(a) defective manufacture;
(b) damage after the disc left the control of the appellant and before it was received by the plaintiff, and
(c) misuse by the plaintiff.
11 The plaintiff's evidence, if accepted, excluded misuse on his part. The appellant's case was that not only was there no known precedent for such an accident, but that the implausibility of a defect in manufacture required not merely that its statutory defences should be upheld, but that the plaintiff's account of the accident should be rejected.
12 Each side called its own expert. Unfortunately, the plaintiff's expert, Mr Murray Peterson, assumed the correctness of the plaintiff's account of events; the appellant's expert, Dr Robert Casey, considered the plaintiff's account implausible. Neither expert could identify a clear mechanism for the disintegration of the disc.
13 In addition, each party gave evidence of events within their respective areas of knowledge. Thus, the plaintiff gave evidence as to the circumstances in which the injury occurred, including the purchase and installation of the new cutting disc and the disintegration of the disc at high speed, before commencing to use the blade to cut a black plastic "creeper", which he used in his business as a motor mechanic. (The "creeper" was a low, flat plastic sheet on castors, on which he could lie and manipulate his position under a vehicle.) The plaintiff also called a friend, Mr Garry Keenan, who had been present when the accident occurred.
14 For the appellant, Mr Portier gave evidence of the systems adopted by the appellant to maintain quality control of the discs which it manufactured.
Legal basis of action
15 The plaintiff's case was based on two provisions of the Trade Practices Act. The first, s 74D, provides that the manufacturer of goods which are "not of merchantable quality" is liable to a consumer who suffers loss or damage by reason of the goods not being of merchantable quality: s 74D(1).
16 A specific relevant defence in respect of that liability is found in sub-s 74D(2) which provides:
"(2) Subsection (1) does not apply:
(a) if the goods are not of merchantable quality by reason of -
(i) an act or default of any person (not being the corporation or a servant or agent of the corporation); or
(ii) a cause independent of human control;
occurring after the goods have left the control of the corporation …."
17 The concept of "merchantable quality" is defined in s 74D(3) in the following terms:
"(3) Goods of any kind are of merchantable quality within the meaning of this section if they are as fit for the purpose or purposes for which goods of that kind are commonly bought as it is reasonable to expect having regard to:
(a) any description applied to the goods by the corporation;
(b) the price received by the corporation for the goods (if relevant); and
(c) all the other relevant circumstances."
18 In the present case, nothing turned on the concept of "merchantable quality": it was conceded that, if the disc disintegrated in circumstances which involved no misuse by the consumer, the disc was not of merchantable quality.
19 The second cause of action was based on s 75AD, which provides that a corporation supplying goods having a "defect" which causes an individual to suffer injury, is liable to compensate the individual for the loss suffered as a result of the injuries.
20 A defence is provided with respect to such liability by s 75AK:
" 75AK Defences
(1) In a liability action, it is a defence if it is established that:
(a) the defect in the action goods that is alleged to have caused the loss did not exist at the supply time …."
21 An action under s 75AD is identified as a "liability action": s 75AA. The goods said to be supplied and to have a defect are defined as "action goods": s 75AA. The term "supply time" means the time when the goods were supplied "by their actual manufacturer": s 75AK(2). Although, in relation to s 74D, goods are defined as goods "of a kind ordinarily acquired for personal, domestic or household use or consumption", there was no dispute that the disc in the present case fell within the reference to "goods": see s 74A(2)(a). A reference to the "quality of goods", as used in s 74D, includes a reference to "the state or condition of the goods": s 74A(2)(b).
22 For the purposes of s 75AD, "goods have a defect if their safety is not such as persons generally are entitled to expect": s 75AC(1). However, as with merchantable quality, nothing turned on the precise limits of the concept of a defect. If the disc had disintegrated in the circumstances as described by the plaintiff, the appellant accepted that the goods had a defect.
Correct approach to the evidence
23 The primary challenge raised by the appellant with respect to liability concerned the manner in which the trial judge dealt with the evidence concerning the disintegration of the disc. The appellant argued that the most likely cause of disintegration was misuse. The claim that the disc disintegrated whilst spinning in the grinder, the appellant suggested, was so implausible as to require its rejection. The grinder had a stated operating speed of 6,600 rpm while the maximum operating speed identified on the discs was 7,500 rpm. Neither expert suggested that the grinder was not operating at the stated operating speed.
24 The plaintiff gave the following evidence as to what happened when he switched the angle grinder on (Tcpt, 20/05/08, p 15):
"Q. And what happened when you pressed the button? What did the machine start to do, before the accident I mean?
A. Started to turn the disc.
Q. How long between the disc starting to spin and the time you noticed something had gone wrong? What time elapsed?
A. About one or two seconds.
Q. Now, the grinder, of course, makes a loud noise, does it not?
A. Yes.
Q. What was the first thing that you noticed that was wrong after you turned on the grinder? Whether you saw it or felt it or heard it, what was the first thing?
A. I heard like a bang noise.
Q. And the next thing?
A. All pieces whizzing past me.
Q. When you say you felt the[m] whizzing past you, could you hear them over the noise of the grinder?
A. Couldn't hear over the noise of the grinder.
…
Q. And what was the next thing you noticed?
A. Something strike my hand."
25 The grinder, with the remnants of the cutting disc still in it, was tendered as a exhibit, the plaintiff having given evidence that he had not taken the disc out whilst it was in his possession: Tcpt, p 17(25).
26 The cross-examination of the plaintiff commenced with a challenge to his business records and tax returns. The trial judge concluded that his "evidence in relation to the tax returns, and the returns themselves, were unsatisfactory and certainly unconvincing": at [32]. He added that "his evidence in relation to his income was almost unbelievable": at [37]. This challenge to his credit went not merely to his evidence as to loss and damage, but also cast doubt on the accuracy of his description of the accident.
27 The cross-examination also challenged his recollection of the accident. It was put to him that, in giving his evidence in chief, he had failed to recount that he had put the grinder on top of a wheelie bin, after inserting the disc. The fact that the grinder had been placed on the wheelie bin was ascertainable from the report prepared by Mr Peterson, who had interviewed the plaintiff on 2 February 2004. His Honour did not see any significance in the omission of that part of the history: at [43]. There was no reason to doubt his Honour's assessment of that aspect of the evidence.
28 The critical part of the cross-examination was as follows (Tcpt, 21/05/08, p 19(5)):
"Q. You'd started using the grinder to cut before anything happened with the disc, hadn't you?
A. No.
Q. Wasn't it the case that it was during the course of cutting with the disc that something happened and pieces started to fly?
A. Definitely not."
29 Mr Keenan was with the plaintiff at the time of the accident. His evidence was consistent with that of the plaintiff (Tcpt, 21/05/08, pp 29-31):
"Q. Well can I take you from the time he walked out of the garage or the shed with you following him, what did he do with the grinder next?
A. Placed it on the wheelie bin.
Q. What did he do while it was on the wheelie bin?
A. He plugged it in.
Q. Do you know where the cord from the little short cord on the grinder you can see to the power point led to, do you know where the power point was?
A. The power point was at the back of the house.
Q. Well after he'd done that, connected up the cord to the grinder I mean, what did he do with it?
A. He switched it on, picked it up and switched it on.
…
Q. Well you heard it start up, correct?
A. Oh well seemed so quick.
Q. Well we'll get to what happened so quickly in a minute. But did you hear it start?
A. Yeah.
Q. Then what did you hear?
A. A bang and.
Q. Now are you able to tell me what sort of time there was and if you can't, in seconds or … whatever measure you find comfortable, tell me are you able to estimate how long there was between the start up and the bang?
A. A couple of seconds, not - no time.
Q. Now when the bang happened what did you do?
A. I was dodging pieces of that flying past me.
…
Q. All right. When you heard that bang had Wayne actually got to the creeper and started cutting it?
A. No.
Q. After you heard the bang and the pieces whizzed past you, did you have a look at Wayne?
A. Yes.
Q. What did you see about him?
A. I seen his hand.
Q. Was there anything wrong with it?
A. Yes.
Q. What was - what did it look like?
A. Mangled.
Q. When you saw that, what did you do?
A. Threw up."
30 Although Mr Keenan had described himself as a friend of the plaintiff from the age of five or even younger (Tcpt, p 26), in cross-examination it became clear that they were no longer close. He said that he had not stayed with the McPhersons since the accident at all and had visited Sydney (from his home on the north coast) about twice and had stayed elsewhere on those trips. The only times he had seen Mr McPherson in the four years since the accident were at court hearings and he said he had spoken to him on the telephone "maybe half a dozen" times: Tcpt, p 35. The cross-examination as to the circumstances of the accident was quite limited (Tcpt, p 36):
"Q. Mr Keenan, were you able to see or do you remember or were you able to see whether or not at the time that Mr McPherson turned the grinder on he turned it on and started to cut the creeper?
A. He didn't cut nothing.
Q. Were you able to see that?
A. Yes.
Q. And you remember that, do you?
A. Yes.
Q. Would you accept that from where you were standing if Mr McPherson was in a crouched position with his back facing you and the grinder and the creeper in front of him that he might have started to cut without you being able to see it?
A. No.
Q. Isn't it the case from the way you've described your position and Mr McPherson's position relative to the grinder and the creeper that you could not have seen whether or not he had the grinder in contact with the creeper at the time he started it up?
A. No."
31 Mr Keenan suffered from epilepsy and deafness. His Honour said of him:
"53. I say at the outset that Mr Keenan impressed me a great deal. Mr Keenan was an unsophisticated man who suffered health problems including difficulties with his hearing and long-standing epilepsy (in fact from age six months). Despite his lack of sophistication I was impressed with his evidence because it struck me as representing the straightforward observations of a simple man. He had no hesitation in telling me that following the accident he was of no assistance to the plaintiff because, having seen what he described as the plaintiff's mangled hand, he was throwing up and was incapable of ministering to the plaintiff's wounds. He said in clear terms and under careful cross-examination, that the bang (ie the disintegration) occurred within a second or two of the machine being started and before any cutting took place.
54. Effectively the defendant has asked me to not believe Mr Keenan and to reject what he says he saw of the accident. This means Mr Keenan would have had to make up, probably in conjunction with the plaintiff, important parts of his evidence. I do not think he would have been capable of the necessary guile to effect this end."
32 Although the trial judge stated (at [8]) that he had not had the benefit of a transcript, his summary of the evidence, and particularly that of Mr Keenan at [55]-[60], was both clear and accurate. His acceptance of Mr Keenan's evidence, in its corroboration of the plaintiff (at [66]), provided a substantial obstacle to the appellant's case on the appeal. Its case at the trial had depended upon the objective evidence, as provided by Dr Casey and Mr Portier, being inconsistent with the descriptions given by the plaintiff and Mr Keenan. The same contention was repeated before this Court.
33 His Honour dealt with this case directly, commencing with a concise statement of the evidence relied upon by the appellant, at [68]:
"The objective evidence referred to included the following:
(a) The presence of the ridge within the guard which was essentially undamaged.
(b) The fact that some 21 million discs a year had been produced over an eight year period and there was no evidence of such a failure before.
(c) The failures that were recalled by Mr Portier involved only two failures causing injury and one causing property damage. There were about 30 others concerning the life expectancy of a disc.
(d) The system in place was obviously working and effective in preventing the distribution of defective discs. …
(e) The ranking by Mr Peterson of the possible source of the defect militated against it having been damaged in manufacture or packing.
(f) The rebounding quality of the disc fragment was such that it was unlikely to have caused any injury."
34 His Honour noted that the appellant's expert, Dr Casey, had stated that the plaintiff's version of events was "implausible" and that the disintegration of the disc must have occurred in some other way: at [73].
"In particular, he put forward the proposition that the blade could have disintegrated for other reasons than an inherent flaw. He said that if the blade became jammed in a cut or the blade kicked back or was misaligned during cutting, then these could be explanations for the fracture…."
35 His Honour stated that his first impression on reading the report was that, in Dr Casey's view, the plaintiff's version was "impossible, having regard to some of his objective observations". In that respect, Dr Casey had placed primary reliance on "a mound of debris" that had built up on the inner side of the guard of the grinder which appeared to have been unaffected from the fracturing of the disc: see (a) at [33] above. In his first report (par 19) Dr Casey stated:
"Within the guard, a large amount of debris has built up in a plane that more or less corresponded to the plane of the blade. … This mounding was intact around the full periphery of the internal wall of the guard. I probed this mounded debris with a sharp implement and found it to be very brittle and fractures away from the wall of the guard, very easily. The fact that this mound was intact around the full periphery indicates that it had not been struck by a large piece of blade-debris as the blade broke, which would be expected to break the debris from the wall. It then follows that the blade only fractured when it was outside the guard and moved away from the guard. Given that the blade had fractured around its full circumference, it is considered highly unlikely that this could spontaneously occur in just one direction so as not to strike the inside of the guard. It is considered that the improbability is to the point as to make this scenario implausible. It is considered far more plausible that the reason for the fracturing of the blade outside the guard relates to the fact that the blade was in contact with something at the front of the machine. The motion of the blade then projected the debris away from the guard."
36 Mr Peterson disputed Dr Casey's assessment in his principal report at par 3.7, stating:
"… I would comment that the guard extends about 50 percent of the disc's circumference and so there is about a 50 percent chance that a segment would break free without striking the guard on start up of a faulty cutting disc. It must be understood that the rotational speed of the disc is 6,000 rpm or 100 rotations per second. Should a flaw have existed as indicated by the photographic evidence, then the failure does not fit the description of Dr Casey as being highly unlikely. With there being a flaw in the chord of the cutting disc at 90 degrees to the radial direction we can expect that the disc would fail immediately on start up at 6,000 rpm without making any cut and a large segment would break free which evidently was ejected which missed the guard."
37 In his supplementary report, Dr Casey disputed the arithmetic, noting that the likelihood of a piece fracturing away from the disc and not striking the guard was approximately 20%, rather than 50%: at par 13.
38 His Honour accepted that if the objective evidence established that the plaintiff's version simply could not have happened, he would have had no choice but to reject the plaintiff's claim: at [77]. However, he concluded that Dr Casey's evidence did not go so far.
39 In his oral evidence, Dr Casey accepted that if there were a crack in the order of 30mm long in the disc, a single fragment might come away and leave the remainder of the disc intact: Tcpt, 22/05/08, p 37. However, he considered that "far more likely than not to have resulted from contact with an external object": p 38(5).
40 Dr Casey also agreed in cross-examination that he had inspected the guard to see if there were any fragments of plastic which might have resulted from use of the angle grinder to cut the plastic creeper, being the purpose for which the plaintiff had obtained the disc. He agreed that he was unable to find any such material: Tcpt, pp 47-48.
41 A significant part of Dr Casey's investigation, and his cross-examination, were directed to the cause of the plaintiff's injury and, in particular, whether it had resulted from a fragment of the disc rebounding off a hard surface: see (f) at [33] above. Dr Casey attempted some crude experiments to see if he could replicate such an event, but was unsuccessful. The evidence concerning the rebounding qualities of fragments of disc was thus largely inconclusive. That evidence provided no firm basis upon which to discount the explanation of the accident given by Mr Keenan and the plaintiff. Ultimately this evidence appears not to have been given much weight and was not relied upon in the appeal.
42 The trial judge did not make a finding as to the likelihood of no fragment hitting the guard upon spontaneous disintegration. That approach was legitimate for three reasons. First, Dr Casey's calculation of a 20% chance was based upon an assumption as to the size of the fragments: as their size was in fact unknown, this was a matter of speculation and must have been subject to a significant degree of error. Secondly, the disparity between a 20% chance and a 50% chance was not of great significance in the circumstances. Even if there were only a 20% chance of the accident occurring as the plaintiff described, the objective evidence did not categorically disprove the plaintiff's claim. Thirdly, the calculation was based on an accepted fact, namely that no piece of the disc hit the inside of the guard. Because it was clear that the disc did disintegrate, the relevant comparison needed to be between the likelihood of that occurrence in a case of spontaneous disintegration, as opposed to its likelihood in a case where external pressure was applied through the use of the disc. No comparative statistical analysis was provided on this basis.
43 For these reasons, his Honour was entitled to treat the undamaged ridge of debris within the guard as inconclusive.
44 In relation to (e) at [33] above, his Honour took into account the evidence of Mr Peterson who, after reading Dr Casey's report, adhered to the view that the plaintiff's description of the event was plausible. (For present purposes, Mr Peterson's view that any defect in the goods was more likely to have been caused during transport than by manufacture, discussed below, was beside the point.) Mr Peterson was cross-examined at some length as to the greater likelihood that the disc disintegrated during a cutting process when some inappropriate lateral force was applied. He agreed that such an event might account for there being no strike marks inside the guard, but maintained the view that it was not consistent with the absence of marks around the circumference of the disc: Tcpt, 22/05/08, p 19(5).
45 Mr Peterson was also cross-examined on the basis that, if each of the pieces were roughly 30mm square, the outer circumference of the disc which disintegrated would have created some 21 pieces. If they were all thrown off at the same point in time, he agreed that the possibility of no piece hitting the guard was zero: Tcpt, p 9. However, he did not accept the underlying assumption that each piece left the disc as part of an independent event. In fact, both experts adhered to the theory that the outer parts of the disc broke off in a process described by his Honour as "akin to the unravelling of an orange peel": at [79], referring to the evidence of Mr Peterson at Tcpt, p 3(25).
46 On this point, the statistical evidence was largely irrelevant. Accepting that the inside of the guard on the grinder demonstrated that no significant fragment had been thrown in that direction, the only question was whether that could have resulted from spontaneous disintegration, or only from the application of an inappropriate force on the spinning disc. The contrary views of the experts provided little assistance in resolving this question. Each involved a level of speculation. His Honour was entitled to treat Mr Peterson's evidence as supporting the opinion that the plaintiff's claim was not implausible.
47 The other three objective factors, (b), (c) and (d), which formed the primary challenge to his Honour's finding on appeal, related to the quality controls inherent in the manufacturing process, together with the fact that producing some 21 million discs a year had given rise to no known failure of the kind asserted by the plaintiff.
Error in assessment of evidence
48 Although the trial judge had identified these matters as relevant to whether the plaintiff's account of the accident should be accepted, he in fact dealt with them only in reference to the statutory defences, after he had found that the disc was defective: at [87].
49 That approach was erroneous, as the appellant was entitled to have this evidence taken into account, as his Honour had apparently recognised at [68], in considering whether such a finding should be made in the plaintiff's favour. However, because the exercise to be undertaken required a determination as to the extent to which the objective evidence should be used to undermine the credible evidence of Mr Keenan, supporting that of the plaintiff, as to how the accident occurred, it would usually be necessary for this Court to remit the matter for a retrial if the error were material. That course would normally invoke the constraints imposed by the Uniform Civil Procedure Rules 2005 (NSW), which provide:
" 51.53 Circumstances in which Court may order new trial
(1) The Court must not order a new trial on any of the following grounds:
(a) misdirection, non-direction or other error of law,
(b) improper admission or rejection of evidence,
(c) that the verdict of the jury below was not taken on a question that the trial judge was not asked to leave to the jury,
(d) on any other ground,
unless it appears to the Court that some substantial wrong or miscarriage has been thereby occasioned."
50 There was no suggestion that this rule did not apply with respect to an appeal by way of rehearing under s 75A of the Supreme Court Act 1970 (NSW), in relation to an appeal from the District Court from a judgment of a judge sitting without a jury. Nor was it suggested that there was any inconsistency between the rule and s 75A. Further, despite the apparent purpose of sub-r (1) in specifying the relevant ground on which an appeal might be upheld, it is clear that no constraint is imposed by that identification of grounds, because of the breadth of par (d). The Court must therefore be satisfied that the particular error has occasioned "some substantial wrong or miscarriage"; if not so satisfied, and if the Court is unable to assess the evidence itself, it must presumably dismiss the appeal.
51 There is another reason why the Court must assess the evidence: the appellant did not seek a retrial, but insisted that it was open to this Court to conclude that the appellant's case should be accepted, despite the trial judge's views about the evidence of the plaintiff and Mr Keenan. Nevertheless, it is appropriate for the Court first to consider whether the manner in which the trial judge addressed the appellant's evidence gave rise to a substantial wrong or miscarriage.
52 The principal witness relied upon by the appellant in relation to this material was Mr Portier. According to the appellant, his evidence established "that 21 million discs were produced in the relevant year (and in each of the previous four years) and not one had failed spontaneously on start up": written submissions, par 41. That was described as an "objective fact": written submissions, par 50(a).
53 This submission significantly misstated Mr Portier's evidence. Mr Portier said that he had been the Product Manager for "thin wheel products" with the appellant since early 2002: Tcpt, 22/05/08, p 75. He was asked whether, in that position, he was "aware of any other occasion on which it has been suggested that a thin wheel disc manufactured by Saint-Gobain has failed on start up": he answered "No": Tcpt, 23/05/08, p 23(45). That evidence was further qualified in cross-examination at p 46:
"Q. Are you saying … that you've never had a complaint from a customer about one of your grinding discs?
A. No.
Q. You have complaints from time to time don't you?
A. Yes.
Q. And from time to time you have complaints concerning the discs breaking don't you?
A. Yes.
Q. From time to time do those complaints include a complaint that property was damaged or personal injury was suffered?
A. Yes.
Q. And frequently those complaints as to personal injury being suffered or property being damaged have to do with discs breaking up, is that right?
A. Not always.
Q. No but frequently?
A. No.
Q. What about often?
A. No.
Q. Sometimes?
A. Yes."
54 Mr Portier appeared to agree that he had received complaints of people being hurt which had involved a defect in a product and that they averaged two a year: Tcpt, p 51. Unfortunately, there was a degree of imprecision in the questioning and in the answers. Mr Portier had no records with him of customer complaints, had not checked the records held by the appellant and was relying on memory: Tcpt, p 52. Mr Portier was treated as a witness of fact, rather than an expert and no prior statement had been provided to the plaintiff. Nor does it appear that the appellant's records of complaints regarding discs were sought prior to trial.
55 In substance, this material tended to establish that discs practically never disintegrated and, if they did, did not cause damage to property or personal injury. However, in the present case it was known that such a disc had disintegrated. The question was whether it had done so spontaneously, or when improperly used. Because there was no evidence of other instances where discs had disintegrated in either circumstance, neither the number of discs produced, nor the low level of complaints assisted in resolving the issue in dispute. To adapt an analogy suggested in the course of argument, the fact that the number of motor vehicle deaths, measured against kilometres travelled, is tiny neither proves that a particular person has not died in a motor vehicle accident nor explains whether his death was more likely to have been caused by a defect in the vehicle or negligent driving. Similar difficulties with statistical analysis were noted by Branson J in Medtel Pty Ltd v Courtney [2003] FCAFC 151; 130 FCR 182 at [58]. In a case dealing with the merchantable quality of a heart pacemaker, her Honour stated:
"It follows, in my view, that for the purposes of the subsection, it is not appropriate to attribute to Mr Courtney's pacemaker any qualities derived by statistical analysis of the total population, or pool, of pacemakers from which Mr Courtney's pacemaker came. So, for example, while statistical analysis of the total population of pacemakers from which Mr Courtney's pacemaker came might show that pacemakers of that type have a 2 per cent chance of premature failure, that analysis would reveal nothing about the quality of Mr Courtney's particular pacemaker."
56 However, even that statement was to be read in context. Mr Courtney's pacemaker had not revealed premature failure, but carried an appreciably higher risk of premature failure than it was objectively reasonable to expect in a pacemaker: at [77]. Thus it appears that, in the particular circumstances of that case, the statistical assessment was significant. As explained above, the present case involved a failure of the goods, with the consequence that any bare statistical analysis of likelihood of failure was beside the point.
57 The appellant also sought to call in aid the reasoning of Heydon JA in Makita (Australia) Pty Ltd v Sprowles [2001] NSWCA 305; 52 NSWLR 705 at [102]. That being a negligence case, the liability of the defendant depended on an inference that particular stairs on which the plaintiff had fallen were "slippery". His Honour accepted a history of incident-free use of the stairs to find that they were not slippery, in preference to the view of an expert that they were.
58 The legitimacy of such reasoning in appropriate cases is undeniable: the question is whether it is applicable in the present circumstances. It is not. This is not a case in which one object has been used on many occasions without incident, but is rather one where an object is used once and disintegrates. It provides a circumstance where the comments of Branson J in Medtel are apt, namely that the quality of this product cannot be derived from a statistical analysis of a pool of similar products. Once the product has failed, the relevant question is not the likelihood of failure but the comparative likelihood of failure as a result of inherent defect as against failure from improper use. The reasoning in Makita provides no assistance in answering this question. (A quite separate issue as to whether either of the expert reports satisfied the requirements of Makita with respect to expert opinion evidence might have been raised, but was not.)
59 It is necessary to turn to the evidence of inspection and checking conducted by the appellant in the present case.
Quality control in manufacturing
60 It is convenient to start with a complaint that his Honour declined to make a finding that the quality management systems operated by the appellant conformed to the Australian and New Zealand standard with respect to "the design and manufacture of coated and resinoid bonded abrasives". The plaintiff's case was not run on the basis of a failure to comply with such standards and there was a certificate of registration granted by what appeared to be an industry regulatory authority dated 3 July 2006. His Honour's doubts as to the degree of conformity to the standard, expressed at [113], may be put to one side.
61 By way of further preliminary comment, it was curious that no attempt was made by the appellant, from its own knowledge, expertise and experience, to explain the mechanism by which the disc failed. One aspect of the sample testing conducted at the factory was that described as "burst speed testing" which involved accelerating sample discs beyond 1.5 times the maximum design rpm. Mr Portier identified the nature of a burst speed test, at Tcpt, 23/05/08, p 5(40):
"Q. What's a burst speed test?
A. A burst speed test is a test undertaken to determine at which speed a wheel would break and that's then compared to a standard.
Q. Is it a destructive test?
A. It is a destructive test yes."
62 The sample testing required by the appellant during manufacture identified the number of discs, also described as wheels, to be tested per batch, depending upon the size of the batch. For a batch size between 11,000 and 21,000, 40 wheels were to be tested and the system permitted an "allowable number of failures" as two per 40 tested. If that number were exceeded, a further 80 wheels were to be tested, with three permissible failures. The evidence included a record produced by the appellant as a result of testing samples of a particular batch on 9 September 2003. The first sample was taken on commencement of production and further samples were taken thereafter at 30 minute intervals. The sampling involved noting aspects of the appearance of two discs on each occasion. Another form included in the evidence showed that burst speed testing for a particular batch reached 15,300 rpm.
63 It appears that the burst speed testing was done in accordance with the requirements noted above, which involved the selection of 40 discs from a batch of between 11,000 and 21,000 discs. The procedures described the burst speed test as "the sole minimum requirement for accepting a batch". In the event that the test failed, the product (presumably the whole batch) was put on "hold" and a relevant manager was required to determine the appropriate action.
64 The evidence given by the plaintiff and Mr Keenan was that the disc in question disintegrated within one or two seconds of the grinder being switched on. This may have occurred during a period of fast acceleration. There was no evidence as to how long the acceleration would last or whether greater pressures were placed on the disc during such a period. Nor did the evidence establish the variety of results achieved by the routine burst speed tests. Such evidence, which must have been available to the appellant, might have demonstrated a narrow band of speeds required to destroy discs, or it might have established a significant degree of variation. Clearly some degree of variation was expected. How much variation was to be expected might well affect the assessment of the likelihood that a particular disc would disintegrate at a speed below its maximum design operating speed.
65 Although, on one view, relatively small numbers of samples were taken, counsel noted that there were numerous manufacturing safeguards designed to ensure uniformity within a batch. On the other hand, it was not entirely clear how that principle applied with respect to the failure rate of sample testing. For example, the maximum operating speed indicated on the disc which failed was 7,500 rpm. The standard required that the sample discs not disintegrate at speeds of 11,250 rpm. However, if two out of 40 sample discs in a batch did disintegrate below the standard required, the batch would apparently be released for sale. Consistency of production suggested there might well be other discs within such a large batch which would similarly disintegrate at a speed lower than the standard. There was no evidence as to what steps were taken (if any) if a disc disintegrated during testing at a speed below the maximum design speed for that disc.
66 How this evidence was to be assessed was quite unclear. That it complied with the relevant standard may be accepted. Further, the testing process may have ensured that a very small number of defective discs would be released for sale. It was not, however, designed to ensure that no discs were released which did not conform to the standard. It appeared that an allowable failure rate amongst the sample pool could rise to 5% without a batch being withdrawn.
67 No doubt the quality control measures demonstrated that defective discs would be rare or, to put the matter differently, that the chance that any particular disc was defective was very low. However, they did not demonstrate a likely absence of defective discs, nor that it was glaringly improbable that one disc in several million would disintegrate at a speed below the maximum design operating speed.
68 If the trial judge had taken this material into account when assessing the likelihood that the plaintiff and Mr Keenan were accurately describing what happened on the day of the accident, that might have affected his Honour's assessment of their evidence. However, it is unlikely that his Honour would have come to a different conclusion. That follows in part from his Honour's conclusion with respect to the statutory defence, namely that the evidence did not show that it was more probable than not that the defect (assumed for that purpose) occurred through rough handling during transportation, rather than during the process of manufacture and packaging. Accordingly, his Honour's error in assessing the evidence as to when the disc disintegrated did not give rise to any miscarriage of justice.
69 The appellant was content to invite this Court to make its own findings on the basis of the objective evidence, expressly stating that this was not a matter which should be sent back for a retrial: written submissions, par 75. Accepting his Honour's appraisal of Mr Keenan and the plaintiff and acknowledging the possibility that while Mr Keenan was genuinely attempting to tell the truth as he saw it, there was room for him to be mistaken, I would not be persuaded that the account given by the plaintiff and Mr Keenan was inherently implausible or glaringly improbable. The appellant's evidence supported the probability that defective discs would, on rare occasions, be released to consumers. It also left open the possibility that such a disc might disintegrate at a speed below the maximum design operating speed. In this respect, the appeal should be rejected.
Statutory defences
70 Accepting that the disc was defective, for the purposes of s 75AK of the Trade Practices Act, the appellant argued that the evidence supported the conclusion, on the probabilities, that the defect in the disc did not exist at the time they were supplied by the appellant.
71 There were in fact two causes of action and two separately worded defences. As the trial judge noted at [115], the close relationship between the two provisions had been identified by Handley JA (Tobias JA and Palmer J agreeing) in Effem Foods Ltd v Nicholls [2004] NSWCA 332; (2004) ATPR ¶42-034 at [19]:
"Section 74D(2)(a)(i) relevantly requires proof that the defect arose from the act of a stranger after the goods left the control of the manufacturer. Section 75AK(1)(a) requires proof that the defect did not exist at that time. At least in this case I can see no practical difference in the scope of these defences. The first requires proof that the defect occurred after supply, the second proof that it did not exist at the time of supply. The defences are simply two sides of the same coin."
72 Effem Foods involved what Handley JA described as "an encounter between a Snickers bar, a safety pin and [the consumer]" at [1]. The trial judge held that similar reasoning applied to the defences in the present case, an opinion which was not challenged on appeal.
73 Again, in a passage in Effem Foods applied by the trial judge without demur on appeal, Handley JA referred to a submission that the liability of a manufacturer was strict, but not absolute and asking what more the manufacturer could have been expected to do in order to discharge its onus of proof: at [30]. His Honour continued:
"His point is valid, so far as it goes, but the manufacturer has to establish these defences on the balance of probabilities and speculation and proof of mere possibilities are not enough. This does not mean that a manufacturer's liability is absolute. There is scope for these defences where an examination of the product after the accident establishes that it has been deliberately tampered with. … A manufacturer is not required to lead direct evidence to support these defences and a case based on circumstantial evidence is capable of discharging the onus. In the present case the defendant's evidence did not rise above the level of a speculative possibility."
74 The appellant also sought support in the decision of the English Court of Appeal in Piper v JRI (Manufacturing) Ltd [2006] EWCA Civ 1344; All ER (D) 181. That case involved a claim under the Consumer Protection Act 1987 (UK) based on liability for a defective product, comparable to the legal principle relied upon in the present case. The loss in question was caused by the failure of a prosthesis inserted as part of a total hip replacement operation. A significant difference between that case and the present was that in Piper, by the time of the trial, expert analysis had demonstrated that the fracture was "a result of fatigue failure initiating from a defect in the titanium alloy from which it was made": at [8]. The case therefore turned on a statutory defence relied on by the manufacturer.
75 The evidence called by the manufacturer was that by the time of the incident, it had produced more than 80,000 of this type of prosthesis, with a failure rate of the kind which arose in that case of only five. (Given the consequences of failure, that figure probably recorded, at a reasonable level of reliability, the actual failures.) The evidence demonstrated that each product was inspected twice after manufacture, that a defect measuring 35 microns would be visible to the human eye and that, if there were any suspicious mark, firstly it would stand out on the highly polished surface and, secondly, inspectors could examine it with magnifying glasses which they carried: at [21]. Although the plaintiff's case was that the defect which initiated the fatigue failure was quite small, it did not suggest that it was smaller than 35 microns. Further, the inspectors were found to have been "highly experienced and conscientious": at [29].
76 While Piper provides a good example of circumstances in which the statutory defence may be established, it provides little assistance in determining the present case. Indeed, there would appear to be a quite dramatic distinction between the mass production of discs subject to random checking, as in the present case, and the individual attention given to the prostheses in Piper.
77 The evidence in the present case demonstrated that, whilst discs for use in angle grinders for cutting metal are extremely hard, they also tend to be brittle and capable of damage through rough handling. The Australian standard for "abrasive wheels" noted, in part:
"In general, inorganic bonds (vitrified, silicate and magnesite) are more brittle than organic bonds (resinoid, shellac and rubber). Hence, inorganic bonded wheels can be broken more readily by handling, shock or impact."
78 Both experts accepted that the disc in question was subject to cracking if dropped on a hard surface or otherwise subjected to sharp shocks. Mr Peterson expressed the view that the disc "may have had damage sustained in manufacture or packing, or more likely damaged in transport due to impact or possible thermal effect": second report, 21 January 2008, par 3.13. No reliance was placed upon thermal effects, and Mr Peterson agreed that his preference for damage during transport had been based upon an assumption that "the manufacturer was competent": Tcpt, 21/05/08, p 85(20). His Honour then invited Mr Peterson to take that element out of the equation which, Mr Peterson said, "greatly increases the probability that it was damaged during or as a part of the manufacturing process": Tcpt, p 85(25).
79 His Honour noted that the evidence did not clearly identify the point of supply by the appellant or, for the purposes of s 74D(2), when the goods "left the control of" the appellant. His Honour concluded that the claim that it was more likely that the damage occurred after the disc left the control of the appellant, rather than whilst it was in its control, "is no more than simply speculating": at [120]. His Honour therefore concluded that the appellant had not discharged its onus under the Trade Practices Act.
80 In Effem Foods, the evidence for the manufacturer was largely focused upon the safety procedures and quality control at the factory, although evidence was given as to the chance that the packet had been tampered with and resealed: at [14]. That possibility was described by the trial judge in that case as "very remote" and as "entirely a matter of speculation rather than any positive evidence". That assessment was accepted on appeal.
81 The mechanism for damage to the disc in the present case was far less implausible than the quite deliberate action which would have been required in Effem Foods, involving opening a sealed packet, inserting a safety pin and resealing the packet, being the only plausible mechanism by which to introduce a safety pin into a Snickers bar after manufacture. In the present case, damage might have been done by inadvertent or accidental rough handling. However, such evidence as there was of complaints about defective discs did not suggest a single case in which the defect had been demonstrated to have resulted from such a cause. Just as the absence of evidence of disintegrating discs failed to assist in attributing such an event to misuse rather than defective manufacture, so it also failed to assist in determining whether rough handling in transport was a more likely cause than the failure of quality control mechanisms in the course of manufacture.
82 In addition, there was an absence of evidence as to the kind of mishandling which could cause a defect liable to lead to disintegration, other than a defect which would be readily apparent to the consumer when inserting the disc in the angle grinder. As already indicated, the appellant's evidence covered two out of three bases. Mr Portier had established the level of quality control maintained within the factory, while Dr Casey supported the inference that the blade disintegrated through misuse. Dr Casey stated in his report of 29 November 2007 at par 27:
"Importantly, the flaw [in the disc] is not identified within [Mr Peterson's] report and as such [he] cannot differentiate between a flaw which resulted due to manufacture or a flaw which resulted from rough handling following manufacture. During my inspection of the blade, I could not find any manufacturing flaws in or around the fracture (voids or absence of reinforcing fibres) and as such, it is considered that it is more likely than not, that if a flaw resulted in the break-up of the blade, it did not result from manufacture. Therefore, if a flaw did result in the break-up of the blade, rough handling is considered to be the most likely candidate to have generated this flaw. It is not possible to put a time scale onto the point at which the flaw was introduced to the blade and so it is not possible to state if the flaw was introduced during shipment, sale, storage, installation or during use of the angle grinder."
83 The closest Dr Casey's evidence came to describing how a flaw might have arisen after manufacture was in seeking to rebut the proposition that fragments of the blade, discharged at 250 kilometres per hour, could have rebounded from a hard surface onto the plaintiff's wrist. To test that proposition, he "conducted a number of trials in which [he] threw similar blades against paved surfaces": at par 29. Such "trials" did not demonstrate any great affection for scientific method, but relevantly he noted that the discs "generally cracked at the point where they hit the paving, which accords to the brittle nature of the abrasive matrix within the [disc]". This evidence did no more than confirm a possible cause of weakness, which was not in dispute. It did not demonstrate, even in combination with the appellant's quality control procedures, that the defect was more likely to have arisen in transit, after leaving the control of the appellant, than at an earlier stage.
84 On the basis of the available evidence, no error has been shown in the assessment of the trial judge that the appellant failed to prove that the defect arose after the goods left its control or that the defect did not exist at the time of supply by the appellant. Accordingly, the challenge to the rejection of the statutory defences must fail.
Conclusion
85 Although in one respect the appellant has demonstrated error in the approach adopted by the trial judge to the assessment of evidence as to how the accident occurred, no miscarriage of justice has been demonstrated, nor has any basis been established for this Court to interfere in the findings of fact made by his Honour. His Honour's findings with respect to the statutory defences also being upheld, the appeal must be dismissed with costs.
86 CAMPBELL JA: I agree with Basten JA.
87 SIMPSON J: I agree with Basten JA.
**********