Australian Iron & Steel Ltd v Greenwood [1962] HCA 42; (1962) 107 CLR 308 (29 August 1962)

HIGH COURT OF AUSTRALIA

AUSTRALIAN IRON AND STEEL LTD. v. GREENWOOD [1962] HCA 42; (1962) 107 CLR 308

Negligence

High Court of Australia
McTiernan(1), Taylor(2), Menzies(2), Windeyer(3) and Owen(2) JJ.

CATCHWORDS

Negligence - Personal injuries - Damages - Whether excessive - Verdict upheld unanimously by Full Supreme Court - Trial by jury - Exercise by High Court of jurisdiction to order new trial.

HEARING

Sydney, 1962, April 12; August 29. 29:8:1962
APPEAL from the Supreme Court of New South Wales.

DECISION

August 29.
The following written judgments were delivered:-
McTIERNAN J. I would not interfere with the award of damages made by the in the summing-up. I think we ought to presume that the jury did their best to assess the damages according to the directions in the summing-up. To uphold this appeal we must be satisfied that with those directions to go upon, the jury's award is not one which four sensible men would make having regard to the circumstances of the case. The view which a jury might take of the injuries and sufferings of the plaintiff, and the damage done to his enjoyment of life and economic future is outlined by the observations made by the learned judges of the Supreme Court on the evidence. The case is not merely that of a man who lost a leg and returned to work without reduction of pay. This summary leaves out the account which the evidence proves of extreme pain and suffering, and the doubts which it creates about the future earning capacity and economic security of the plaintiff. At the time of the accident he was thirty years of age and on the evidence had prospects of promotion. The amount of 17,000 pounds is assumed to have been awarded as general damages. Mr. Allen by way of argument suggested that 10,000 pounds to 12,000 pounds would mark the limits of a reasonable award. The argument comes down to a denial that 17,000 pounds is reasonable damages and nothing is advanced to support the argument but the figure itself. The onus in this Court is on the appellant to make out that the learned judges of the Supreme Court ought to have said that the general damages awarded are too much to be within any reasonable limit. I am not prepared to decide that their Honours were wrong. In my opinion no good reason was advanced for so holding. (at p310)

TAYLOR, MENZIES AND OWEN JJ. This is an appeal from a decision of the Full Court of the Supreme Court of New South Wales refusing an application by the appellant (who was the defendant in the action) for a new trial on the question of damages on the ground that the jury's verdict was excessive. The claim was for damages for personal injuries and the verdict was for 19,528 pounds. (at p311)

2. The jurisdiction of this Court to hear and determine an appeal from the Full Court of a State granting or refusing a new trial on the ground that a jury's verdict was, or was not, outside the bounds of what is reasonable is part of the jurisdiction conferred by the Constitution and regulated by the Judiciary Act which specifically gives the High Court power to grant a new trial in any case where there has been a trial with a jury. Upon such an appeal this Court has power to give the judgment which the Full Court ought to have given and in considering what judgment ought to have been given when a Full Court has decided that a verdict was, or was not, excessive or inadequate it is necessary for this Court to consider the evidence given at the trial and the quantum of the verdict. From its establishment this Court upon such an appeal has consistently in this way decided for itself whether in its assessment of damages a jury failed to act reasonably: see among the reported cases Miles v. Commercial Banking Co. of Sydney [1904] HCA 54; (1904) 1 CLR 470, at pp 477, 478 ; Rowe v. Australian United Steam Navigation Co. Ltd. (1909) 9 CLR 1, at pp 13, 14-17, 33-34 ; Triggell v. Pheeney [1951] HCA 23; (1951) 82 CLR 497 ; Coates v. Carter [1951] HCA 30; (1951) 82 CLR 537, at p 542 ; but in deciding for itself, in a matter that is very largely one of impression, it of course pays particular respect to the decision under appeal: Thatcher v. Charles [1961] HCA 5; (1961) 104 CLR 57 . Furthermore, when, as here, a new trial is sought upon the ground that a jury's verdict is not reasonably based upon the evidence, the Court must assume that the jury has taken the view of the evidence most favourable to the respondent and only in what seems to it to be a clear case will a new trial be ordered. There is, we think, no difference between our function vis-a-vis the Full Court of a State and that of the House of Lords vis-a-vis a Court of Appeal in considering whether a jury's verdict should be allowed to stand - and it is clear that the House of Lords considers for itself whether a verdict was so high or so low that reasonable men confining their attention to relevant matters could not have arrived at such an estimate: see, to take a recent instance, the statement of Lord Tucker with which the other members of the House agreed in Cavanagh v. Ulster Weaving Co. Ltd. (1960) AC 145, at pp 163, 164 . Moreover, in discharging its duty to substitute its judgment for what it regards as a mistaken judgment of a Full Court of a State with regard to a jury's verdict, this Court has seen no reason to draw any distinction between appeals from the Full Court of the State of New South Wales and those from the Full Courts of the other States for the law which determines the powers and duties of this Court in every such case is Commonwealth law and it is the law of every State where civil actions may be tried with a jury that a jury's award of damages which is outside the bounds of reason should be set aside on the simple ground that the jury has failed in its duty to act reasonably. These are the principles to be applied in deciding this appeal. (at p312)

3. The accident in which the plaintiff received his injuries occurred in April 1956, he then being thirty years of age. He had been employed by the defendant for some years as a shift man and electric locomotive driver in the defendant's colliery and at the time of the accident was earning 19 pounds per week. The trial took place in February 1961 and it was agreed by the parties that his medical and hospital expenses and wages lost up to that time amounted to 2,528 pounds so that the sum of 17,000 pounds represents the jury's award for general damages including, of course, future loss of earning capacity. The injury which the plaintiff sustained was a serious one involving the crushing of his right heel with consequent fractures to the heel bone which had to be removed by surgical treatment. Attempts to save the foot involving a number of skin-grafting operations were made and the plaintiff was in hospital for about three months and during that period suffered considerable pain and discomfort. Unhappily the treatment to the foot proved unsuccessful and ultimately, in August 1957, it was found necessary to amputate his right leg below the knee. In March 1958 he was fitted with an artificial limb which he has worn since. He has since suffered a good deal of pain and discomfort in the stump of his leg and this may well continue. In July 1958 he returned to work as a locomotive driver in the defendant's mine and at the date of the trial he was still working for the defendant as a locomotive driver and was earning the same weekly wages as he would have received had there been no break in his service. There is nothing in the evidence to suggest that the defendant will not continue to be willing to employ him until he reaches the retiring age of sixty, fixed by statute for those who work in mines. The plaintiff said, however, that he found some difficulty in doing the work because of his artificial limb and the jury could have concluded that at some future and necessarily uncertain date before he reached the age of retirement his disabilities might compel him to give up work as a locomotive driver in the mine. The loss of the lower part of his leg has necessarily diminished to a very considerable extent his enjoyment of life and a further fact to be taken into account is that his post-accident employment requires him to work between 1 p.m. and 9 p.m. each working day. (at p313)

4. The account we have given of the plaintiff's injuries and of the disabilities, pain and suffering they occasion and his reduced capacity to work and enjoy life would, of course, justify a heavy verdict but we are satisfied that an award of 17,000 pounds for general damages - a sum which, if invested, would produce an income equal to the plaintiff's pre-accident earnings leaving the capital intact - is substantially beyond the highest figure which in our view could reasonably have been awarded. (at p313)

5. Accordingly we think that the order of the Full Court dismissing the appeal should be set aside and in its place an order made that there should be a new trial of the action on the issue of damages. (at p313)

WINDEYER J. This case comes to us as an appeal from a unanimous decision of the Full Court of the Supreme Court of New South Wales upholding the verdict of a jury. All that is said in support of the appeal is that the amount the jury awarded the plaintiff as general damages for personal injury is excessive. It is not said that the learned judges who sat in the Supreme Court overlooked some matter that they should have considered or considered anything that they ought not to have considered, or were in any way mistaken as to the principles of law applicable. Their conclusion was that the amount of the jury's verdict was not in fact beyond the bounds of reason. I think that, in these circumstances, this Court should not take it upon itself to say that what the Supreme Court thought was within reason was in fact beyond all reason, no question of law being involved. As I differ from other members of the Court as to the attitude that this Court should adopt, I think I should explain why I consider that, in cases such as this is, this Court ought not to interfere with the decision of the Supreme Court. (at p313)

2. The case is not one in which the relevance of some element or head of damage was in dispute. It is not one in which on the facts being found the damages were precisely calculable or in which pecuniary loss alone had to be estimated. It is not one in which some consequence of a breach of contract or of a tort was said to be too remote to sound in damages. It is not one in which some misdirection or other irregularity in the conduct of the trial might have affected the minds of the jury. What is said is that the Full Court should have said the damages were excessive, and that we should now reverse its decision and order a new trial. There are, in my view, two objections to this. They coalesce, but I shall consider them separately. The first lies in the nature of the jurisdiction the Full Court exercised. The second arises from the nature of the question that their Honours had to consider. (at p314)

3. The first depends upon considerations arising from the discretionary nature of the common law power to grant a new trial of an action tried by a jury. These considerations apply, more or less strongly, wherever common law actions are tried by juries according to common law principles, but perhaps more so today in New South Wales than in other places, because of the statutory requirement there that damages are to be assessed by a jury, and because of differences between a system based upon the Common Law Procedure Acts and systems based upon the Judicature Acts and rules: see Phillips v. Ellinson Brothers Pty. Ltd. [1941] HCA 35; (1941) 65 CLR 221 , especially per Rich J. (1941) 65 CLR, at pp 228-231 . The system of trials in civil actions in the Supreme Court of New South Wales is still, in essentials, that which prevailed in the common law courts at Westminster before the Judicature Acts. It is not for us to try to mould it in accordance with doctrines that are of later growth in a different system, however desirable they may perhaps seem. (at p314)

4. In New South Wales it has been provided by statute since 1844 that, with some irrelevant exceptions, all actions at law and all civil issues of fact in the Supreme Court shall be tried and damages shall be assessed by a jury of four persons. And moreover, subject to certain statutory alterations, every such trial or assessment shall "as far as practicable be subject to the same rules and manner of proceedings as would in England have been observed in an action at law in the courts of Westminster on a trial at nisi prius, before the seventeenth day of September 1847". That is still the law: Jury Act, 1912-1957 (N.S.W.), ss. 29 and 31. The New South Wales civil jury differs from the English jury in being ordinarily composed of only four persons. But in other respects its processes are the same. Hargrave J. once said, "I cannot understand a single section of our colonial jury laws without assuming the meaning of every word in the sense known, and known only to the English law, which meaning must, therefore, be part of our colonial law": Reg. v. Valentine (1871) 10 SCR (NSW), at p 131 . The history of the New South Wales civil jury is traced in the judgment of Else-Mitchell J. in Caledonian Collieries Ltd. v. Fenwick (1959) 76 WN (NSW) 482, at 488-491 and in an article in 33 Australian Law Journal 235. I need not refer to it further. Since 1900, issues of fact may be tried and damages may be assessed in the Supreme Court by a judge without a jury: Supreme Court Procedure Act, 1900. But this can only be by consent of both parties. When that course is adopted, the Act gives a right of appeal to the Full Court; and it is by way of rehearing. But there is no appeal from the verdict of a jury, except, of course, by motion for a new trial. The Supreme Court can order a new trial, but its power to do so is restricted within limits that the common law sets. (at p315)

5. Broadly stated, there are two classes of case in which at common law a new trial may be had of an action tried by a jury at nisi prius. (at p315)

6. The first is for an error of law made by the judge. Examples are cases in which there was a wrongful reception or rejection of material evidence; or in which a non-suit or verdict by direction was improperly granted or refused; or in which there was a misdirection on some point of law, as distinct from an insufficient direction on matters of fact. Errors of law of that kind could in the past have been made the subject of a bill of exceptions and a writ of error or, in some cases, be a ground for a venire de novo. In cases of that sort a new trial is thus a substitute for older remedies. (at p315)

7. Cases of the second class are those in which there was no error of law. They include those in which the jury, disregarding the evidence, returned a verdict that was clearly wrong. The practice of granting new trials in such cases began, as is well known, in the seventeenth century because the Court would not accept a wrong verdict as a warrant for pronouncing an unjust judgment. As error did not lie to correct a verdict, and as attaint was virtually obsolete after Bushell's Case [1823] EngR 38; (1670) Vaughan 135 (124 ER 1006) , the only means the courts had of controlling juries' verdicts was by ordering new trials. By the middle of the seventeenth century the practice had become well established: Bright v. Eynon [1757] EngR 109; (1757) 1 Burr 390 (97 ER 365) . At one period new trials were somewhat readily granted if the judge who had presided at nisi prius expressed himself as dissatisfied with the jury's verdict. But it was always for the court in banc to decide, in its discretion, whether or not there should be a new trial; for the question was whether the party for whom the verdict had been given should have the judgment of the court pursuant to the postea. If judgment had not been signed, the court could set the verdict aside and grant a new trial: or if, after 1831, judgment had been signed out of term pursuant to 1 Wm. IV, c. 7, it could vacate the judgment and grant a new trial: Tidd's Practice (1833) pp. 177, 178. (at p315)

8. The distinction between the two kinds of cases in which new trials might be granted was reflected in the attitude of the courts on the motion. Anything on which a bill of exceptions might have been founded, that is some error of law, came to be regarded as a ground for a new trial as of right, at all events if objection had been made at the trial - at common law such an objection was not in all cases essential: Holford v. Melbourne Tramway and Omnibus Co. Ltd. (1909) VLR 497 . But the granting or refusing of a new trial on other grounds was in the discretion of the court in banc. This distinction was adverted to by the Chief Justice, then Dixon J., in Hocking v. Bell (1945), 71 C.L.R., at p. 499. . I referred to it in Jones v. Dunkel (1959) 101 CLR, at pp 313, 314 ; Balenzuela v. De Gail (1959) 101 CLR, at p 244 ; Transport & General Insurance Co. Ltd. v. Edmondson (1961) 106 CLR, at pp 31, 32 . (at p316)

9. This is a case of the second class for a complaint that a jury's award of damages was excessive could not have been the subject of proceedings in error. The question is one of fact not of law: Henderson v. Clifford Watmough & Co. (1939) 161 LT 233 ; Cole v. De Trafford (1917) 1 KB 911, at pp 913, 914 . None the less, once the common law courts had assumed jurisdiction to set aside the verdicts of juries and grant new trials, it was never doubted that their power to do so extended to cases where the damages awarded were excessive. Indeed, even before then, a verdict could in some cases have been challenged by attaint on the ground that the damages awarded were "outrageous and excessive": see Duncombe, Tryals Per Pais or The Law of England concerning Juries by Nisi Prius 4th ed. (1702) p. 260. It may even be that the practice of granting new trials began with a case of an excessive verdict; for the first reported case of a new trial is often said to be Wood v. Gunston [1658] EngR 423; (1655) Style 446 (82 ER 867) , in the Upper Bench during the Commonwealth. Perhaps it was not really the first: see Bright v. Eynon [1757] EngR 109; (1757) 1 Burr 390 (97 ER 365) and Thayer, Preliminary Treatise on Evidence. Whether or not it was is immaterial. What is material is that the report states that "upon the supposition that the damages were excessive, and that the jury did favour the plaintiff, the defendant moved for a new trial". This was granted, Glyn C.J. saying, "It is in the discretion of the court in some cases to grant a new trial, but this must be a judicial, and not an arbitrary discretion". That states the law as it still is. (at p316)

10. Until 1854 a decision of any of the common law courts in England granting or refusing a new trial was final and conclusive. For there is no right of appeal from the decision of a superior court unless one be given by statute: Attorney-General v. Sillem (1864), 2 H. & C. 581, at pp. 608-609 (159 E.R. 242, at p. 253); [1864] EngR 352; 10 H.L.C. 704, at pp. 719, 724 [1864] EngR 352; (11 E.R. 1200, at pp. 1207, 1209). ; Victorian Stevedoring and General Contracting Co. Pty. Ltd. and Meakes v. Dignan [1931] HCA 34; (1931) 46 CLR 73, at p 108 . In 1854 the Common Law Procedure Act of that year provided, by s. 35, for appeals from decisions of the common law courts in England granting or refusing a new trial. But only when the court in banc was not unanimous, or itself thought fit to allow an appeal, could its decision be carried to the Exchequer Chamber. Moreover, the Act expressly provided that "where the application for a new trial is upon matter of discretion only, as on the ground that the verdict was against the weight of evidence or otherwise, no such appeal shall be allowed". This was a recognition of the distinction that had become firmly rooted in the common law and of the discretionary character of orders for new trials when verdicts are said to be against the weight of evidence. Complaints of excessive, or inadequate, damages are only a particular form of the general objection that a verdict was against the evidence: Cole v. De Trafford (1917) 1 KB, at pp 913, 914 ; Praed v. Graham (1889) 24 QBD, at p 55 . (at p317)

11. It is by virtue of the common law principles described above that the Supreme Court of New South Wales exercises its power to set aside the verdict of a jury and grant a new trial. By 9 Geo. IV, c. 83, s. 3, it was given jurisdiction "as fully and amply to all intents and purposes in New South Wales as His Majesty's Courts of King's Bench, Common Pleas and Exchequer or either of them lawfully have or hath in England". But, unlike those Courts, its decisions granting or refusing a new trial on the grounds that a verdict was against the evidence were never unappealable. At all relevant times there was an appeal from decisions of the Full Court to the Privy Council - either by leave pursuant to the prerogative or, later, in cases where a sufficient amount was involved, by virtue of 7 & 8 Vict., c. 69 and the Order-in-Council made thereunder. The history of this matter is referred to in the judgment of Isaacs and Rich JJ. in The Commonwealth v. Limerick Steamship Co. Ltd. and Kidman [1924] HCA 50; (1924) 35 CLR 69, at pp 100-104 . A decision of the Supreme Court on a motion for a new trial could therefore from the first be the subject of an appeal. But the tradition of the common law was powerful. In only a few cases has the Privy Council heard appeals from decisions of the Supreme Court granting or refusing a new trial where no question of law was involved. I am not aware of any in which the application for a new trial had been made on the ground only that a verdict was for an excessive amount. (at p317)

12. The position of this Court in a matter of this sort is different from that of the Court of Appeal hearing an appeal from a judgment entered after a trial by a jury in the King's Bench Division. Since the changes made by the Judicature Acts there has not, at all events since Divisional Courts ceased to grant new trials, been any court in England corresponding for present purposes with the Supreme Court in banc. And - except for the abortive Court of Appeal under 4 Geo. IV, c. 96 and the Charter of Justice - there has never been in New South Wales any intermediate court of error or court of appeal comparable with the Exchequer Chamber or the Court of Appeal in England. That is to say there is not, and never has been, an appellate court standing between the Supreme Court and the Privy Council. The jurisdiction that we are asked to exercise in this case is thus comparable with the jurisdiction that before Federation the Privy Council might have exercised in a similar case. By the Constitution the judgments of this Court in its appellate jurisdiction are final and conclusive, subject only to the prerogative preserved by s. 74. In my opinion, we should therefore govern ourselves by much the same considerations as would the Privy Council. This Court was, no doubt, intended by the founders of the Constitution to be more accessible to litigants than in those days the Privy Council could be. But the terms of s. 73 of the Constitution and its history contain nothing to suggest that there was any intention to restrict the discretion of the Supreme Court in controlling the conduct of proceedings at nisi prius and supervising the verdicts of juries. The Judiciary Act s. 73 states that, "The High Court in the exercise of its appellate jurisdiction may affirm reverse or modify the judgment appealed from and may give such judgment as ought to have been given in the first instance . . . ". The important words, for present purposes, are "such judgment as ought to have been given in the first instance". It is hardly necessary to say that "in the first instance" does not refer to the verdict of a jury or to a judgment entered pursuant to such a verdict, for from that there is no appeal: Musgrove v. McDonald [1905] HCA 50; (1905) 3 CLR 132 ; The Commonwealth v. Brisbane Milling Co. Ltd. [1916] HCA 39; (1916) 21 CLR 559 ; Fieman v. Balas [1930] HCA 39; (1930) 47 CLR 107 . (at p318)

13. It has been suggested that the function of this Court in relation to the Supreme Court of a State is analagous to that of the House of Lords in relation to the Court of Appeal in considering juries' verdicts. I do not think the analogy at all close. And in any event I have not read of any appeal to the House of Lords from a decision of the Court of Appeal in any way like the present case. On the first occasion when Phillips v. London & South Western Railway Co. (1879) 4 QBD 406; 5 QBD 78; 5 CPD 280 was before the Court of Appeal it was said that the case was to go to the House of Lords; but apparently it did not. I do not know of any case concerning damages for personal injuries that did, or of any case at all in which their Lordships entertained an appeal on a question of quantum of damages alone in a case tried by a jury. In Tolley v. J. S. Fry & Sons Ltd. [1931] UKHL 1; (1931) AC 333 there was no appeal from the finding of the Court of Appeal that the damages were excessive; so that the House of Lords did not have to consider the matter directly. Moreover, damages for libel have a different basis from damages for personal injury. Mechanical and General Inventions Co. and Lehwess v. Austin and The Austin Motor Co. (1935) AC 346 , was a case in contract. So far as it turned on quantum of damages for breach of an agreement, it was held that the Court of Appeal had followed a wrong principle in reversing the jury's verdict and entering judgment for nominal damages. Taff Vale Railway v. Jenkins (1913) AC 1 , was a case under Lord Campbell's Act, which turned on the relevance of a head of damage claimed as well as on the amount awarded by the jury. The three members of the Court of Appeal in that case had taken different views. The House of Lords upheld the verdict of the jury. I find nothing in these or other cases in the House of Lords that assists in the present question, except of course the well-known passages as to the principles which should guide an appellate court in reviewing the verdict of a jury that appear in the judgment of Lord Wright in the Mechanical & General Inventions Co's Case (1935) AC 346 , principles of which the learned judges in the Supreme Court were ully aware. (at p319)

14. The jurisdiction of this Court is not in question. The question is only as to the manner of its exercise. The question, as I see it, is whether we ought to say that the Supreme Court has erred, when their Honours were unanimous upon a matter that is largely one of discretion, and which is not referable to any precise criterion of law, but which is a matter rather of fact and opinion. A court of appeal may sometimes interfere with the exercise of a discretion by a judge, although it is not positively shown that he has acted on some erroneous principle: House v. The King (1936) 55 CLR, at pp 504, 505 ; Pearlow v. Pearlow (1953) 90 CLR, at pp 76, 77 . But this is not a case of a statutory power to entertain appeals from the exercise of a statutory discretion. Here what is asked is that the Court in the exercise of its general appellate jurisdiction should override the exercise of a common law court in banc of its traditional discretionary power on a motion for a new trial sought not for error of law by the court, but for error of fact by the jury. (at p320)

15. It is worth noticing that considerations similar to those that lead me to the conclusion that, on principle, we ought not to reverse the decision of the Full Court have for a very long time influenced courts in the United States in relation to this very question - the grant or refusal of a new trial on the ground of excessive damages. To begin with, the rule was there quite strict. A new trial could only be had from the court where the verdict was given or to which the record was returnable; and a Federal Court could not review the decision of a State Court, or of a Federal Court of first instance, granting or refusing a new trial on the ground that a jury's verdict was for an excessive or inadequate amount: see e.g. Public Utilities Corporation v. McNaughton (1930) 39 F 2d 7 ; Baltimore Line v. Redman [1935] USSC 126; (1935) 295 US 654, at p 657 . The decisions are numerous. They go back to the doctrine enunciated by Story J. in Parsons v. Bedford [1830] USSC 57; (1830) 3 Peters 433, at pp 447, 448; 28 US 432, at p 446 . They were based sometimes on the letter of the seventh amendment, the words being "according to the rules of the common law", sometimes on the more general ground that the grant or refusal of a new trial is a matter within the trial court's discretion. The rule is perhaps now less strict than originally it was: see Fairmount Glass Works v. Cub Fork Coal Co. [1933] USSC 5; (1933) 287 US 474, at pp 481-485 [1933] USSC 5; (77 Law Ed 439, at p 446) . But an order of a trial court granting or refusing a new trial on the grounds that damages for personal injury were excessive will still, speaking generally, not be set aside unless it appears that a wrong principle was applied by that court: see Houston Coca-Cola Co. v. Kelley (1942) 131 F 2d 627, at p 628 ; South Pacific Co. v. Guthrie [1949] USCA9 31; (1949) 180 F 2d 295 ; St. Louis Southwestern Railway Co. v. Ferguson [1950] USCA8 163; (1950) 182 F 2d 949, at p 954, 955 . In the last of these cases the prevailing view in the United States is summarized as follows: "We have said many times that the excessiveness or inadequacy of a verdict is not a matter for our consideration, but that the entreaty for any such vice lies solely to the judgment and the conscience of the trial judge on motion for a new trial. This is because the amount of a verdict is primarily a factual evaluation on inabsolute elements, while our function traditionally has been regarded as extending only to a testing of the soundness of the processes by which such a result has been achieved" (1950) 182 F 2d, at p 954 . In the latest American case on the topic that I have seen, Agnew v. Cox [1958] USCA8 80; (1958) 254 F 2d 263, at p 268 , the Eighth Circuit Court once again dealt with the question in a case of personal injury in a motor car accident, and quoted a passage from an earlier judgment: "This Court has consistently adhered to the proposition that the responsibility for keeping jury awards within reasonable bounds is essentially that of the trial courts and not of this Court [1953] USCA8 76; (1953) 203 F 2d 864, at p 868 ". So far as the American decisions are based on the seventh amendment, they are not directly in point. But, so far as they reflect general common law doctrine, they seem pertinent. (at p321)

16. I do not think that this Court has ever, or not until recently, reversed a unanimous decision of the Supreme Court refusing a new trial as to damages and upholding a jury's verdict, unless it appeared that that Court had proceeded on some wrong principle or under some misapprehension of fact. Indeed it has seldom considered objections to amounts awarded as damages except when that question was entangled in some way with other grounds of appeal. The Court allowed the appeal in Rowe v. Edwards [1934] HCA 32; (1934) 51 CLR 351 , but in doing so indicated that the Supreme Court had apparently misconceived its own powers. In the present case there is no indication that the Full Court mistook its task or misunderstood the evidence, and no error of principle in its approach is manifest. (at p321)

17. I pass now to the second of the two considerations that, as I said at the beginning of this judgment, determine my attitude. (at p321)

18. The case is different from one in which we are called upon to review an award of damages for personal injuries made by a judge. Then our task is clear, if not esy. We examine the evidence for ourselves, guided by whatever findings of fact the primary judge made and knowing his reasons for the conclusion that he reached. We can draw whatever inferences from the evidence as to past events we think proper, and make whatever assumptions we think should be made as to the probabilities of the future, remembering only that we have not seen the witnesses. Having done so, we must assess the damages according to our own opinions of what is reasonable compensation. We may be, indeed necessarily must be, influenced by knowledge of what was done in other cases. We must give due weight to the opinion of the trial judge and not lightly set it aside, but in the end we must decide whether the amount he awarded should stand or some other amount be substituted for it. But in this case we cannot do any of that. This case was tried by a jury. (at p321)

19. "The whole theory", said Lord Halsbury, "of the jurisdiction of the Courts to interfere with the verdict of the constitutional tribunal is that the Court is satisfied that the jury have not really acted reasonably upon the evidence, but have been misled by prejudice or passion": Watt v. Watt (1905) AC 115, at p 118 . The jurisdiction to interfere is not now restricted, as perhaps it originally was, to cases of verdicts that are, in the strict sense, perverse. When the damages awarded by a jury are so large or so small as to be out of all proportion to the harm suffered the verdict can be set aside. But the question is always whether the verdict shows that the jury "failed to perform their duty", as Lord Wright, quoting Lord Halsbury, expressed it: Mechanical and General Inventions Co. and Lehwess v. Austin and The Austin Motor Co. (1935) AC, at p 375 - or, as it has been put in the United States, whether the verdict shocks the sense of justice and raises a strong presumption that it is based on prejudice or passion rather than on sober judgment: Bangert v. Hubbard (1955) 67 AmLR 2d 395 In some cases, and on some issues, it may not be difficult to say of a verdict that it is contrary to reason, because the evidence concerning some element or incident essential to the conclusion may be overwhelmingly one way. And when special damages and pecuniary losses have to be calculated, it is often possible to say with assurance whether or not a particular result is reasonable, in the sense of being a logically possible inference from the evidence. That is because in such cases no resort to external standards beyond rational norms is required. But the assessment of general damages for personal injuries involves a resort to standards of evaluation that are imprecise. And in the result, judges as well as jurymen may, and as we know often do, hold widely differing views of what would be a proper sum even when they do not differ materially as to the nature, extent and consequences of the injury suffered. (at p322)

20. The appellant conceded that this Court should give due weight to the opinion of the Full Court. But, except as conventional politeness, I did not follow what that meant here; for, as I understood it, what was urged for the appellant was that, if our opinion differed from that of the Full Court, we should simply substitute our opinion. It is said that we must examine the evidence that was before the jury, not merely for the purpose of seeing whether the learned judges in the Full Court were under any misapprehension as to the facts, but for the purpose of ourselves testing the verdict. But in what way should we proceed to do this? It is not suggested that the plaintiff's injuries were not permanent and serious. They were. And the damages to which he is entitled must include compensation for elements of harm which were, as in all such cases they are, incapable of precise proof or accurate assessment in money. The general damages which might properly be awarded were, in Baron Martin's words, "such as the jury may give when the judge cannot point out any measure by which they are to be assessed, except the opinion and judgment of a reasonable man": Prehn v. Royal Bank of Liverpool (1870) LR 5 Exch 92, at pp 99, 100 . (at p323)

21. What is reasonable, whether a reasonable price, reasonable time, reasonable valuation or reasonable damages is a question for a jury and not for a court. And the question whether a jury's assessment of the damages shows that they failed in their duty is a question of fact. The common law in its early days did not deviate. A jury's verdict would only be set aside when, in proceedings for attaint, another jury, of twenty-four, unanimously found on the same evidence that the original jury of twelve had found a false verdict. That was long ago; but Lord Morris in Scott v. Musial (1959) 2 QB 429 used words in which there is an echo of the old law. "If the figure of an award seems to be outrageous, or so extravagant that no other jury would repeat it, then there might, in some cases, be ground for suspecting that a jury has been partial or perverse" (1959) 2 QB, at p 438 . When, today, a court is asked to set aside a jury's award of damages as excessive or inadequate, the question it has to consider is a question of fact, not of law. I have already referred above to decisions that this is so. To them may be added the observations of Cussen A.C.J. in Driver v. War Service Homes Commissioner (No. 1) (1924) VLR, at pp 532, 533 . Damages for pain and suffering, for the loss of a limb and the changed circumstances of life it entails "cannot be calculated and can only be measured according to the standards that generally prevail, and a reasonable conception of what is adequate to the occasion": per Dixon J. in Pamment v. Pawelski (1949) 79 CLR, at p 411 . Where are the measure of reasonableness and "the standards that generally prevail" to be found? Mainly, it is said, by considering what damages have been awarded in other cases. Birkett L.J. said in Bird v. Cocking & Sons Ltd. (1951) 2 TLR 1260 , that "when a particular matter comes for review one of the questions is, how does this accord with the general run of assessments made over the years in comparable cases?" (1951) 2 TLR, at p 1263 : and see Rushton v. National Coal Board (1953) 1 QB 495 ; Waldon v. The War Office (1956) 1 All ER 108 . But in those cases the Court of Appeal was dealing with assessments made by judges whose awards it had power to alter. The "general run of assessments" were decisions of judges either at first instance or in the Court of Appeal. Verdicts of juries stand in a somewhat different position. They may naturally be expected to differ more widely than opinions of judges do, because jurymen have generally no comparable accumulation of experience. And "when on a proper direction the quantum is ascertained by a jury, the disparity between the figure at which they have arrived and any figure at which they could properly have arrive must, to justify correction by a court of appeal, be even wider than when the figure has been assessed by a judge sitting alone": Nance v. British Columbia Electric Railway Co. (1951) AC 601, at p 614 . This may not seem logically satisfactory. And, as the day to day work of courts is nowadays so much concerned with the accidents that are the consequences of a mechanized age, it is perhaps inevitable, where damages are ordinarily assessed by juries, as they are in the Supreme Court of New South Wales, that judges should try to produce some consistency or apparent consistency. Justice, it was said, requires this. Without some predictability of the monetary consequences of possible occurrences, the whole of the existing system of accident insurance could, it is said, break down, and the settlement of claims be frustrated. Courts have made pronouncements recognizing the influence of these considerations. And some judges have consciously tried to maintain a general level of damages in cases of the same type by setting aside egregious verdicts. Challenged verdicts are professedly tested against the general run of awards made by judges and juries in other cases. Thus guided, the Supreme Court considers whether a verdict is in fact beyond the bounds of reason. If it is out of bounds, the only remedy is the clumsy procedure of a new trial before another jury; for the Court is governed by the principles of the common law and has no power to substitute its own assessment for that of a jury. (at p324)

22. I confess that I have misgivings about much of this. I have stated my views on some aspects of the matter in Thatcher v. Charles [1961] HCA 5; (1961) 104 CLR 57 . Even if we are to discard altogether as obsolete or old fashioned such statements as that of De Grey C.J. in Spong v. Hog [1746] EngR 491; (1772) 2 Wm Bl 802 (96 ER 472) , that "we cannot impeach the opinion of a jury in one cause by the opinion of a jury in another, especially where the evidence is different, and the point in issue not exactly the same", still for me a difficulty remains. No two cases of personal injury are in fact alike. The most important factors in producing a particular award may have been peculiar to the individual concerned. No real scale can be applied, because no basis of classification is satisfactory. What in some classifications are taken as constants are in particular cases submerged by variables. A subjective adjudication, and here it can be nothing else, depends upon reasoning from examples or by analogies, but not perhaps quite in Aristotle's sense. Adjudication involves the detection of decisive differences, as well as the perception of significant similarities. What weight is to be given to similarities and to differences can only be a matter of opinion. Furthermore, the experience and knowledge of other cases varies with the persons called upon to adjudicate. Since courts have said that conformity with empirically induced standards is desirable, text-book writers and commentators have begun to analyse the results of trials and to produce tables of awards that have been given for different kinds of physical injuries. One American tabulation made in 1951 covers three hundred and sixty double column pages referring to cases of excessive damages, and fifty referring to inadequate damages, arranged alphabetically from "abdomen" to "vertigo" and "wrist", with an index of the occupations of the injured persons and other matters: 16 American Law Reports Annotated 2d. 29; see also the American classification referred to in Munkman, Damages for Personal Injuries (1956) p. 123. These painstaking productions, of which there are now examples in England and here as well as in the United States, are it seems intended to assist in that form of the judicial process that Cardozo J. deprecated, the mere matching of the colours of the case in hand against the colours of samples spread out upon a desk. Yet in this field, more perhaps than in any other, circumstances alter cases. Judges have therefore expressed doubts about the value of these collations of desiccated fact. Nevertheless, when the validity of a monetary measure of what is in reality incommensurable is in question, judges must, consciously or unconsciously, depend upon their knowledge of awards in other cases, whether it be derived from personal experience or gathered otherwise. But what it is said we should look to is not the mean of other awards. It is what, in the language of statistics, would be called the mode. And it is not the mode of verdicts that juries have given, but the mode of those verdicts that courts have not set aside. (at p325)

23. Assuming that in some way it can be said that a particular amount would be a reasonable award in a given case, how does one determine how far it may be exceeded before it becomes too much and liable to be set aside? It was at one time not unusual, both in England and in New South Wales, in cases where damages were considered excessive, for a new trial to be ordered unless the plaintiff would consent to judgment for the sum the court thought proper. This practice, which is followed in the United States under the name "remittur", was declared unlawful by the House of Lords in 1905: Watt v. Watt (1905) AC 115 . In terms of money, verdicts are today very much higher than they were in the past, the not far distant past. This is the result of various factors, to some of which I referred in Thatcher v. Charles [1961] HCA 5; (1961) 104 CLR 57 . Notably they are the decrease in the purchasing power of money, the effect of insurance, the influence of a generally higher standard of living in the community and of some lessening of thrift in its older forms. Apart from the other factors the depreciation of the currency has influenced verdicts for damages so greatly that comparisons become of no use unless some adjustment is made. This has perhaps been longer appreciated in America than here: see the note, Fluctuating Dollars and Tort Damages Verdicts, in (1948) Columbia Law Review 264 Furthermore, awards of general damages tend to be higher in some localities in Australia than in others. The same thing has occurred in different parts of the United States. (at p326)

24. Further difficulties remain. A court cannot know what the jury thought of the evidence, how they looked at the plaintiff's experiences in the past or how they foresaw his future. When damages are said to be excessive, it is to be assumed that the jury took the most favourable view possible of the plaintiff's claim, that is that they took the most serious view possible of his plight and entertained the utmost pessimism as to his prospects. But what that actually means in a case like this it is impossible to know. Even the future economic consequences of the injury could not be predicted with any certainty. How the jury saw the plaintiff's future we cannot know; and we must resist the temptation to draw our own inferences from the evidence and test the jury's verdict by them. (at p326)

25. Juries are told, of course, that they must use their common sense, that they must not give an extravagant sum, that they must not be unreasonable, that if they do their award will be set aside, that they must give fair compensation. And, as the law is, that perhaps is all that they can properly be told. Their task may, therefore, be fittingly described by a sentence Lord Goddard, then Goddard L.J., used in his judgment in Mills v. Stanway Coaches Ltd. (1940) 2 KB 334 "Of course, different minds have different ideas as to what is moderate and seeking for a mean, a normal or an average, where there is really no guide is very like Lord Bowen's illustration of a blind man looking for a black hat in a dark room" (1940) 2 KB, at p 349 . (at p326)

26. Jurymen are carefully denied knowledge of what was done in other cases. The validity of their verdict does not depend upon whether a court approves it, but on whether it is the result of an honest performance of their duty. Of that, conformity with a supposed standard of measure that is both imprecise and unknown to them seems hardly an appropriate test. Courts should, I think, be very slow to accuse a jury of having failed to perform the duty they were sworn to perform. If honestly, without prejudice or partiality they considered the evidence and arrived at a verdict, they did their duty. I may quote here a passage from Duncombe's Tryals per Pais referred to above. The first edition was published in 1665. The earliest I have seen is the fourth published in 1702, where the passage appears at p. 332: "And it's a reasonable kindness to jury-men to make good construction of differing judgments among them, while we see how oft judges themselves differ in their opinions, on a matter stated equally to them all, and that not only as to matter of law but as to matter of fact". (at p327)

27. In this case a jury of four found a verdict which three judges of the Supreme Court think should stand. Three judges of this Court think it should not, and one thinks that it should. Personally it seems to me to be a high verdict. But most awards of general damages seem to me to be high, and getting higher. But that does not mean that they ought to be diminished. Phrases such as "an amount out of all proprortion to the facts", "damages so large that a sensible jury could not reasonably have awarded them", "an entirely erroneous estimate", and all the rest with which we have become so familiar, have no certain meaning - that is to say they are not referable to any established objective criterion. They describe limits but they do not, for me, define them. And they are, I think, at times conveniently quoted to give an appearance of objectivity to a subjective opinion. If, however, such statements are of assistance I would say that, having regard to the different opinions in this case, I am not convinced that the decision is "such as would be clearly wrong in the judgment of the great majority of ordinarily reasonable and fair men". The words are those of Brett J., as he then was, in Bridges v. North London Railway Co. (1874) LR 7 HL, at p 235 . (at p327)

28. Summed up, it comes to this: Within the limits - variously described, undefined, and as time goes on shifting - of reasonableness the jury had a discretion to assess the damages according to their conclusions from the evidence. The Supreme Court, being the Court in which the action was tried at nisi prius, had a discretionary jurisdiction, according to common law principles, to grant a new trial. It unanimously refused to do so. Nothing was established that in my view compelled it to find that, in Lord Halsbury's words, the jury were "misled by prejudice or passion" or "failed to perform their duty". In these circumstances I think this Court ought not to interfere with the decision of the Supreme Court. I would dismiss the appeal. (at p328)

ORDER

Appeal allowed with costs.

Order of the Full Court of the Supreme Court of New South Wales set aside. In lieu thereof order that the appeal to that Court be allowed with costs, that the verdict of the jury be set aside so far as it assessed the amount of damages and that there be a new trial of the action limited to the assessment of damages.