new trial one of the learned judges pointed out that on a very close scrutiny of one of the photographs it was possible to make out the figure of a dog standing in front of the house. This leaves little doubt what the rejected evidence would have proved. It is a fact which the jury might have regarded as important, namely that the place which Adams purported to describe to them was not the actual site of the accident; perhaps that might be true too of Godden's evidence and possibly even of the defendants themselves. It is difficult to see why this evidence was not relevant and admissible. It related to an objective fact serving to identify the place about the nature of which evidence had been given and the question whether it was the same place as that to which other evidence ascribed the accident was important. That it was admissible the judges of the Full Court of the Supreme Court all thought. The conclusion, however, that the evidence was erroneously rejected seems to me necessarily to mean that there must be a new trial. When material evidence has been erroneously rejected at the instance of the party who succeeds, then to deny nevertheless to the unsuccessful party the remedy of a new trial the Court must have some sure ground for saying that the reception of the evidence would not have affected the result or that it ought not to have done so. I do not rest my conclusion in this case upon the supposed difference between the common law rule which prevails in New South Wales and the provision made in the rules under the judicature system. The provision stands in England now as O. 58, r. 10 (2), but formerly it was O. 39, r. 6. The difference between the common law rule and the effect of that provision can easily be exaggerated by over-estimating the operation of the judicature provision in widening the discretion of the court and by under-estimating the effect of the common law rule in allowing a discretion to the court. In Holford v. Melbourne Tramway and Omnibus Co. Ltd. [1] , Cussen J. after referring to the provision said, "I think that the rule is very little different from the view which was taken by the courts before the Judicature Act. The courts would not necessarily direct a new trial if the misdirection was on an immaterial or collateral matter, or if the trial resulted in a verdict against the person in whose favour the misdirection operated, or if the mis-direction was in respect of a pure question of fact, and the judge's attention was not called to the mistake. But it is an error to think there never can be a wrong or miscarriage unless it can be shown that the jury were in fact influenced in giving their verdict by a mis-direction. There is a wrong or miscarriage occasioned by a mis-direction in law, or as to the application of evidence, if, as a final result of what has been said by the judge, the jury retire to their room under a wrong impression in relation to these matters, and the result of the case is such as to show that they may have been influenced in their verdict by the misdirection. "Miscarriage" is a technical word, and includes this technical meaning. The plaintiffs' counsel contended that the onus of showing the miscarriage is on the party asking for the new trial. I think this is clearly right, but I think that onus is satisfied when the facts appear to be as above set out, and that unless the party opposing the grant of the order for a new trial can point to some further fact, the conclusion that there was a miscarriage must be drawn" [2] . In Pratten v. The Labour Daily Ltd. [3] McArthur J. pointed out that, although in the foregoing passage Cussen J. had put the burden upon the party complaining of an error at the trial of satisfying the court of appeal that it amounted to a substantial wrong or miscarriage, the contrary had been said. "It has been said", remarked his Honour, "that if misdirection is established the onus of showing that it did not influence the result is on the party showing cause against a new trial" [4] . McArthur J. referred to two cases, Anthony v. Halstead [5] and White v. Barnes [6] . The second confirms a statement made in the first case by Grove J. during the argument, and by Lindley J., in his judgment, to the effect that it was for the party showing cause against a new trial to show that the misdirection did not influence the result. One might suppose that upon a question depending not upon proof of an issue by evidence but upon reasoning as to the effect of what from a record appears to have occurred at a trial, it would not matter much upon which side the duty lay of persuading the court that a given error involved a substantial wrong or miscarriage. In the first of the cases referred to there seems to have been a question as to the use of counsel's notes to show what had occurred at the trial. The judges spoke of producing "authentic evidence" to establish the miscarriage of justice. This expression, which doubtless related to the use of notes, has been repeated in a way calculated to cause misunderstanding. But, however this may be, both in the case of the common law rule and that under the judicature system inconsistent judicial statements may be found as to where the burden lies of satisfying the court of the tendency, or lack of tendency, of the misdirection or misreception of evidence or other error at the trial to affect the result. For example, in what is perhaps the most recent judicial discussion of the law and practice in New South Wales, I refer to that by Manning J. in his dissenting judgment in Honnery v. Smith [1] since considered in Lakeman v. Finlay [2] , the conclusion is expressed that "where the appellant complains that evidence has been wrongly rejected a new trial should be granted only if he satisfied the court that if it had not been so rejected the result of the trial might reasonably have been different" [3] . But the opposite burden of persuasion is stated in the formulation made in this Court in Piddington v. Bennett and Wood Pty. Ltd. [4] by Evatt J. of what was "at that time the accepted practice in New South Wales on the civil side" [5] . The words "at that time" refer to the time of the decisions in Makin v. Attorney-General for New South Wales [6] and Macleod v. Attorney-General for New South Wales [7] . His Honour's formulation of the then accepted civil rule was expressed in terms relating to the wrongful admission of evidence. The learned judge said "that the court would as a rule grant a new trial where evidence had been improperly admitted: but that in its discretion the court might refrain from granting a new trial if it was affirmatively satisfied that the actual verdict returned could not have been affected by the inadmissible evidence" [8] . If the question really be of any importance, resting as it does rather on processes of reasoning than upon proof of an issue of fact, the true view, it may be suggested, is that at common law it was necessary to grant a new trial unless the court felt some reasonable assurance that the error of law at the trial whether in a misdirection or wrongful admission or rejection of evidence or otherwise was of such a nature that it could not reasonably be supposed to have influenced the result or because, in any case, as a matter of law the same result must have ensued, while under the judicature rule the burden is the other way. The form in which the judicature rule is cast seems to indicate an intention that the court should not grant a new trial unless it reached a positive opinion, in other words unless it was persuaded that a substantial wrong or miscarriage had been occasioned by the error. This may form one distinction between the two rules. It is the view adopted in Best on Evidence, 12th ed. (1922), p. 70. Another distinction perhaps may be that a rather more substantial wrong or miscarriage has been required under the judicature rule than had been required at common law. But the existence of the distinction is doubtful. It must be remembered that the law relating to the grant of new trials was one that had continued to develop. That the remedy of a new trial was in many respects a substitute for a bill of exceptions was a consideration that long tended to prevent the refusal of a new trial where a bill of exceptions would have provided an effective remedy. But as time went on this consideration had ceased to govern the court's discretion. Probably what Higgins J. said in Robinson and Vincent Ltd. v. Rice [1] is justified in substance, namely that the same principle as that obtaining under the judicature rule applied both before the Common Law Procedure Act 1852 and since that in England and was applied in New South Wales. But whether it be the rule at common law or under the judicature system a much more important restraint must be observed upon the discretion of the court to refuse a new trial. The court, where the determination of questions of fact is not entrusted to it, cannot substitute its judgment for that of the jury in purporting to decide whether a substantial wrong or miscarriage has occurred. Care must be taken lest in exercising an authority to decide whether an error of law occurring at the trial is likely to have influenced the result, what is really done is to examine the evidence as if the court were forming a conclusion of fact for itself. The basal distinction between the court's duty and the function of the jury cannot be confused in this way. The question whether an error of law made at a trial calls for a new trial depends on definite considerations involving a legal criterion. Cases may be put on one side where, although there has been some error of law committed at the trial, whether in the judge's direction to the jury or in his rulings as to the reception or rejection of evidence, the verdict may be sustainable because the successful party in any case is entitled as a matter of law to such a verdict or because had a contrary verdict been returned it must have been set aside. Apart from such cases it is enough if it appears to the court deciding an application for a new trial that an error of law has been made at the trial which may operate to do any of the following three things - (a) improperly to limit or enlarge the evidentiary materials by which it is not an unreasonable hypothesis to suppose the judgment of the jury might be affected, even if illogically; or (b) in the case of misdirection touching standards or criteria of liability, wrongly to expose the party to a hazard that is appreciable and not illusory of a verdict for or against him that otherwise might not have been found; or (c) in matters of burden of proof, of the legal need of corroboration or of other like incidents of the jury's consideration of the case, to provide an erroneous guidance which it is not unreasonable to regard as capable of contributing to the result. These categories are probably not exhaustive but it is enough if a case falls within one of them; if it does it is necessary to treat it as involving a substantial wrong or miscarriage. What has been said relates entirely to misdirection, the erroneous reception or rejection of evidence and other errors of law. As to erroneous dealing with facts, for present purposes it is enough to refer to what Cussen J. says as to the distinction in Holford's Case [1] . But where the error is of law and is one of the foregoing description, it is not for the court to proceed to inquire into the facts of the case and form a conclusion as to what the jury would or should have done had the trial proceeded entirely in conformity with law and without any misdirection or misreception or erroneous rejection of evidence. That is what the decision of the House of Lords in Bray v. Ford [2] means. See per Lord Greene M.R. in Braddock v. Bevins [3] .

1. (1909) V.L.R. 497. 2. (1909) V.L.R., at p. 526. 3. (1926) V.L.R. 115. 4. (1926) V.L.R., at p. 131. 5. (1877) 37 L.T. 433. 6. (1914) W.N. 74. 7. (1957) S.R. (N.S.W.), at pp. 607-609; 74 W.N., at pp. 380, 381. 8. (1959) S.R. (N.S.W.) 5; 75 W.N. 479. 9. (1957) S.R. (N.S.W.), at p. 609; 74 W.N., at p. 381. 10. (1940) 63 C.L.R. 533. 11. (1940) 63 C.L.R., at p. 563. 12. (1894) A.C. 57. 13. (1890) 11 L.R. (N.S.W.) 218; (1891) A.C. 455. 14. (1940) 63 C.L.R., at p. 563. 15. (1926) 38 C.L.R. 1, at p. 10. 16. (1909) V.L.R., at p. 527. 17. (1896) A.C. 44. 18. (1948) 1 K.B. 580, at pp. 599, 600.

In the present case, as it seems to me, it would be contrary to principle to refuse a new trial on the ground that there was no substantial wrong or miscarriage. The basal fact is that material evidence was erroneously excluded from the consideration of the jury, evidence that touched the question upon which the case turned. It was something the party was entitled to lay before the jury for its consideration. It lies outside the province of the Court to inquire into the effect which the evidence if admitted would produce upon the Court if the Court were the tribunal of fact, and it lies outside the province of the Court to speculate on the effect which it would have produced on the jury. It is enough that evidence definitely material to the determination of the case was excluded at the instance of the defendants. That leaves the unsuccessful plaintiff entitled to a new trial.

In my opinion the appeal should be allowed, the order of the Full Court of the Supreme Court should be discharged and in lieu thereof it should be ordered that a new trial be had between the parties.

The evidence of Barnes which was rejected at the trial was admissible as tending to show that Adams, when describing the state of the road at the place where he understood that the collision had happened, was speaking of a place a hundred yards or so away from the corner where in fact it occurred. The only question is whether the plaintiff, having failed in the action, is entitled to a new trial because of the rejection of this evidence. If the answer depended upon whether we thought it likely that the evidence, if it had been admitted, would have altered the jury's view of the case, my opinion would be that the appeal should fail. A reading of the transcript record of Adams' evidence produces in my mind a strong feeling that Adams was under no misapprehension as to where it was that the collision occurred, and that the whole trouble arose from his having become confused, in the course of his cross-examination, as to the distance of the place of collision from the house to which Barnes referred as the house where the dogs were sold, and to his not being given an opportunity to make himself clear by being shown the photographs which were in evidence. I should think that the nature of Adams's confusion must have been quite plain to the jury. But in the present state of the law of New South Wales, as appears from the cases which have been and are to be cited, a new trial must be granted unless we can say that the jury, proceeding according to law and within the bounds of reason, could not have been led by the rejected evidence, if it had been before them, to find a verdict for the plaintiff. It is not possible to go as far as that, and I must therefore agree that the appeal should be allowed.

It is with some regret that I have reached the conclusion that this appeal should be allowed and a new trial ordered. The evidence which was rejected by the learned trial judge was, it seems to me, on the fringe of the case and, during the argument on the appeal, I was inclined to think that it might reasonably be urged that if it had been before the jury it could not have influenced their verdict. But upon consideration and after examination of the whole of the evidence in the case I am satisfied that this conclusion cannot be reached by the application of those principles which, in New South Wales, govern the discretion of the Supreme Court in dealing with applications for new trials in nisi prius matters on the ground of the wrongful rejection or reception of evidence: see Chitty's Archbold's Practice, 12th ed. (1866), p. 1519; Gordon v. Bank of N.S.W. [1] ; Goodsell v. National Bank of Australasia [2] ; Piddington v. Bennett & Wood Pty. Ltd. [3] and Crease v. Barrett [4] . Upon those principles an appellant is entitled to a new trial once it is seen that material evidence has been rejected unless it plainly appears that its rejection could not have affected the jury's verdict. In spite of the observations of Higgins J. in Robinson and Vincent Ltd. v. Rice [5] , this rule is quite different from that formulated by O. 39 r. 6 of the Judicature Rules (now O. 58, r. 10 (2)) under which the discretion of the appellate court is to be exercised against an an appellant unless, in its opinion, "some substantial wrong or miscarriage has been occasioned" by the wrongful rejection or reception of evidence.

1. (1886) L.R. (N.S.W.) 122, at pp. 129, 130. 2. (1890) 11 L.R. (N.S.W.) Eq. 156, at pp. 159, 160. 3. (1948) 63 C.L.R., at p. 554. 4. (1835) 1 C. M. & R. 919 [149 E.R. 1353]. 5. (1926) 38 C.L.R. 1, at pp. 9, 10.

On the whole, I am of the opinion that it is not possible to say that the trial could not have resulted differently if the rejected evidence had been admitted. It was material evidence, it was tendered to establish a fact not otherwise proved or admitted and it may have been thought to have introduced a factor relevant to consideration of the evidence concerning the state of the road in the vicinity of the collision. It is true, of course, that it may be thought to deal with a matter on the very fringe of the case and I would readily agree that in the circumstances of the case it is quite unlikely that its admission would have affected the jury's determination. But to rest our decision on this ground would, in my view, be to depart from established principles and, in the language of Parke B. in Crease v. Barrett "in a degree" to "assume the province of the jury" [6] . It is, perhaps, unfortunate that the discretion of the Full Court should be so circumscribed in matters of this character but as the matter stands there must, in my opinion, be a new trial.

1. (1835) 1 C. M. & R., at p. 933 [149 E.R., at p. 1359]. Menzies J.

At the trial of this action the trial judge, as the Full Court of the Supreme Court of New South Wales decided, wrongly rejected evidence to identify a house shown in a photograph that was part of the evidence. The Full Court (Owen J. dissenting) nevertheless refused the plaintiff appellant a new trial because the majority (Street C.J. and Herron J.) were not satisfied that the error "did affect or was capable of affecting the ultimate result of the trial".

The action was for damages for negligence causing a collision between the plaintiff's motor-cycle and side-car and the defendants' truck, and an important issue in the case was the state of the road at or near the point of collision.

Two witnesses, called for the defendants, who had no first-hand knowledge where the collision occurred did describe the condition of the road late in 1953, that is, at the time of the collision, in the vicinity of a spot which was pointed out by one of the defendants as the place where the collision occurred in a way helpful to the defendants if the section of the road they described was, in fact, where the collision occurred. One of these witnesses (Adams) waveringly fixed the place he described as 100 yards up from a house where boxer dogs were sold and the reason for seeking to identify the house in the exhibit as the house where boxer dogs were sold was for the purpose of persuading the jury that the section of the road which he and the other witness described was not the place where the collision occurred because it was not in dispute that the collision had occurred on the bend in the road where the house stood.

Where admissible evidence is wrongly rejected the party aggrieved is prima facie entitled to a new trial but a new trial will not be ordered if the evidence rejected could have had no effect with the jury: Crease v. Barrett [1] ; Reg. v. Gibson [2] , per Lord Coleridge C.J. [3] ; Goodsell v. National Bank of Australasia [4] ; Piddington v. Bennett and Wood Pty. Ltd. [5] , per Dixon J. [6] and per Evatt J. [7] . In jurisdictions where there is a special rule dealing with the matter different considerations apply: cf. English R.S.C. (O. 58, r. 10).

1. (1835) 1 C.M. & R. 919 [149 E.R. 1353]. 2. (1887) 18 Q.B.D. 537. 3. (1887) 18 Q.B.D., at pp. 540, 541. 4. (1890) 11 L.R. (N.S.W.) Eq. 156. 5. (1940) 63 C.L.R. 533. 6. (1940) 63 C.L.R., at p. 554. 7. (1940) 63 C.L.R., at pp. 560-564.

I am unable to say that the evidence which was rejected in this case could not reasonably have affected the minds of the jury and that if the evidence had been given the result must have been the same; beyond that, the significance of the evidence is a matter for the jury not an appellate court and I think it unnecessary to discuss its relative importance in the case.

It was argued for the respondent that in any event a new trial should be refused because this was a case where the trial judge should have directed a verdict for the defendants on the ground that the evidence for the plaintiff established contributory negligence on his part. The trial judge rejected a submission to this effect and I think he was right in doing so.

For these reasons I agree with Owen J. that there should be a new trial.

At the trial the plaintiff's and the defendants' witnesses differed as to the nature of the roadway on its southern edge - that is on the defendants' correct, or near, side. This was a material matter. The learned trial judge's summing-up and the evidence led for the plaintiff show that the plaintiff contended that the defendants had been negligent in not keeping their vehicle to the left of the roadway as they approached a curve. The plaintiff relied upon reg. 66 under the Motor Traffic Act 1909 as amended. This provides that the driver of a vehicle approaching a crest or curve shall drive so as to keep the centre line of the carriage-way on the right-hand, or off side, of his vehicle whilst his view is obstructed within a distance of five hundred feet ahead by reason of the crest or curve. This is an obvious rule of prudence quite apart from any regulation. The collision occurred some fifty yards before the defendants' vehicle reached the curve, and at a point not visible to the plaintiff until he was actually rounding the curve. There was no centre line marked on the roadway; and, on this aspect, the question between the parties became really whether the defendants' vehicle was being driven as close to the left of the usable road as it could reasonably be. The bitumen surfaced portion of the roadway was narrow; but the plaintiff's contention was that the defendants could have driven their vehicle further to the left by using the unsurfaced shoulder of the road beyond the bitumen. Whether it was practicable for them to have done so at the time of the accident was in dispute. The trial took place more than three years after the accident. The road had been repaired and altered in the meantime; so that witnesses spoke of their recollection of its condition three years before. It was common ground that on the eastern side of the bitumen the surface had been much eroded by water. The plaintiff admitted that this was so. He said, however, that, although a motor car could not have used it, this side of the road was "trafficable" for a truck. A witness, Bird, at the time of the accident a police constable, who had inspected the road shortly after the accident said that it was trafficable. The defendants, on the other hand, said that immediately beyond the bitumen there was a drain of such depth and irregularity that no vehicle could be driven with its wheels off the bitumen on that side. One of the witnesses, Adams, called for the defendants had what was said to be the place of the accident pointed out to him by another witness, Godden, immediately before he went to the court. He then gave evidence of the state of the road at that place as he remembered it to have been at the time of the accident. His description accorded generally with the evidence of the defendants and of Godden. He was cross-examined as to the surroundings of the place he had inspected. After some hesitations, inconsistencies and uncertainty he fixed it as a hundred, or a hundred and fifty, yards above a house where there was a notice advertising puppies for sale; but he said that his attention had been given more to the part of road pointed out to him than to the houses in the vicinity. Several photographs showing short stretches of road including the curve had been put in evidence in the plaintiff's case. Adams was not shown any of these photographs while he was in the box; but the plaintiff's counsel sought to call evidence in reply to prove that a house shown in one of the photographs was the house where puppies were for sale. How far this house is from the point on the road where the collision occurred was not proved, and counsel were unable to tell us. But, so far as one can tell from the photographs, the house is less than a hundred yards from any part of the road appearing in any photograph. Moreover the place of collision was, it seems, below, that is downhill from, not above, or uphill from, the house. Therefore it was said that by establishing that the house in the photograph was the house to which Adams referred the plaintiff's counsel would have been in a position to suggest to the jury that the part of the road which Adams and Godden had looked at and described, being a hundred and fifty yards above the house, was somewhere outside the area shown in the photographs. And thus it could be suggested they had looked at the wrong place. The evidence, if admitted, would have been far from conclusive of this. Adams, who was obviously somewhat confused as to the houses, might have wrongly estimated the distance from the place where he stood. Moreover he was not describing the condition of the road at a precise point so much as the general character of its edge as affecting the possibility of driving with the near side wheels off the bitumen. The plaintiff's case was not that the defendants' truck should have moved further to the left in the emergency, but rather that, when going up the road and towards a "blind" curve, it was too far to the right and in contravention of the regulation. In these circumstances it seems to me that evidence about the house shown in a photograph, which photograph had not been shown to Adams or Godden, would have been a very unsure, and indeed somewhat disingenuous, ground on which to ask the jury to infer that the witnesses had gone to and looked at the wrong place. I agree with what Herron J. said as to the effect of Adam's evidence. Thus it is not, I think, surprising that the learned trial judge failed to appreciate what is now urged as the significance of the questions asked; and this it seems was not explained to him; so that when the evidence was objected to by the defendants' counsel he rejected it. But it was admissible; and I agree with the view of Owen J., in his dissenting judgment in the Supreme Court, and with other members of this Court that the consequence of its rejection is that there must be a new trial. I am unable to say that if admitted it might not have influenced, and probably influenced, the jury. The Chief Justice has stated the principles of law which govern the matter. With all that he has said I agree; but, because of some of the arguments addressed to us, I wish to add something for myself.

Since Crease v. Barrett [1] it has been generally accepted that, notwithstanding earlier decisions to the contrary, if admissible evidence which might have influenced the jury be rejected, there must be a new trial; unless, of course, the case was one in which any verdict other than that found by the jury would be so unreasonable that it would be set aside. Except in circumstances such as those referred to in Chitty's Archbold's Practice, 12th ed. (1866), p. 1506, and apart from the modern rule under the judicature system about substantial miscarriage, to which I shall refer later, a court of appeal has no discretion to overlook an improper rejection of evidence. The consequences of an improper admission of evidence are, generally speaking, similar. But, in my view, the rule is there, if anything, less strict. This is because evidence may be inadmissible by reason of irrelevancy and because it relates only to some matter remote from and immaterial to any matter in issue. Its reception in such circumstances might be only an immaterial error; and a new trial could properly be refused if its reception could not reasonably be supposed to have affected the jury's verdict. But the position is, I consider, logically rather different if evidence which ought to have been admitted be rejected. Such evidence can never be irrelevant. It must ex hypothesi be material to some issue. The very reason why it should be received is that it might assist the jury in arriving at a verdict. Its rejection, therefore, necessarily deprives the jury of knowledge of some testimony which the party tendering it was entitled to have them know. Subject to, this distinction, which in some circumstances may make wrongful rejection an a fortiori case, the general principles stated in Piddington v. Bennett and Wood Pty. Ltd. [1] by Dixon J. (as he then was) [2] and discussed by Evatt J. [3] are as applicable to improper rejection as to improper reception, which was there under consideration.

1. (1835) 1 C. M. & R. 919 [149 E.R. 1353]. 2. (1940) 63 C.L.R. 533. 3. (1940) 63 C.L.R., at p. 554. 4. (1940) 63 C.L.R., at pp. 560-564.

We were pressed with the view that a new trial is always a most "deplorable result". But whenever there has been a significant error in law it is a necessity of justice, to be deplored but not refused. No doubt it is probable that the jury were aware, at all events after the rejected questions had been asked, that the house in the photograph was the house of which Adams spoke. The learned Chief Justice of New South Wales seems to have thought this to be so, for he said: "the shutting out of what I am satisfied must have been common knowledge, namely, that this house shown in ex. B (a photograph) was the house where the dogs were sold, could not have such an influence on the result of the trial as to require this court now to order a new trial for the purpose of redressing an injustice". If I could feel sure that the jury did know what the rejected evidence would have established I would agree. But I cannot be sure; I think we are not justified in speculating as to what influence the rejected evidence might have had with the jury had it been admitted. General statements that new trials should not be lightly granted have been frequently made and are frequently quoted. The most eloquent is perhaps that of Lord Penzance in Scott v. Scott [4] and see Dakhyl v. Labouchere [5] and Turnbull & Co. v. Duval [6] . But such statements ought not to be read apart from the context in which they were made, or as meaning that an appeal court has in every case a complete discretion to grant or refuse a new trial. A new trial cannot be refused just because the court thinks the jury's verdict right, any more than a new trial can now be granted merely because the court thinks the verdict wrong. There is a clear distinction between, on the one hand, applications for a new trial on the ground that the verdict was against the weight of evidence or because of discovery of fresh evidence or because the judge's summing-up was, in relation to the facts, insufficient, and, on the other hand, complaints of misdirection of law or wrongful rejection and reception of evidence. In the former cases a new trial is, within limits, a discretionary remedy to be applied only if the court thinks there has been a miscarriage of justice. But in the latter cases there has been an error in law; and the court must assume that it has, or may have, resulted in a miscarriage of justice, for a party has a right to have his case tried according to law. The distinction between the two classes of cases is referred to in, among other places, passages in the judgments of the present Chief Justice in Hocking v. Bell [1] ; Orr v. Holmes [2] and Wollongong Corporation v. Cowan [3] . Motions for new trials became common after the seventeenth century as a means of redressing perverse, corrupt or altogether unreasonable verdicts of juries at nisi prius. They were at one stage readily granted if the trial judge reported his dissatisfaction with the verdict. The courts in banc at different periods exercised different degrees of discretion in allowing or refusing new trials. Where the ground alleged was error in law, a motion to a court in banc for a new trial was, as the Chief Justice has pointed out, a substitute for a writ of error upon a bill of exceptions for a venire de novo; and this tended at one time to restrict the court's discretion. But when the complaint was not of error of law but of the jury's verdict, the courts in banc seem to have allowed themselves a wide discretion to grant or refuse a new trial: Blackstone Commentaries Bk III c. 24; Wilkinson v. Payne [4] .

1. (1863) 3 Sw. & Tr. 319, at p. 322 [164 E.R. 1298, at p. 1299]. 2. (1908) 2 K.B. 325, at p. 327. 3. (1902) A.C. 429, at p. 436. 4. (1945) 71 C.L.R. 430, at p. 499. 5. (1948) 76 C.L.R. 632, at pp. 640-642. 6. (1955) 93 C.L.R. 435, at p. 444. 7. (1791) 4 T.R. 468 [100 E.R. 1123].

I agree too with the Chief Justice that the common law principles obtaining in New South Wales and the rules under the judicature system are, in relation to new trials, not so far apart as might appear. To deprive a jury of knowledge of a material fact which could influence them in arriving at their verdict must, in my view, ordinarily be a "substantial wrong" within the meaning of the rule under the judicature system. That rule does not empower a court of appeal to usurp the function of the jury. It may make it necessary, in jurisdictions where it operates, for an appellant to satisfy the court that rejected evidence might, if it had been admitted, have influenced the jury. But the field is not one in which questions of where the onus lies can often be decisive. In Lakeman v. Finlay [5] the Supreme Court of New South Wales recently recognised that where there has been an error in law there is no such general discretion to refuse a new trial or any so pronounced presumption against granting one as was contended for here. I think the Supreme Court was right in that view and that error in law in the conduct of a trial, whether it be by misdirection or in relation to the reception of evidence, if proper objection were taken at the time, prima facie furnishes a ground for a new trial ex debito justitiae.

It was also argued for the respondent that the plaintiff's case must fail because of contributory negligence. But the trial judge was right in refusing to direct a verdict on that ground.