High Court - Appeal - Decision of State Supreme Court on appeal from primary judge - Inferences of fact involved - Function of High Court.

Master and Servant - Liability of master for injury to servant - System of work - Demolition contractor - Failure to instruct employees - Contributory negligence.

APPEAL from the Supreme Court of Western Australia.

BARWICK C.J. The appellant, a labourer, was employed by the respondent along building was of brick, with an iron roof. The roof consisted of three gables and a skillion section. The pitch of the gables was sufficient to make it difficult for a man to stand on the roof without supporting himself by some means. The respondent's foreman told the plaintiff and two fellow workmen, Ellio Pedri and Erico Bartolomei, to go up on the roof and take the iron off. No other instructions were given on behalf of the respondent. The workmen gained access to the roof by backing a truck against the portion of the building which had the skillion roof and by climbing from the top of the truck on to the roof. The appellant and Pedri went to a gable section. It was understood between them that the appellant would work on one side of the ridge and Pedri on the other. Apparently, Bartolomei was to work on the skillion section of the roof. (at p194)

2. The appellant and Pedri had worked together before on demolition work and the appellant had had a little experience of such work. The procedure which they each understood had to be employed was to remove the ridge capping first and then the sheets of iron from the roof. The nails fixing the ridge capping also went through and secured the upper ends of the sheets of iron nearest the ridge. It would seem that the roof was formed of sheets of iron each six feet in length. The tool employed by the men to remove the nails and the sheets of iron was a pinch-bar. The appellant said that he understood the method to be employed on a roof of this pitch was that after the ridge capping had been removed the remaining nails would be taken from each sheet of iron in turn, the sheet being dropped from the roof when all the nails had been removed from it. Using this method of work there would not at any time be any sheet in position on the roof which was not secured by a nail or some nails. The appellant agreed that at times on a flatter roof all the nails from all the sheets of iron may be removed before an attempt is made to dislodge any of the sheets of iron. But, at best, only an exercise of judgment on somebody's part could determine which procedure in a particular case was a safe method of work in demolishing an iron roof. Of course, if a workman skilled in such work was working alone, he could, in my opinion, properly be left to decide for himself which method he would employ. Cf. O'Connor v. The Commissioner for Government Transport [1954] HCA 11; (1954) 100 CLR 225 . But here there were at least two workmen, neither of whom was in charge of the other and the method of working on the demolition of the roof was not specified. (at p194)

3. The two men began working together each on opposite sides of the ridge taking out the nails from the ridge capping. Apparently they arranged that as between them it would be the appellant's duty to remove the ridge capping when it was free for removal. As they worked along the ridge removing the nails Pedri seems to have been the faster workman and reached the end of the ridge on his side before the appellant reached the end on his side. Pedri then began to work back along the roof removing some nails from the sheets of iron. He had progressed the width of two or three sheets of iron away from the end of the ridge before the happening of the incident out of which this action has arisen. (at p195)

4. When the appellant had completed removing the nails from his side of the ridge capping and understood that Pedri had removed all the nails from the other side of the ridge capping the appellant attempted to remove the first section of the ridge capping which would be some six feet in length. However, it did not yield to his pressure, applied whilst he was wholly on his own side of the ridge. He checked and found that there were no remaining nails on the capping on his side of the ridge. He concluded that something must be holding the capping, perhaps a nail left in position by Pedri by oversight or error. Accordingly, in order to examine what was impeding the removal of the capping, the appellant threw his leg over the ridge in order to straddle it. At this point of time in fact Pedri had removed all the nails at least from the sheet immediately opposite where the appellant had completed working : the one on which he put his foot as he attempted to straddle the ridge. Pedri may also have removed all the nails from each of two other sheets. However, the appellant, as he manoeuvred himself to mount the ridge, would see on Pedri's side of the roof a sheet of iron in position and undisturbed. He could not of necessity see that all the nails had been removed from it. He said that he knew that Pedri "had taken off the nails but did not know Pedri had taken off all the nails and left the sheet behind", although he knew Pedri was some minutes ahead of him in the work. He could have called out and asked Pedri what he had done but he said "that they had never done that job before like that" meaning the removal of all nails from a sheet of iron leaving it in place on a roof of that pitch. The appellant assumed that because of the way they had removed iron from a roof on prior occasions, all the nails had not been removed from a sheet of iron left in its place on the roof. (at p195)

5. When the appellant "sort of swung" his leg over the ridge and placed the weight of his foot on the sheet it collapsed and the appellant fell to the ground. His account of what happened is brief but adequate :

"You came up one side crouching with your hand on top of

the ridge cap? - - - Yes.

You put your foot over the other side and as soon as you

put your foot on the iron it moved? - - - Yes, and I fell

straight down."

At that point of time Pedri was the width of two or three sheets of iron away from where the appellant was. Pedri, who gave evidence, said that he had decided to remove all the nails in all the sheets in the top row of the sheets of iron before dropping any of the sheets of iron off the roof because it was easier for him to work in that fashion. When he began to drop the sheets down to the ground he said he would have both his hands free and not be encumbered by a pinchbar in his hand. There was some dispute at the trial as to whether the sheets of iron at least on the side of the gable on which Pedri was working were secured by three rows of nails or only two rows of nails. The trial judge found that the sheets were fixed by only two rows of nails so that, nails having been taken out of the ridge capping, if Pedri removed all the nails in the second row of nails the sheets of iron which lay between him and the appellant would have been totally unsecured, as in fact the sheet on which the appellant placed his foot was. (at p196)

6. The primary judge found that the roof was steep enough to make it difficult to stand on and that workmen had to hold on to the ridge capping with one hand while removing the nails. He said :

"When told by the foreman, Panizza, to get up on the roof

and take the iron off, he had not spoken to the plaintiff about

how they would do the job. The plaintiff also said in evidence

that there had been no discussion between them about how

the job was to be done except that when they had climbed up

onto the roof, Pedri, who had got up first and was on the far

side of the ridge, had said 'I go this side' and he, the plaintiff,

on the near side, said 'I start this side'. The foreman,

Panizza, had given them no instructions as to which method

they were to adopt but had simply told them to go up on the

roof and take the iron off." (at p196)

7. In finding that the plaintiff had proved negligence on the part of the defendant which caused or contributed to his injuries, his Honour said :

"Neither the plaintiff nor Pedri was a skilled workman,

doing work which he had been trained to do. . . . Once again

neither was given any instructions as to the precautions to be

taken, but were simply told what to do and left to themselves

as to how they did it. In those circumstances it is hardly

surprising that they should in some circumstances have different

ideas as to the proper way to go about the jobs they were

given to do, and if in such a case they were working together

on the same job, where the manner in which one did his work

might affect the safety of the other, one would have expected

someone in authority to decide on the method to be adopted

and to see that the method was used by the workers in the

team." (at p196)

8. After pointing out that there were methods which differed according to the slope of the roof and that the question of whether the slope was appropriate to one method or another was a question for judgment he said :

"In those circumstances where different workers might use

different methods of work and where the use of such different

methods might create unnecessary dangers to the workers in

doing the work, it seems to me that an employer properly

mindful of his duty to use reasonable care for the safety of his

workmen, would by his foreman either have directed the method

to be adopted by the workers in each particular case or at

least have directed the workers themselves to decide on which

method should be used." (at p197)

9. The trial judge however also found that the appellant showed some lack of care for his own safety which was contributory to the injuries he suffered. In this connexion he said this :

"He" (that is the appellant) "knew that Pedri had finished

first and was working back towards the end at which they

began and should therefore have realized that Pedri had

removed some nails, additional to those which first went through

the ridge capping, from the sheets of iron on which he was

proposing to stand. It is true that the plaintiff might not then

have noticed whether there were two or three rows of nails in

each sheet of iron and it is also true that he thought there was

still a nail through the ridge capping on Pedri's side, because

he could not lift the ridge capping off. Nevertheless, that was

only a supposition - there could have been other reasons for

the difficulty in moving the ridge capping - and I think that in

those circumstances ordinary prudence for his own safety

should have demanded either that he first test the security

of the sheet of iron or, if that were impracticable, ask Pedri,

who was only some three or four yards away from him, as to

whether the sheets of iron were still secure."

Accordingly the primary judge found the plaintiff to be fifty per cent responsible for his own injuries. (at p197)

10. The appellant appealed against this decision to the Full Court of the Supreme Court of Western Australian claiming that there should have been no apportionment of responsibility : the respondent cross-appealed seeking to set aside the verdict. That Court, by majority, dismissed the appeal and allowed the cross-appeal. The judgment of the primary judge was set aside and judgment entered for the respondent. (at p197)

11. The majority judgment, referring to appropriate decisions of this Court, rightly emphasized that an employer's duty is to take reasonable care for the safety of his workmen. This principle of law the trial judge undoubtedly accepted and applied. He concluded that in point of fact the respondent did not take reasonable care for the safety of his workmen. There was, in my opinion, unquestionably evidence upon which his Honour could so conclude. The roof on which he asked the workmen to work was evidently of such a pitch that it would be unsafe to allow sheets of iron to remain in position apparently undisturbed and not secured in any way. The danger to workmen inherent in a sheet of iron thus left in place is to my mind obvious and was something which a reasonable employer ought to have foreseen as a potential source of danger to a workman who was unaware of the fact that the sheet had in fact been unnailed and was free to move. The possibility that a workmen intent on his task might very well step on the sheet thinking it safe to do so because of its undisturbed condition and the unlikelihood that it would be left on that roof in position unsecured were each something which, in my opinion, a reasonable employer ought to have foreseen. The danger which such a loose sheet on such a roof would constitute was something which was very easily avoided. Clear instructions to the workmen either as to a settled manner of doing the work or as to the need to co-ordinate their efforts by arranging as between themselves how the work would be done was to my mind both a simple and an obvious precaution to have been taken in the circumstances. Consequently there was to my mind abundant evidence upon which the primary judge could reach the conclusion which he did. (at p198)

12. The matter before the Full Court was whether or not the primary judge was wrong to so conclude. The majority of the Full Court seemed to think that the right view of the facts established at the trial was that the appellant, knowing or in a position where he ought to have known tht the sheet in question had been fully unnailed and was in a dangerous condition, deliberately stepped upon it in a foolhardy manner. Of course, if that had been the fact it would be right to conclude that he was the author of his own injuries. Cf. Commissioner for Railways v. O'Brien (1958) 100 CLR where the workman was aware of the dangerous condition of the roofing. But, in my opinion, that was not a proper view of the facts. The primary judge believed the appellant. The appellant expressly said that he did not know that all the nails had been taken out of the sheet upon which he put his foot and that the system which he understood to be the proper system to work on such a roof would not have left the sheets in position unsecured. Consequently, in my opinion, it is not proper to conclude that in fact the appellant did a foolhardy thing in placing his foot on the sheet which was in position and apparently undisturbed. It is of course quite true had he been in the position of a person having time to reason the whole situation out he might possibly have inferred from the fact that his fellow workman had progressed so far along the roof as he obviously had that there may be a question as to whether or not the sheets had been wholly unnailed, a question he would resolve by enquiry. But the appellant was not in the position of a person calmly considering what inferences might be drawn from what he said. He was busy at his task and was not, in my opinion, neglectful of his own safety in placing his foot on the sheet which so far as his actual knowledge was concerned was not in a dangerous situation, and in circumstances in which he was entitled to assume that the sheet would not have been left in position in an unsecure state. As I am of opinion that there was material upon which the primary judge could find that the respondent failed to take reasonable care in the circumstances for the safety of his workmen and as I cannot agree that the facts were as the majority of the Full Court seemed to think they were, I am of opinion that the Full Court ought not to have disturbed the decision of the primary judge. As an additional reason for thinking that the appeal should be allowed I may say that in my own personal view the primary judge was right in concluding as he did that the respondent was negligent. (at p199)

13. However, there remains the question whether the primary judge was wrong in concluding that the appellant failed to take reasonable care for his own safety and that this failure contributed to his resulting injury. For my own part, had I the initial task of deciding that question I would conclude that the appellant did not fail to take care for his own safety when he placed his weight on the sheet of iron. My reasons for thinking so sufficiently appear in what I have so far written. However, this is an appeal. My own view as to the limits within which the appellate court ought to confine itself in reviewing findings of fact have been expressed elsewhere. See Whiteley Muir and Zwanenberg Ltd. v. Kerr (1966) 39 ALJR 505 . Further reflection has not caused me to modify that view. The question of contributory negligence in this particular case is, it seems to me, one on which reasonable minds may differ, both in the answer given and in the reasons therefor. After consideration and not without some hesitation, I have reached the conclusion that notwithstanding my own opinion I ought not to hold that the trial judge was wrong to conclude as he did that the appellant failed to take reasonable care for his own safety. (at p199)

14. Consequently, in my opinion, the Full Court ought not to have disturbed the verdict and judgment of the trial judge. In my opinion, this appeal should be allowed, the order of the Full Court set aside and in lieu thereof both the appeal and the cross-appeal to the Full Court dismissed. (at p200)

MENZIES J. The plaintiff, while working upon the demolition of a small house, fell through the roof and suffered injuries. He sued his employer for damages for negligence. The learned trial judge found that the plaintiff and the defendant were both negligent and were equally blameworthy. Accordingly, he awarded the plaintiff one half of the assessed damages, viz. $10,463.05, less workers compensation payments already made $2,797.62, i.e. $7,665.43. The Full Court, by a majority, dismissed an appeal by the plaintiff and allowed a cross-appeal by the defendant, resulting in judgment for the defendant in the action. (at p200)

2. The plaintiff and a fellow worker named Pedri were removing corrugated iron from a roof, each working upon one side of a ridge. Each worker began opposite the other removing nails which held the top of the first row of iron sheets and the capping which covered the ridge. Each worked in the same direction and Pedri, having worked more quickly than the plaintiff, descended his side of the ridge to remove the nails at the bottom of the first row of iron, this time working back in the opposite direction. It fell to the plaintiff, as the last to finish the first row of nails, to take off and throw down the capping. He found some difficulty in doing this, and, thinking that Pedri might have left a nail undrawn, he stepped across the ridge upon a sheet of iron from which Pedri had drawn the nails at the bottom as well s at the top. This action dislodged the sheet of iron and the plaintiff fell through the roof and was injured. (at p200)

3. The workers were not instructed how to do their work and the plaintiff gave evidence that once he had drawn all the nails holding a sheet of iron he would remove the iron and throw it down before proceeding to draw nails in other sheets of iron. Pedri did not follow this practice, and, as I have said, was working along the line of nails at the bottom of the sheets from which the top nails had already been taken, leaving the sheets in position until he had withdrawn all the nails in the row. He would then throw down the sheets. It was obviously not the practice of either to draw some nails and leave others to be withdrawn later. (at p200)

4. The mode of working which the plaintiff said he intended to follow and that in fact followed by Pedri were both normal ways of doing what was a simple and straight-forward job for men accustomed to the work. (at p201)

5. It must have been apparent to the plaintiff that Pedri was working back leaving the sheets in position after he had withdrawn the bottom nails. The plaintiff knew that Pedri had finished the first row of nails and had drawn some nails from the second row but had not removed any sheet of iron. Pedri had worked back to a point beyond that at which the plaintiff stepped over a ridge. (at p201)

6. In these circumstances I think that the finding of the learned trial judge, supported by the decision of the majority of the Full Court, that the plaintiff was negligent in stepping upon a sheet of iron from which all the nails had been taken, is not one with which this court should interfere. This I regard as the critical finding. Moreover, like the majority of the Full Court, I consider that it follows that the plaintiff's negligence in stepping upon a sheet of iron from which all the nails had been drawn was the cause of the accident. Pedri was not negligent in working as he did, nor was the defendant negligent in not instructing two experienced workmen how to do a simple job. (at p201)

7. I agree with the judgment of Hale J. and would dismiss this appeal. (at p201)

WINDEYER J. This is an appeal from the Full Court of the Supreme Court of Western Australia which reversed a finding of Nevile J. in favour of the plaintiff in a negligence action. His Honour had found negligence on the part of the defendant, and contributory negligence of the plaintiff making him responsible for half the damages. (at p201)

2. Lord Denning once said, "we must not condemn as negligence that which is only a misadventure" : Roe v. Minister of Health (1954) 2 QB 66, at p 87 . If I had had to try this case at first instance, I would have reminded myself of that ; and, had I found the facts as they are now put before us, I feel fairly sure that my decision would have been in favour of the defendant. But I was not the trial judge. Nevile J. was. And, strictly, the question for us in this Court is not was his conclusion right. It is, was the Full Court right in saying that it was wrong, and in setting his judgment aside? (at p201)

3. An appeal has been described by the Privy Council as "the formal proceeding by which an unsuccessful party seeks to have the formal order of a court set aside or varied in his favour by an appellate court" : Commonwealth of Australia v. Bank of New South Wales (1950) AC 235, at p 294 ; [1949] HCA 47; (1949) 79 CLR 497, at p 625 . Lord Westbury has said in 1864 that "an appeal is the right of entering a superior court, and invoking its aid and interposition to redress the error of the court below" : Attorney-General v. Sillem [1864] EngR 352; (1864) 10 HLC 704, at p 724 [1864] EngR 352; (11 ER 1200, at p 1209) . Appeals are creatures of statute. They were not known to common law as distinct from Chancery procedure : Grierson v. The King [1938] HCA 45; (1938) 60 CLR 431, at p 436 . The common law process to correct a mistake of law made by a judge at nisi prius was writ of error. After attaints had disappeared there was, practically speaking, no process to remedy mistakes of fact by a jury until the advent of new trials in the second half of the seventeenth century. Appeal being a remedy given by statute, it is necessary to observe what powers are conferred upon the appeal court by the relevant statute, for all such courts are not in the same position. And the word "appeal" has itself more than one sense for modern law : e.g. Ex parte Australian Sporting Club Ltd.; Re Dash (1947) 47 SR (NSW) 283 . The origin of this Court's jurisdiction to hear appeals from the Supreme Court of the States is s. 73 of the Constitution. The manner of its exercise is regulated by the Judiciary Act 1903-1959 (Cth), ss. 36 and 37. The latter provides 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 . . .". That, as I understand it, involves our considering the judgment of the learned trial judge in this case, not so much to determine whether we would have come to the same conclusion on the facts, but rather to determine whether or not the Full Court should have disturbed it. (at p202)

4. I shall first state the essential primary facts as the learned trial judge found them. The appellant in this Court was the plaintiff in the action, the respondent was the defendant. I shall for convenience call them by those names throughout. (at p202)

5. The defendant company is a contractor engaged in the demolition of buildings. The plaintiff was a labourer employed by it. On 28th August 1967 the plaintiff and one Pedri, another servant of the defendant, were removing the corrugated iron roof of an old building at Fremantle which the defendant was demolishing. The plaintiff and Pedri had done work of this kind for the defendant on other occasions. On this day they were simply told by the defendant's foreman, under whose direction they were, to get up on the roof and take off the iron. They were each provided with a pinchbar with which to extract the nails holding the iron sheets to the purlins. At the critical time they were working at a fairly steep gable of the roof, one on each side. They had taken out the top 124 C.L.R.) row of nails on each side. The plaintiff, working from his side, then tried to remove the metal ridge cap. He found it did not come away readily. Assuming that perhaps a nail had not been removed, he sought to pull it harder, and for this purpose to straddle the ridge. He put one foot over from his side to the other side, on to a sheet of iron still in position there. It collapsed under his weight. All the nails holding it, not only the top row, had been taken out of it by Pedri ; and there was thus nothing to hold it in place. The plaintiff fell to the ground. He sustained a fracture of his right elbow resulting in some impairment of function of his right arm. He says that he had assumed that the sheet on which he had sought to stand was firm because it was there. His own practice, when removing a sloping roof, was to throw each sheet to the ground as soon as it was loose, and he supposed that Pedri would probably have done the same thing. Pedri gave evidence that that was not what he usually did. Even when working on a sloping roof, he preferred to loosen all the sheets of a row before removing any of them, his reason being that he might need both hands to send a sheet to the ground, and he did not want to put his pinch-bar out of his hand until he had completed a row of nails. (at p203)

6. The accident obviously occurred in the course of the plaintiff's employment, and he has been in receipt of compensation pursuant to the workmen's compensation law of Western Australia. However, in January 1969 he commenced this action for damages, alleging negligence on the part of the defendant, of which particulars were given with the statement of claim. The defendant alleged that he was guilty of contributory negligence. The plaintiff's particulars seem designed to embrace in general terms different aspects of the duty of an employer to take due care for the safety of his workmen, matters which are nowadays compendiously called the duty to provide a safe system of work. Some of the allegations seem to be derived from law reports rather than based on the facts of the instant case. However, they are clear enough. Added to them, or rather inserted among them, are what seem to be not so much allegations that the defendant had adopted an unsafe system of work as allegations that it was vicariously liable for particular negligent acts and omissions by Pedri. As to these, I agree entirely with what Hale J. said of them in his judgment. It cannot be said that Pedri failed in any duty of his of care for the plaintiff or that any conduct of his made the defendant liable. (at p203)

7. I set out the particulars of negligence in full because they were quoted by the trial judge in his judgment and he had to consider how far they were established in fact. They were as follows :

"The defendant was negligent in that it by its servants :

(a) Failed to provide safe and suitable equipment for removing

the roofing iron without unnecessary risk to the plaintiff.

(b) Failed to organize a safe system of removing the said iron

without unnecessary risk to the plaintiff.

(c) Failed to supervise the said work so as to prevent

unnecessary risk to the plaintiff.

(d) Failed to warn the plaintiff that nails had been removed

from the said iron, or that it was unsafe for him to place

any weight on it.

(e) Removed the said nails from the lower edge of the said

sheet before the nails had been removed from the top edge.

(f) Failed to take reasonable care for the safety of the plaintiff

while he was engaged in the said demolition work."

Paragraphs (d) and (e) apparently refer to Pedri's actions or inactions. The latter of them does not seem to have any bearing on the case in the light of the evidence. Hale J. has said that "on the facts of the case it is meaningless". I agree. Paragraph (d) can hardly be said, as a practical matter, to allege a want of care on the part of Pedri or anyone else. It was not shewn that Pedri had any warning of the plaintiff's intention to straddle the ridge which would require him to warn him against his doing so. Paragraph (f) is merely an assertion in general terms that the defendant failed in the duty of care an employer owes to his workmen. It is in no sense a particular of the manner of the failure. That leaves only pars. (a), (b) and (c). Paragraph (a) can be ignored. No evidence was given to support it. The plaintiff was supplied with a pinch-bar and simply told to get on with the job. There is nothing to suggest that he needed any other equipment or that the tool he was given was not safe and satisfactory. (at p204)

8. The case for the plaintiff therefore turned on the general allegations, in pars. (b) and (c), that the defendant had failed to provide a safe system of work or had failed to exercise proper supervision of the work to prevent unnecessary risk to the plaintiff. Nevile J. considered that in substance these charges were made out as the plaintiff and Pedri were allowed to go on the job without any instructions as to how they were to perform it and without any supervision of how they were in fact doing their work. What was principally relied upon was a failure by the defendant's foreman to co-ordinate their methods - with the result, it was said, that the plaintiff might not unreasonably assume that work on Pedri's side of the ridge was being done in the way he was doing the work on his side, and that he was thus led to think that any sheet of iron that had not been removed was safe to stand on. Nevile J. accepted this view of the facts. He held that the defendant was negligent. This necessarily meant that it could and should have foreseen the possibility of such a mishap as occurred, that it was at fault as it had taken no steps to avoid such a happening. I feel bound to say that this is a possible view of the facts. Nevertheless I would myself have thought it far-fetched, for the reasons that Hale J. vigorously expressed in his judgment in the Full Court. (at p205)

9. The case, as presented, seems to me to be another instance of a tendency to say that simply because an accident happens to a man at work, which conceivably could have been by some means avoided, the system of work was unsafe and the injured man's employer was responsible. That, as Taylor J. said in Rae v. Broken Hill Pty. Co. Ltd. [1957] HCA 33; (1957) 97 CLR 419, at p 430 , has been said many times to be "a completely erroneous approach to the problem". Yet it is an approach still sometimes taken, and this has been one of the factors producing what Kitto J. in Electric Power Transmission Pty. Ltd. v. Cuiuli [1961] HCA 3; (1961) 104 CLR 177, at p 180 spoke of as "the tendency which has been shown by some courts in recent years to put at altogether too high a level the duty of an employer with respect to the safety of his employee". In that case Taylor J., Dixon C.J. agreeing, said as to the facts there in question (1961) 104 CLR, at p 183 :

"If one seeks far enough it is possible to perceive an element

of risk in the performance of any task however simple but the

duty of an employer does not extend to guarding against every

conceivable risk however remote or fanciful."

The same thing was said again in Vozza v. Tooth & Co. Ltd. [1964] HCA 29; (1964) 112 CLR 316, at p 318 . It is always important to remember that the fact that an accident happens to a man at work is not of itself evidence that those in control of the work or responsible for the manner of its performance were negligent. If a man be unreasonably exposed to the risk of injury that fact must ordinarily be apparent before he was injured. An intelligent, instructed and reasonable observer must then have foreseen the possibility of an accident. It must have been clear then that the employer was failing in his duty to take reasonable care for the workman's safety. (at p205)

10. I would not myself have thought that in the present case the defendant ought to have made known to the plaintiff and Pedri how each was going about his work or tell them how to do it. An experienced man who works at a simple task in company with another man can, I would think, reasonably be left to ascertain from his workmate anything as to his method of work that may affect his own safety. It is not as if there was evidence of some regular recognized practice from which Pedri had departed. Nor was there any evidence of it being the usual practice in demolition work to direct men as to their precise way in which each should act in a task they were to do together. I have said so much to explain why I should have reached a conclusion different from that of Nevile J. on the question whether the defendant failed to take due care for the safety of its workmen. (at p206)

11. I turn now to his Honour's finding that the plaintiff was himself negligent. I would not question that his action, based on a belief, without any test or inquiry, that the iron would hold him, was risky. But whether it should be said that he failed to take the care for his own safety that a reasonable man in his position ought to have taken is not to my mind clear. And the proposition is made for me the more obscure when I find that other judges think it clear, yet see it differently. Had it been my task to try the case, I would probably have seen the whole occurrence as an unfortunate accident for which neither party was to blame. It seems to me that what happened could be fairly regarded as a mere mishap the proximate, but not the sole, cause of which was a natural mistake that the plaintiff made in a heedless moment when going about his work. However, the question of contributory negligence, like that of negligence, was for the trial judge. I am far from saying that there was not evidence on which he could reasonably reach on each the conclusion that he did. I am not prepared to say that he was wrong simply because I would probably have taken a different view. The case, as I see it, thus raises again what this Court has called "the ever-recurring question how far a court of appeal is justified in going in reviewing a finding of fact made on oral evidence" : Riebe v. Riebe [1957] HCA 66; (1957) 98 CLR 212, at p 222 . A brief formulation by this Court of the answer to that question is as follows :

"On an appeal from a judge of fact it is the duty of an

appellate court to examine the evidence itself, due regard being

had in weighing that evidence to any opinion formed by the

trial judge of the credibility of the witnesses and to those

advantages he derives from the general atmosphere of the

trial" : Committee of Direction of Fruit Marketing v. Spence

(1953) 27 ALJ 427, at p 428

.

There are more elaborate statements of the same principle in many cases which are collected and analyzed in the judgment of Dixon C.J. and Kitto J. in Paterson v. Paterson [1953] HCA 74; (1953) 89 CLR 212, at pp 219-224 . When the decision of a judge sitting without a jury is open to appeal on both law and fact, an appellate court must have regard to the evidence, and its effect so far as the written transcript reveals it. Lord Halsbury said in Riekmann v. Thierry (1896) 14 RPC 105, at p 116 - in a passage quoted by Isaacs J. in Dearman v. Dearman [1908] HCA 84; (1908) 7 CLR 549, at p 560 , and more recently by Lord Reid in Benmax v. Austin Motor Co. Ltd. (1955) AC 370, at p 376 - that on an appeal there is no presumption that the judgment in the court below is right. But ambiguity lurks there in the word "presumption". For Lord Esher M.R. had said sixty years before that where a case came before the Court of Appeal on appeal from the decision of a judge without a jury the Court would presume that the judge's decision on the facts was right and would not disturb his judgment unless the appellant satisfactorily made out that it was wrong : Colonial Securities Trust Co. Ltd. v. Massey, (1896) 1 QB 38 . That case is cited in Halsbury's Laws of England, 3rd ed., vol. 30, p. 471, for a statement that "the presumption is that the decision appealed against is right". In the article, now in The Law Quarterly Review, vol. 71, p. 402 - to which Lord Simonds acknowledged his indebtedness in the Benmax Case (1955) AC 370 - Professor Goodhart, expressly preferring the word "presupposition" to "presumption", said that it had become accepted doctrine that "the appellant, in seeking for the reversal of the judgment below, must establish his case, and that there is a presupposition that the trial judge was correct in his conclusion". It has been said many times that the decision of a judge of fact ought not to be set aside on appeal unless the appellate court, duly considering the advantages the judge had had in hearing the evidence, is satisfied that his conclusion was wrong. In Powell v. Streatham Manor Nursing Home (1935) AC 243, at p 250 , Lord Sankey said it must be "plainly wrong". There have been differences in emphasis in judicial affirmations of this doctrine. At one extreme are the statements that the judge of the facts must have been "plainly wrong" or "clearly wrong" before his judgment will be set aside. At the other end is Lord Cave's insistence that it is "the duty of the Court of Appeal to make up its own mind, not disregarding the judgment appealed from and giving special weight to that judgment in cases where the credibility of witnesses comes into question, but with full liberty to draw its own inference from the facts proved or admitted, and to decide accordingly" : Mersey Docks and Harbour Board v. Procter (1923) AC 253, at pp 258-259 . Whatever words be used, I think that it must appear convincingly that the judge of fact came to a wrong conclusion before it is reversed. I consider that the statement by the Chief Justice in Whiteley Muir and Zwanenberg Ltd. v. Kerr (1966) 39 ALJR 505 points the approach for this Court to take. (at p208)

12. There is a technical distinction between a simple right of appeal and an appeal expressed to be by way of re-hearing : see Attorney-General v. Vernazza (1960) AC 965, at pp 974-975, 978, 982 . This was considered historically in its bearing on the jurisdiction of this Court, and the relevant authorities were collected, by Dixon J. in Victorian Stevedoring and General Contracting Co. Pty. Ltd. and Meakes v. Dignan [1931] HCA 34; (1931) 46 CLR 73, at pp 107-110 . His Honour's observations there shew that, in the exercise of its appellate jurisdiction, this Court has full authority to decide whether a judgment of the court below ought or ought not to have been given on the evidence before it seen in the light of the law as it then stood. Such an appeal, on both fact and law, although limited to the material before the court below, may be conveniently called a re-hearing, provided it is not thought that the use of this term without statutory authority enlarges the powers of the court. But, strictly speaking, it may not be appropriate to call an appeal to this Court from the Supreme Court of a State a re-hearing, as Isaacs J. pointed out in the course of his dissenting judgment in Werribee Council v. Kerr [1928] HCA 41; (1928) 42 CLR 1, at pp 20-21 . (at p208)

13. I turn now to the position of the Full Court of the Supreme Court of Western Australia on an appeal from a judgment of a judge of that Court sitting without a jury. This, as I understand it, depends upon ss. 58 and 59 of the Supreme Court Act, 1935-1967 (W.A.). The origin and history of these provisions in the form they had in 1957 appears from the judgment in Riebe v. Riebe [1957] HCA 66; (1957) 98 CLR 212, at pp 217-221 . It suffices here to say that, in the form these sections now have, s. 58 (1) (a) confers appellate jurisdiction upon the Full Court in any cause or matter tried or heard by a judge or before a judge and jury : and s. 59 (6) provides that in any cause or matter where there has been a trial by a judge sitting without a jury every application to set aside his verdict, finding or judgment is to be made in accordance with the relevant Rule of Court. The rule referred to appears to be O. 58 r. 1 which provides that all appeals shall be "by way of rehearing". This does not mean that the appeal is a complete rehearing as a new trial is. It means that the case is to be determined by the Full Court, its members considering for themselves the issues the trial judge had to determine and the effect of the evidence he heard as appearing in the record of the proceedings before him, but applying the law as it is when the appeal is heard not as it was when the trial occurred : see Attorney-General v. Birmingham Tame and Rea District Drainage Board (1912) AC 788, at pp 801-802 , and Attorney-General v. Vernazza, (1960) AC 965 . Order 58 provides too that the Full Court shall have power to draw inferences of fact and to give any judgment and to make any order that ought to have been made. This is more or less standard form. It confers a wide power. But it does not I think curtail the recognition or respect that an appeal court should accord to the decision of a trial judge. That has been recognized in many cases, among them in the judgment of Jordan C.J. in Carberry v. Gardiner (1936) 36 SR (NSW) 559 . Whether the express authority to draw inferences of fact be necessary may be doubted, for it would seem to be necessarily implicit in a power to hear appeals on a question of fact. Nevertheless it is a significant provision for it leads on to the matter discussed by the House of Lords in the Benmax Case (1955) AC 370 . (at p209)

14. I take one sentence from the judgment of Lord Reid in that case as a statement of the relevant principle. His Lordship said (1955) AC, at p 376 :

"But in cases where there is no question of the credibility

or reliability of any witness, and in cases where the point in

dispute is the proper inference to be drawn from proved facts,

an appeal court is generally in as good a position to evaluate

the evidence as the trial judge, and ought not to shrink from

that task, though it ought, of course, to give weight to his

opinion."

The distinction thus made between primary facts and inferences of fact is a sound one ; and its bearing upon the way in which an appeal court should approach its task has long been appreciated. For example in 1905, quite early in the history of this Court, Griffith C.J., in Luke v. Waite [1905] HCA 5; (1905) 2 CLR 252, at p 265 said :

"It was pressed upon us that this Court ought not to disturb

the conclusions of the Court appealed from on a mere question

of fact. This is, no doubt, the general rule, but it has no

application when, as in this case, there is no conflict of

testimony, and the only question of fact is as to the effect of the

facts proved in raising further inferences of fact : Thurburn

v. Steward

(1871) LR 3 PC 478

."

More recent recognitions of this in this Court, in addition to Paterson v. Paterson [1953] HCA 74; (1953) 89 CLR 212 and cases there collected, may be found in the judgments in Jones v. Capaldi [1956] HCA 37; (1956) 98 CLR 615 ; Mann v. Mann [1957] HCA 68; (1957) 97 CLR 433 . But this doctrine ought not, I consider, to be taken by judges in courts of appeal as a grant of an uninhibited liberty to review all conclusions of ultimate fact of a judge of first instance. An appeal court is, it is said, in the circumstances envisaged in as good a position to evaluate the evidence as the trial judge. This does not necessarily mean that its members are in a better position to do so than he was. Moreover it is a rule easier to state than to apply : and in Salmond on Torts, 14th ed. (1965), p. 318, it is said of the Benmax Case (1955) AC 370 that :

"While it is useful to have such clear and authoritative

guidance, it may perhaps be regretted that the law should

have been laid down in such very wide terms. The reports

contain many cases in which judges of the greatest eminence

have differed as to the evaluation of admitted facts and if

appellate tribunals are to make full use of the powers given to

them by Benmax v. Austin Motor Co. Ltd. the uncertainty of

the law will greatly increase."

15. The prevalence today of insurance against liability for negligence has tended to protract litigation till all appeals are exhausted. Many negligence actions are now tried by a judge alone, not by judge and jury : and a court hearing an appeal from a judge has his reasons and his findings of primary facts; and it can examine his conclusion in the light of those reasons and those facts. When, as in this case, the matter on which legal consequences ultimately depend is the evaluation of the evidence by the standard of care expected of a reasonable man, judges in courts of appeal are more free, and often more ready, to set aside the decision of a trial judge than they would be the verdict of a jury. And they are free to differ among themselves, so that the final determination of legal rights and liabilities may be arrived at by a bare majority in a final court of appeal after differences of opinion among judges below. And the trial judge was a reasonable man : so were they all, all reasonable men. The question whether conduct was or was not negligent is not a question of law. Nor is it strictly a question of fact. It is a matter of judgment upon facts, a matter on which minds may rationally differ. I shall say more of that later. Some of those who like Latin tags may say resignedly "tot homines quot sententiae" : but others, remembering "interest reipublicae ut sit finis litium" and "interest reipublicae res judicatas non rescindi", may have grave misgivings about the social value of the whole process, as I have. (at p210)

16. The present case is a striking illustration of the conflicting outlooks that judges may have upon facts not in dispute. Nevile J. held that the defendant was negligent in that it "did not take reasonable care for the plaintiff's safety" but that "the plaintiff also showed some lack of care for his own safety", amounting to contributory negligence. In the Full Court, Hale J., with whom Virtue J. agreed, reached a different conclusion. He thought that it could not be said that the defendant had failed to take reasonable care for the safety of its workmen : he agreed that the plaintiff was careless of his own safety, and placed responsibility for the accident entirely on his "foolish act". Burt J. took a different view again. Speaking of the trial judge's conclusion that the defendant was negligent, he said : "After much consideration I find myself unable to say that he was wrong" ; and he concluded on that aspect : "In my opinion the finding of negligence against the employer was open on the evidence and should not be disturbed." He went on to consider the finding of contributory negligence. As to that, he considered that the trial judge had overlooked "as a critical fact in the situation", his finding that the plaintiff had a natural expectation that the sheet of iron which collapsed under him was secured by nails at least at one end when he stepped on to it, and therefore that when proper weight was given to that fact there was "no basis for the finding of contributory negligence". The result of the proceedings in the Supreme Court thus is that two judges - one of them the trial judge - held that the defendant was negligent. Two other judges of the Court said that it was not negligent ; and, as they sat in the Full Court and formed a majority there, their opinion on that prevailed. On the question whether the plaintiff himself was negligent, one judge says that he was not, but three, including the trial judge, say that he was - two of them saying that because of this he is to be regarded as wholly responsible for the accident, the third, the trial judge, saying that he was but partly responsible. (at p211)

17. The primary facts are not now in dispute. No question now remains of the credibility of any witness. The trial judge rejected some of the evidence of Pedri as to the progress and state of the work on his side of the roof at the time of the accident. His Honour described Pedri's evidence on this point as "highly unlikely" and as "highly improbable". As a matter of inference he then concluded, contrary to Pedri's testimony, that there were only two rows of nails in each sheet of iron. This finding was not challenged. And there is no challenge really to his Honour's findings of what was in the plaintiff's mind : what he "assumed" to be the reason for the ridge cap being stuck and not readily removable ; and that it was "natural for him to expect" that an iron sheet left in its position on the roof would provide him with a secure foothold. Matters of that kind are relevant in considering whether the plaintiff's conduct was that of a prudent man. But the extent to which the beliefs, assumptions or expectations of a party have weight when his acts are in question may depend on whether they were reasonably foreseeable as influencing his conduct. It may depend too on how reasonable and firm were those beliefs, assumptions and expectations. In the due assessment of that the trial judge has indefinable advantages over an appellate court. They are among the advantages that, using Lord Sankey's words, are "sometimes broad sometimes subtle". They transcend mere credibility. (at p212)

18. I am sceptical of applying to a finding of negligence the principle that an appeal court is as competent to determine the proper inference from proved facts as is the trial judge. There is, of course, no difficulty in this proposition when the inference of fact is itself of a physical fact or happening, something which could itself have been observed or otherwise perceived, to use Professor Goodhart's word, by the senses of a person actually present at the relevant time. Thus an appeal court can legitimately infer from circumstantial facts that adultery was committed : Mann v. Mann [1957] HCA 68; (1957) 97 CLR 433 . And marks on the road, the position where vehicles came to rest and so on, may enable members of an appeal court to determine, just as well as anyone else could, on which side of the road a motor car accident occurred and other facts of that kind. That is simply to infer an observable fact from observed facts. In the same way a court of appeal may decide whether circumstantial facts justify an inference that a death was caused by suicide : Dominion Trust Co. v. New York Life Insurance Co. (1919) AC 254 . Similarly if the question was whether or not at a particular time there was a fence in a particular position on land : Craine v. Australian Deposit and Mortgage Bank Ltd. [1912] HCA 60; (1912) 15 CLR 389 . The same general principle can, in some cases, be applied to conclusions of mixed fact and law. For example, when the question was whether there was such user of a place by the public as would found an inference of dedication for a public purpose : Montgomerie & Co. Ltd. v. Wallace-James (1904) AC 73 . Or, as in the Benmax Case (1955) AC 370 itself where the issue was whether an alleged invention was really new and an invention which could be the subject of letters patent. Many similar illustrations can be found in the reports. But inferences of fact from proved specific facts seem to me to be logically in a different position from the evaluation or appraisal of the quality of a man's conduct. Was he negligent? (at p213)

19. It is true that we, as lawyers, using phrases derived from the separate functions of judge and jury in trials at nisi prius, say that whether a man has been negligent, the issue that was submitted to the jury in an action of case, is a question of fact. The answer to it is then called a finding of fact. And then, misled I think by our own terminology, we say that a decision that a man was negligent, based upon specific facts that have been found either by direct evidence or by inference therefrom, is itself an inference of fact. That in my view is a confusion of concepts and the beginning of error. A true inference of fact is an inference of an actual matter or event. It is reached by a consideration of actual facts proved and the probabilities to which they give rise. But the evaluation of conduct in terms of reasonableness is a value judgment upon facts rather than an inference of a fact. Negligence carries an ethical overtone, fault or blame : liability results from culpability. It depends upon measuring conduct by a norm, the conduct of a hypothetical reasonable man in the circumstances : and of that there is no certain touchstone. Reasonableness is here a concept as indefinite and imprecise as it is important in the common law. I may be pardoned for quoting Sir Alan Herbert's Uncommon Law for the statement that "the common law of England has been laboriously built about a mythical figure - the figure of the reasonable man". I cannot regard a judgment that a man failed to act as a reasonable man would act as in the same logical realm as an inference that something actually happened or was done. (at p213)

20. I appreciate that the doctrine of the Benmax Case, (1955) AC 370 , has been invoked in negligence cases, and that it was referred to by the Privy Council in deciding that certain acts of a dentist did not answer the statutory description, infamous or disgraceful conduct in a professional respect. But the question there was one of mixed fact and law and it was not really necessary to describe it as involving a finding "in the nature of an inference from the specific facts found" : Felix v. General Dental Council, (1960) AC 704, at p 717 . Moreover, an "appeal" from a disciplinary tribunal is not in the same position as an appeal from a court ; and too in a very recent decision in such a case the Privy Council said only that the facts proved were sufficient to justify the determination by the tribunal of infamous conduct : Sloan v. General Medical Council (1970) 1 WLR 1130, at p 1132 . Their Lordships did not expressly evaluate the matter for themselves. I think that I am justified in putting the Benmax Case (1955) AC 370 aside from the present case. (at p214)

21. In short, in a case of this kind, the primary facts not being in question, I would treat the decision of the trial judge as the equivalent in all respects of the verdict of a jury, unless from his reasons it appeared that he had in some way misdirected himself. It seems to me that - adopting the words of Lord Esher, then Brett J., in Bridges v. North London Railway Company (1874) LR 7 HL 213, at p 235 concerning the verdict of a jury - the question in the end is still whether the finding is "such as reasonable and fair men might not unfairly arrive at". I expect that some will say that this view is a departure from orthodoxy as now declared, and that I am obdurate in heresy. Some may say too that I evade the duty of a member of a court of appeal ; for in London Bank of Australia Ltd. v. Kendall (1920) [1920] HCA 53; 28 CLR 401, at p 407 , Isaacs and Rich JJ., holding that the finding of a judge should not be treated as equivalent to the verdict of a jury, said :

"Where the law says that the court, and not a jury, is to

determine the facts, and also says that an appellate court can

be asked to reconsider them, and therefore should reconsider

them, it is the duty of the appellate tribunal (and it is the

statutory right of the litigant who invokes it to require of it

the performance of that duty) to determine for itself the true

effect of the evidence so far as the circumstances enable it

to deal with the evidence as it appeared in the court of first

instance."

Their Honours went on in that case to quote pronouncements in the House of Lords and Privy Council to the same effect. There are, I realize, many eloquent statements by eminent people from then up to the present time that can be arrayed against the position I seek to defend. Nevertheless in deciding to stand upon it I think it is fortified by logic, principle and policy. I do not stand alone. For example, Lord Normand in Bolton v. Stone [1951] UKHL 2; (1951) AC 850 said that "the decision (of the trial judge) should have been respected as equivalent to a verdict of a jury on a question of fact". (at p214)

22. It cannot be said that there were no facts or no rational interpretation of the facts which would sustain the value judgment of the learned trial judge : and I do not think that it can be said to have been clearly wrong when other judges think that it was right. Therefore I am not prepared to say that it was proper for the Full Court of the Supreme Court to reverse it. So putting aside what might have been my own conclusion had I been in the position of the trial judge, I would allow the appeal and restore his judgment. (at p215)

WALSH J. In the Supreme Court of Western Australia the Full Court by majority (Virtue S.P.J. and Hale J., Burt J. dissenting) dismissed an appeal brought by the present appellant, allowed the cross-appeal of the present respondent and ordered that the judgment which Neville J. had directed to be entered in the action be set aside and that the appellant's action against the respondent be dismissed. The judgment of Nevile J. was based upon findings that an injury which the appellant sustained when employed by the respondent was caused by the negligence of the respondent and that there had been contributory negligence on the part of the appellant. (at p215)

2. In the Full Court Hale J., with whom Virtue S.P.J. agreed, decided that it had not been established that the respondent was in breach of a duty of care owed by it to the appellant. I agree with that conclusion and with the reasons given by Hale J. for it. It is, therefore, not necessary for me to decide whether or not the finding of Nevile J. that the appellant was guilty of contributory negligence was a proper finding on the evidence. (at p215)

3. Hale J. considered that upon the facts disclosed by the evidence the Court ought not to conclude that the respondent should have foreseen the possibility of an accident, such as that which occurred, and taken steps to prevent it. I agree with that view of the case. I do not think that there was anything in the nature of the work which the appellant and his fellow worker Pedri were told to do which would suggest to an employer acting reasonably that instructions ought to be given as to how they should carry out the work or that steps ought to be taken to ensure that each of them was informed of the manner in which the other intended to do his share of the work. It was not shown that anything had happened before this accident occurred to cause the employer to suppose that there was a risk that the two men, if not instructed, might create a situation of danger by each using without the knowledge of the other a different method of doing the work. After the accident had occurred and after evidence had been given at the trial of this action about the methods which were used, it could then be seen that one view of the matter which might reasonably be taken was that it was because of an assumption which the appellant had made as to the method of work being used by Pedri that the appellant believed it to be safe to move to the other side of the ridge and to put his weight upon the sheet of iron from which in fact all the nails had been removed. But before the accident had happened it could not have been reasonably foreseen, in my opinion, that there was a risk of an accident occurring in that way in the course of the carrying out of a fairly simple task by two men working close to each other. At that time there was nothing to suggest to the employer that one of the workers might put himself in peril by acting upon such an erroneous assumption when the relevant fact, that is whether or not the sheet of iron was still secured by nails, was so easily discoverable either by inspection or by speaking to the other worker. (at p216)

4. In his dissenting judgment in the Full Court Burt J. adverted to the duty which an employer has in some circumstances to prescribe a system to be followed by each worker or to give appropriate instructions to the workers. His Honour went on to consider the application of that principle to the facts of the case before the Court and he said :

"In its application to the facts of this case the question is

whether the respondent as a reasonable employer of labour in

what might be described as the demolition industry should

have foreseen that a worker possessing the experience, or

lack of it, possessed by the appellant unless told of the manner

in which iron sheets constituting the roof upon which in a

general sense he was required to work were to be removed

might reasonably assume, contrary to the fact, that a sheet of

galvanized iron forming part of that roof and yet still in its

original position had not had all the nails previously attaching

it to the roof timbers removed so that it was safe to stand on."

His Honour added that this was, in substance, the question asked by the learned trial judge and answered by him adversely to the respondent. Burt J. found himself unable to say that in reaching this conclusion the trial judge was wrong. But with respect, I am of opinion that the question posed by Burt J. ought to have been answered in favour of the respondent. In my opinion, the right conclusion was that the failure of the respondent to instruct the appellant as to the manner in which the iron sheets were to be removed from the roof or to ensure that he knew what procedure his fellow worker would follow did not constitute a breach of its duty to take reasonable care for his safety. The circumstances were not such that it was reasonably foreseeable by the employer that its failure to take that course might be a cause of injury to the employee. (at p216)

5. The learned trial judge did not make any finding upon the question whether the worker Pedri had by acting negligently caused injury to the appellant. In the Full Court, Hale J. referred to this question as being one which "does not seem to have been pursued in the evidence". He came to the conclusion that it had not been shown that Pedri was negligent. At the hearing of the appeal to this Court counsel for the appellant did not contend that Pedri was guilty of negligence for which the respondent should be held liable in the action. In my opinion, it was not proved that Pedri acted negligently but it is not necessary to pursue this matter any further. (at p217)

6. In my opinion, it was proper in the circumstances of this case for the majority in the Full Court to give effect to the conclusion to which their Honours came upon the question whether or not it should be found that the respondent had been in breach of its duty to the appellant. The conclusion was one which did not depend upon the resolving of any conflict of evidence or even upon the evaluation of the conduct of a person who had given evidence and in respect of whom the trial judge was in a position to form general impressions which might be of assistance in deciding the matter. The conduct to be evaluated was that of the respondent, acting through servants or agents who did not give evidence. Whether or not the trial judge, when deciding whether or not the appellant had acted reasonably, had advantages of a kind which would render it difficult for an appellate court to interfere with his finding on that issue, I am of opinion that their Honours in the Full Court were not precluded from giving effect to their own conclusions upon the question whether or not it had been established that the respondent had been negligent. I am of opinion that the decision of the majority of the Full Court on that question was correct. I am also of opinion that in setting aside the judgment which had been entered in favour of the appellant the majority in the Full Court acted within the principles which have been laid down as to the manner in which and the limits within which an appellate court should exercise its functions when hearing an appeal from the decision of a judge sitting without a jury in an action of this kind. (at p217)

7. In my opinion the appeal should be dismissed. (at p217)

GIBBS J. I have had the advantage of reading the reasons for judgment prepared by the other members of the Court. I need not restate the circumstances in which this appeal arises. No doubt exists as to the legal principles which the learned trial judge (Nevile J.) was called upon to apply, and there is now no controversy as to the ultimate facts, but there is nevertheless room for considerable difference of opinion as to the result of the application of the legal principles to the facts of the case. (at p218)

2. The first question for decision is whether the Full Court of the Supreme Court was entitled to disturb the conclusion of the learned trial judge that an employer properly mindful of his duty to use reasonable care for the safety of his workmen would have directed the workmen as to the manner in which the work should be done or at least would have directed them to agree among themselves on the method to be used. In my opinion the conclusion reached by Nevile J. was correct. Although the demolition of the galvanised iron roof of a single-storey building may be a simple enough operation, it involves certain hazards. In considering what precautions the exercise of reasonable care requires in such a case, the possibility of inadvertance or thoughtlessness on the part of a workman should be taken into account - Smith v. Broken Hill Pty. Co. Ltd. [1957] HCA 34; (1957) 97 CLR 337, at p 343 . To allow unfastened sheets of iron to remain in position on a steeply sloping roof in circumstances in which a man walking on the roof is likely to believe that the sheets are still nailed and thus safe to walk upon, seems to me dangerous. The workmen, who were only unskilled labourers, had some previous experience in demolition work, but they had never been given instructions to follow a procedure that would avoid this danger. The learned judges who constituted the majority in the Full Court of the Supreme Court were of opinion that it was not reasonably foreseeable that the appellant would cross the ridge and entrust his weight to the iron in the manner that he did. With all respect, in my opinion a reasonable employer should have foreseen that a workman in the appellant's position might wish to cross the ridge for a variety of reasons, such as to retrieve a dropped tool, or to remove the ridge capping, and that he might do this impulsively or thoughtlessly and without first ensuring that the iron on the other side was secure. It would have been a simple matter to warn the workmen against the danger that might result if one, without the knowledge of the other, failed to remove a sheet of iron immediately it was unfastened, and to give appropriate insutrctions to avoid this danger. In the circumstances it was, in my opinion, the duty of an employer, acting reasonably to safeguard his employees from danger, to give proper instructions as to the performance of the work. I conclude, therefore, that Nevile J. was right in holding that the respondent failed to take reasonable care for the appellant's safety and that the Full Court of the Supreme Court was in error in reversing his decision. (at p218)

3. The question whether the appellant was guilty of a want of reasonable care for his own safety is a more difficult one but after full consideration I have reached the conclusion that Nevile J. was right in answering this question in the affirmative. It is true that the appellant believed that the sheet of iron on to which he stepped was fastened and the fact that the iron had been left in position provided grounds for this belief. I should have thought that the appellant, if he had looked before he crossed the ridge, would have been able to see that, in fact, the nails had been removed, but this was not established by the evidence and I therefore do not rely upon it. However, the appellant did know that Pedri, his fellow workman, had been removing some nails from the iron, and that, after having finished work on the top row of nails, he had moved along the bottom row, a distance of ten or twelve feet away from the appellant. It is true that this did not necessarily mean that all the nails had been removed from the sheet in question, but in my opinion these facts should have warned the appellant that all the nails might have been removed from the sheet. The fact that it was natural for the appellant to assume that any sheet of iron left in position was secured by nails, at least at one end, did not mean that the appellant, taking reasonable care for his own safety, could act blindly on that assumption in the face of other circumstances that should have caused him to doubt its correctness. In all the circumstances, it seems to me that the appellant failed to take reasonable care for his own safety in crossing the ridge and stepping on to the sheet of iron on the other side without first making sure, e.g. by inquiry from Pedri, that all the nails had not been removed from the sheet. I consider that his action in proceeding as he did went beyond mere thoughtlessness or inadvertence and amounted to contributory negligence. (at p219)

4. I would therefore allow the appeal and restore the judgment of Nevile J. (at p219)

Appeal allowed with costs. Order of the Full Court of the Supreme Court of Western Australia set aside and in lieu thereof order that the appeal and the cross appeal to that Court be each dismissed with costs.