The appellant, a boy of fourteen years of age, was employed by the respondent in work at a power press. On the second day on which he so worked his right hand was crushed by the press and seriously injured. He sued the respondent for damages for negligence, alternatively for breach of statutory duty, and alternatively for compensation under the Workmen's Compensation Act 1932-1941 S.A.. The learned judge (Napier C.J.) found for the defendant upon the first and second claims. The right to workers' compensation was admitted by the defendant, but the assessment of compensation was postponed.

Upon appeal to this Court the appellant relied upon breach of statutory duty. The defendant was the occupier of a factory. The Industrial Code 1920-1936 S.A., s. 321, so far as relevant, provides that the occupier of a factory shall securely fence or safeguard all dangerous parts of the machinery therein, and that the occupier shall cause all fencing and safeguards to be constantly maintained in an efficient state while the parts required to be fenced or safeguards (obviously a misprint for "safeguarded") are in motion or use for the purpose of any manufacturing process. It is contended for the appellant that the evidence accepted by the learned judge showed that the machine in question was dangerous and was not securely fenced and safeguarded. Section 321 is plainly a provision made for the protection of workmen, and if the plaintiff was injured owing to a failure to observe the requirements of the section he has a cause of action.

The press consisted of a power-operated ram carrying a die upon the end of an arm and striking upon a metal plate or table with a force of twenty tons. The press was used for the purpose of stamping out, punching, cutting and bending metal. Power was applied through a loose pulley which could be connected by a clutch with a fast pulley mounted upon a shaft to which the arm carrying the die was attached. The connection could be made by a pedal, or by hand. A metal guard was provided between the operator and the die and plate upon which it descended. The machine was so constructed that the arm could not strike until the guard was within an inch or two of the plate, because the pulling down of the guard to that position was required in order to operate the clutch so as to put the shaft in motion. The ram would go on striking as long as the guard was held down. When the guard was released it sprang upward, the clutch was necessarily disengaged, and a brake stopped the revolution of the shaft and prevented any further movement of the ram.

In his evidence the plaintiff said that he was using the pedal, and that he put a piece of metal in "and it worked out all right. I put another piece in and when I was putting it in the die came down. The guard was nearly up to its proper place when I put my hand in." This account means that while the guard was well above the plate the plaintiff put his hand into the machine and it was struck by the ram. The learned Judge was of opinion that the plaintiff's evidence was unreliable and could not possibly be accepted. It is not a matter for surprise that the boy did not have any accurate recollection of what happened. The learned Judge found that the plaintiff was manipulating the guard by hand and not by means of the pedal. He further found that the plaintiff could not have put his hand under the guard, because when the guard was raised sufficiently high to make it possible for him to put his hand under it the machine necessarily stopped. He also found that the machine was in good working order and that theories of the press "wandering" or "repeating" and of seizure between the pulley wheels or between the loose pulley and the shaft were not feasible. These findings are not challenged; it is admitted that they are supported by evidence.

But the accident did in fact happen. The plaintiff somehow got his fingers under the descending ram. The learned Judge found that what had happened must have been that the plaintiff, not using the pedal, had pulled the guard down with his left hand in the ordinary method of operating the machine by hand and had put his right hand behind the guard while the ram was working.

The position, therefore, is that the machine had a guard, but not such a guard as to prevent what must have happened. If there had been a guard which safeguarded not only the front of the machine but also the right-hand side, the accident could not have happened. It was pointed out for the respondent that there was no express evidence of any witness that such a guard could have been provided. It is equally true that there is no evidence that it could not have been provided. We are left to form our own opinion on the matter upon the evidence which describes the character and mode of operation of the machine. No reason was suggested to show that the guard could not have been continued round the right-hand side of the press.

The defendant alleged contributory negligence on the part of the plaintiff - a defence which is available (Piro v. W. Foster & Co. Ltd.[1]). There is no evidence as to the circumstances in which the plaintiff put his hand into the machine. The accident which happened is the kind of accident which the statute is designed to prevent, and the happening of the accident cannot itself be regarded as evidence of contributory negligence. The onus is on the defendant to establish this defence, and there is no finding in his favour on this issue, and, in my opinion, no evidence which would support such a finding.

The statute requires dangerous machinery to be securely fenced. It was not denied that the machinery in this case was dangerous. It could have been more securely fenced than it was, and if it had been so fenced the accident as it is found to have happened could not have taken place. In these circumstances it is not necessary to consider the problem raised in such cases as Davies v. Thomas Owen & Co. Ltd.[2] and Lewis v. Denye[3], where opinions are expressed on the question whether legislation requiring dangerous machines to be securely fenced means that "if a machine cannot be securely fenced while remaining commercially practicable or mechanically useful the statute in effect prohibits its use." There is no evidence in this case that the machine could not be commercially used if it were securely fenced and therefore, in my opinion, this question does not arise.

The appeal should be allowed, the judgment for the defendant set aside, judgment entered for the plaintiff upon the claim for breach of statutory duty, and the case remitted to the Supreme Court for assessment of damages.

After the trial of the action had been completed an application to re-open the case was made on behalf of the plaintiff. An order was made that the hearing be re-opened and that, subject to any order the Court or Judge might see fit to make, the costs of the first hearing and of the application to re-open should be the defendant's costs in any event. The Court or Judge might well, I think, see fit to make another order, but that is a matter for the discretion of the Court to be exercised after considering (inter alia) whether any costs were rendered useless by reason of the re-opening of the case.

This appeal is from a decision of the Supreme Court of South Australia, which dismissed a claim by the appellant for damages founded upon the provisions of the Industrial Code 1920-1936 S.A., s. 321. So far as material that section provides: - "The occupier of a factory shall securely fence or safeguard ... all dangerous parts of the machinery therein" and "cause all fencing and safeguards to be constantly maintained in an efficient state while the parts required to be fenced or safeguards" (? safeguarded) "are in motion or use for the purpose of any manufacturing process." The obligation under that section is absolute in the sense that the defendant is liable, be he ever so blameless, if the dangerous part or parts of his machinery are not securely fenced or safeguarded (Piro v. W. Foster & Co. Ltd.[4]).

In the present case the learned Chief Justice of the Supreme Court said: - "The manifest intention of s. 321 is to protect workmen who are careless; but there must be reason in all things, and I do not think that the employer is required to anticipate and provide for everything that could possibly happen. I think that machinery is not dangerous, or is safeguarded, in the fair meaning of those words, if the only risk of injury is in the event of someone acting as no reasonable person could be expected to act."

The obligation, however, does not, I think, depend upon how a reasonable person might be expected to act, for the section is designed to protect careless and even stupid persons. The section envisages dangerous machinery in operation, but it does not provide an absolute warranty against all injury. In my opinion the obligation imposed by the statute would be discharged if everything were done to fence and safeguard the machinery within the range of human foresight and experience. And it is necessarily a question of fact whether dangerous machinery has been securely fenced or safeguarded. But an employer cannot safeguard a workman who intentionally injures himself. The injury in such a case would not result from any breach of the statutory duty but from the act of the workman himself. Again, an employer cannot guard against many unexpected and violent operations of nature, however securely the machinery is fenced or safeguarded.

And according to the latest decisions the section affords no protection to a workman guilty of contributory negligence (Piro v. W. Foster & Co. Ltd.[5]).

In the present case it is not suggested that the appellant intentionally injured himself. Nor is there any finding of contributory negligence (the proof of which rested upon the respondent) against the appellant, and it would not be reasonable in the circumstances of this case. The sole question is whether the machinery was securely fenced and safeguarded. The case gave the Chief Justice considerable anxiety, but he saw and heard the witnesses and inspected the machinery in operation. To my mind his conclusions of fact should be accepted without conjecturing or speculating in this Court how the appellant was injured.

He was a boy of fourteen or fifteen years of age, and was employed in the respondent's engineering workshop or factory at a power press. He was engaged in shaping small pieces of metal. The press was fitted with dies, the upper die being affixed to the ram and the lower die to the table. The metal blanks were fed to the lower die by hand one at a time, and they were cut or stamped into shape by a single blow of the upper die affixed to the ram. The press was fitted with a guard in the form of a plate approximately twenty inches long and nine inches high, and it was attached to a block and interlocked with the clutch gear, so that the machine would not operate until the guard was down. The machine was operated by a pedal arrangement or by hand. But the appellant was directed to operate it by hand, and the Chief Justice found that he did so operate it. The ram was capable of striking a hundred strokes per minute if the guard were down, but it was necessary to raise the guard every time the metal blanks were fed to the dies. The guard, it was said, prevented the operator from coming into contact with the dangerous parts of the machine during its operation: that it was impossible for a man's right hand (which was the hand of the plaintiff that was injured) to come into contact with the moving parts during the operation of the machine. The Chief Justice said there was a serious conflict of evidence with respect to the circumstances in which the accident to the appellant happened. The result was that he was left to conjecture how the accident happened. The Chief Justice also said that the possibilities were (1) that owing to some wear or defect in the mechanism of the clutch the press was in the habit of wandering; that the clutch failed to disengage and allowed the press to go on striking after the guard had gone up, or (2) that the appellant must have put his hand around and behind the guard whilst it was still down, or (3) that the clutch seized on the fly wheel. If the clutch seized without engaging, the press could start working while the guard was up. In the end the Chief Justice found that he was unable to believe that the guard was up when the accident happened and that the evidence precluded the possibility of the seizure of the clutch, and, difficult as it was to accept the second possibility, still it was less improbable than either of the other solutions, and he therefore adopted it. But he said that the plaintiff in putting his hand behind the guard acted as no reasonable person could be expected to act and, as before indicated, taking this view he found that there had been no breach by the defendant of his statutory duty under s. 321. The obligation of the defendant, however, was, I think, more onerous than that stated by the Chief Justice. That obligation I have already indicated. When the defendant allowed a youth of fourteen to operate this fast-moving machine, which required the raising of the guard every time a small piece of metal was inserted between the dies, he at least should have provided every safeguard which human foresight and experience suggested. A youth who had to move a guard every time he fed the machine would almost inevitably look for an easier way of working, and what more likely than putting his arm around and behind the guard? Admittedly, I gather, there was no mechanical difficulty in constructing the guard so that it extended further round on each side of the machine and prevented an arm being put around and behind the guard or, at least, making such an act so difficult that even a careless and stupid workman would not attempt it. The omission so to construct the guard resulted in the injury to the appellant.

The result, in my opinion, is that the appeal should be allowed.

The statutory duty which fell upon the defendant respondent as occupier of the factory where the plaintiff appellant was injured required him securely to fence or safeguard all dangerous parts of the machinery and to cause the fencing and safeguards to be constantly maintained in an efficient state while such parts were in motion or use for the purpose of any manufacturing process.

The machine at which the plaintiff, a boy then fourteen years and three months old, was set to work was a press, which, of course, included a moving ram or plunger highly dangerous to the fingers and hands of an operator, that is unless the machine was equipped with a guard making accidental contact with the end of the ram or plunger in the process of work practically impossible. The ram of the press was raised and lowered by a revolving crank shaft. There was a fly wheel upon the shaft, and close to it was the power wheel, which engaged and disengaged with the fly wheel and crank shaft by means of a clutch designed with a key, spring and slot. The clutch was worked by a descending rod normally connecting with a foot pedal, which drew it down and so engaged the clutch. Upon release of the pedal a strong spring brought back the rod and declutched the shafting. A plate was attached to the descending rod for the purpose of bearing a guard. To the plate an arm was screwed about twenty inches long. The arm carried at right angles to it a shield consisting of a steel plate twenty inches wide and nine inches high. It was designed so that it should descend to the table of the press as and when the rod was pulled down so as to engage the clutch, with the result that there would be a solid shield in front of the ram as it descended to the table. The height of the steel table of the press was about that of an ordinary table. When the shield was raised, and the clutch consequently disengaged, the lower edge of the shield was about fifteen inches from the table. To engage the clutch the rod with the shield attached was drawn down far enough to bring the lower edge of the shield close to the table. How close was a matter which does not appear to me to be free from doubt, but the shield did not hang perfectly horizontal and the right-hand side, which was the side attached to the arm by which the shield was borne, was about an inch from the table when the left-hand side touched it.

Napier C.J., who tried the action, said that the clutch would not engage until the guard was brought down to the table and would disengage whenever the guard was allowed to rise about one and a half inches from the table. His Honour had the advantage of seeing the machine, but two years had elapsed and I do not feel certain that the conditions existing at the time of the accident were exactly reproduced.

Though the usual way of working the machine was by means of a pedal or treadle, it was possible to do so by means of the guard. The treadle could be detached and then, by pressing down the shield, the rod could be drawn down to engage the clutch. According to the foreman; he considered that for the boy the machine was safer without the pedal because, in that state, it would keep both his hands employed. His left hand would be employed in pressing down the shield to engage the clutch and cause the ram or plunger to make its stroke, and his right hand would be employed in placing the metal on the lower die upon the table when the shield was up, the clutch disengaged and the stroke finished with the ram held by the brake in an up position pending its next stroke.

At the time of the accident the machine was stamping out half hinges for ammunition boxes. They were stamped from steel blanks placed by hand on the lower die, which was four and a half inches long and one and a half inches wide and about two inches high from the surface of the steel table. Upon the lower die the upper die attached to the plunger descended with great force. If the spring which effected the declutching after each stroke were counteracted by holding down the pedal or the shield, as the case might be, the shaft would make a hundred revolutions a minute, and of course, the plunger or ram would make as many strokes. For some work it was used in this way, but, where it was necessary to feed it with individual blanks, the operation required that the clutch should disengage and that the ram should be held at the top of its stroke. The disengagement was timed so that the ram would reach its up position and there be held by a brake band on the shaft placed at the opposite end from the fly wheel.

Napier C.J. was under the impression that, in addition, there was a check or stop (which was pushed in as the clutch engaged) and that it prevented the die falling when the clutch was not engaged, even if the brake was not operating. But that is not the effect of the evidence and we were told that it was a mistake. (It may have been a misapprehension of the meaning of the evidence of Ingerson.)

The speed at which the machine could be worked when it was declutched and fed after each stroke depended upon the skill and facility with which the operator used it. An experienced girl could stamp a blank every three seconds. It is obvious that to do this the hand feeding the die must be passed quickly under the shield as it rises and withdrawn as the other hand presses down the shield or the foot presses the pedal, as the case may be. In other words, when the pedal is out of use, it means continual and co-ordinated movement of the hands in regular sequence and at a fast rate. The boy's performance reached only one every six seconds, or half the rate of out-turn of an experienced operator. He had been instructed in the use of the machine for the first time on the day preceding the accident. The accident occurred immediately on the resumption of work after lunch. He had been at work at the machine throughout the morning. The ram took off parts of three fingers of his right hand. The middle finger has practically gone and the fingers on either side have lost the top joints, an injury consistent with his fingers having come under the die at a corner.

The boy's own account of the accident was clear, and intelligible enough, but it involved the use of the pedal. The foreman swore definitely that the pedal had been disconnected and that, when he inspected the machine after the accident, he found the pedal disconnected. For that and kindred reasons, the boy's explanation was rejected. The difficulty then was to say how the accident did happen. The defendant attempted to demonstrate by evidence, argument and exhibition of the machine, that there was no way in which the ram could have descended unless the shield was down and that there was no possibility of the boy's arm going under the shield when it was down. The foreman said at one place in his evidence: "We could not come to any conclusion as to how the boy could have got his hand in. We could not make it out." Experience suggests the comment that, if this were so, the defendant's case proved too much. But a little earlier in his evidence he had been asked his opinion whether the boy might have put his hand in at the side of the shield, a question to which he gave the following answer: "It would be a little awkward, it would depend on the position in which he was standing. In the normal operation, it would be very awkward. If he went further to the right, it would be easy. If he went to the right of the machine, it would be easy to do so. I don't consider it would be easy standing right at the front of the machine. It would be difficult."

Now the boy had a tin of blanks at the right-hand side of the table. At its nearest point the plate or shield came to rest seven inches from the die, which means that from the edge of the shield the die was little more than a foot away. On the assumption that the pedal was detached, he was bringing down the shield with his left hand. There is nothing to my mind unnatural or improbable in supposing that he tended to move to the right corner of the table. Nor is it at all unlikely that, in attempting to gain speed at a new and not yet familiar task, he did not preserve proper and precisely regular sequence in the movements of his two hands. If he failed to do so and brought down the shield with his left hand before he had withdrawn his right hand, it might well be, if he was standing to the right, that his right hand would not be under the shield as it descended, but to the right of it, where there would be nothing above his forearm except the bar by which the shield was attached to the rod connected with the clutch. To say that this happened would be speculative, but to say that it might have happened shows that the shield did not form a secure fence or safeguard. A further consequence follows, however, from the conclusion that the imperative requirement of the statute that a dangerous part of the machinery should be securely fenced or guarded was not fulfilled. In Vyner v. Waldenberg Bros. Ltd.[6], an accident had occurred to an employee working at a circular saw, but how it really happened no one knew. It appeared, however, that the riving knife at the back of the saw did not conform with the statutory regulations for safety. Scott L.J., in delivering the judgment of himself, Mackinnon and Morton L.JJ., laid down the rule as follows: - "If there is a definite breach of a safety provision imposed on the occupier of a factory, and a workman is injured in a way which could result from the breach, the onus of proof shifts on to the employer to show that the breach was not the cause. We think that that principle lies at the very basis of statutory rules of absolute duty."[7]. It is not necessary to inquire whether their Lordships meant more than that the breach of duty coupled with an accident of the kind that might thereby be caused is enough to justify an inference, in the absence of any sufficient reason to the contrary, that in fact the accident did occur owing to the act or omission amounting to the breach of statutory duty. In the circumstances of this case that proposition is enough. For, in my opinion, the facts warrant no other inference inconsistent with liability on the part of the defendant. Napier C.J., having rejected the boy's account of the accident, said that he was left to conjecture how the accident happened. Having examined various theories, and being of opinion that the shield was a sufficient guard, he felt forced to the conclusion that, difficult as it was to accept the explanation advanced by the defence, it was less improbable than the other solutions he had considered. That explanation, as I understand it, was that the boy must have put his hand around and behind the guard whilst it was still down and the plunger or ram was striking at the rate of one hundred strokes a minute. This explanation appears to me to be a priori extremely improbable. It is not clear whether the hypothesis is that the boy did this owing to confusion or inadvertence or as a deliberate course of conduct. Positive proof of contributory negligence lies upon the defendant, and confusion and inadvertence would not justify a finding upon that issue against the boy (Piro v. W. Foster & Co. Ltd.[8]). It is not for the plaintiffs to prove that the accident took place otherwise than by some intentional act: Cf. per Greene M.R., Stimpson v. Standard Telephones & Cables Ltd.[9]. If, however, the meaning is that, owing to confusion or the like, and inadvertently, the plaintiff so acted, the question would again arise whether the shield was a sufficient safeguard or should have been more extensive. But, in any case, these do not appear to me to exhaust the possible explanations of the accident. There is one other that I would not be prepared to exclude, but it is one that implies liability on the part of the defendant. There was strong evidence that on other occasions the ram or plunger of the press in question had come down, notwithstanding that the treadle and guard had been up and had not been depressed (Transcript, Dunstan, pp. 107-108, 110; Coombs, 112-113; Jean Kraft, 116, 117, 118, 120; Walter Kraft, 122, 123, 124; Fry, 125-128, 130, 131, 132) and, although it was answered, it does not appear to me to have been answered very satisfactorily. Nor do I think that the attempt to exclude the mechanical hypotheses put forward to explain the happening of such a thing is completely convincing. The coming down of the ram in this fashion was called "repeating," and more than one hypothetical explanation was suggested. It is enough to mention one of them. It is the possible failure of the brake owing to its having been unscrewed to aid the movement by hand of the ram for inserting dies or else through lubrication. His Honour thought that for the machine to repeat, even so, would be impossible, but his conclusion could not but be affected by the mis-apprehension concerning the existence of a check or stop preventing the die falling when the clutch is not engaged. To suppose that the machine did repeat appears to me to be on the evidence a much more reasonable hypothesis than that the boy held down the guard and, notwithstanding the obstruction it afforded and the rapidly repeated strokes of the ram, either worked or experimented at the dies.

For the foregoing reasons I am of opinion that the appeal should be allowed and that judgment for the defendant should be discharged and judgment entered for the plaintiff for damages to be assessed. The plaintiff should have the costs of this appeal and of the action, including, in my opinion, the costs of the hearing on 3rd, 4th and 5th May 1944, but not of the application made by summons, dated 9th May 1944, the defendant's costs of and incidental to which should be paid by the plaintiff's next friend. Costs should be set off.

The only claim among those sued upon at the trial which the appellant has pursued in this appeal is the claim for damages for a breach alleged on the part of the respondent of his duty under s. 321 of the Industrial Code 1920-1936 S.A.. The duty alleged to have been broken is the duty created by that part of the section which provides that the occupier of a factory shall securely fence or safeguard all dangerous parts of the machinery therein and cause all fencing and safeguards to be constantly maintained in an efficient state while the parts required to be fenced or safeguarded are in motion or use for the purpose of any manufacturing process.

A dangerous part of the power press at which the appellant was working at the time of the accident was that part consisting of the ram and the lower die on the table where the upper die fixed to the end of the ram struck when the press was in action. There was a plate fencing this part of the machinery, and it was not possible for a person to get his hand over or under this guard when the guard was down, although it was not flush with the table. The ram would not strike while the guard was up.

But the guard did not extend on the right or left side of the ram or the die. The respondent's foreman said in evidence that while in the normal operation of the machine it would be very awkward for the operator to get his right hand round the right-hand side of the guard while it was down, yet if he went to the right it would be easy to do so. There was nothing to prevent the operator from going to the right side. There was evidence that when the guard was down and the machine was in action the ram would continue striking the die on the table with great speed and force.

Section 321 says that the occupier of a factory "shall securely fence or safeguard all dangerous parts of the machinery." I think that the respondent did not perform this duty by putting the guard in front of the die and ram, but leaving the sides entirely unfenced. The Act does not add any qualification to the words quoted from s. 321. The duty is owed to the careless as well as the negligent workman. It is not limited to machinery which is dangerous in its ordinary careful working. The duty is the prevention of injury (Cofield v. Waterloo Case Co. Ltd.[10], per Isaacs J.).

Napier C.J. said that he could not rely upon the appellant's recollection of how he came to be injured; no other witness saw the accident. His Honour held that the guard was not up in front at the time of the accident, and that the appellant could not have got his right hand over or under it. But his Honour was satisfied that the appellant had got his hand under the ram and it was injured in this way. There was cogent evidence that it would have been possible for the appellant to have got his right hand round the right side of the guard while it was down and the ram was striking the die on the table. The accident could not have happened if the ram and die had been securely fenced and safeguarded by extending the guard on either side.

The respondent committed a breach of statutory duty by leaving the ram and die insecurely fenced and safeguarded against the possibility of a young and inexperienced boy like the appellant putting his hand round the side of the existing guard. In my opinion the facts established prima facie that the cause, or substantial cause, of the appellant's injury was this breach of statutory duty, and the appellant is entitled to recover damages unless the facts establish that the appellant was guilty of contributory negligence (Stimpson v. Standard Telephones & Cables Ltd.[11]). The onus was upon the defendant to establish this defence. It was not established by proof of the fact that the appellant had got his hand under the ram while it was working.

Having regard to the standard of care laid down in Caswell v. Powell Duffryn Associated Collieries Ltd.[12], by which this question has to be determined, and also the inexperience of this boy, I think that a finding of contributory negligence could not possibly be justified.

In my opinion the appeal should be allowed and the judgment of the Supreme Court set aside. I agree with the order for costs proposed by the Chief Justice of the High Court.

Appeal allowed with costs. Judgment for defendant set aside. Judgment for plaintiff upon claim (a) in the statement of claim for damages to be assessed. Cause remitted to Supreme Court for assessment of damages and to make such order as to costs of the action in Supreme Court as may be just.

Solicitors for the respondent, Baker, McEwin, Millhouse & Wright.

[1] [1943] HCA 32; (1943) 68 C.L.R. 313.

[4] [1943] HCA 32; (1943) 68 C.L.R. 313, at p. 326.

[5] [1943] HCA 32; (1943) 68 C.L.R. 313.

[8] [1943] HCA 32; (1943) 68 C.L.R. 313, at pp. 322, 323, 327-329, 337, 338.

[10] [1924] HCA 18; (1924) 34 C.L.R. 363, at pp. 372-374.

[12] (1940) A.C. 152, at pp. 178, 179.