Johansson v. General Motors of Canada Ltd.

Johansson v. General Motors of Canada Ltd. Court Court of Appeal Date 2012-12-04 Citation 2012 NSCA 120 Docket CA 357820 Judge/Registrar/Adjudicator MacDonald, J. Michael (Honourable Chief Justice) (CA); Saunders, Jamie W. S. (Honourable Justice) (CA); Fichaud, Joel E. (Honourable Justice) (CA) Document Type Decision Relations Library Sheet - Johansson v. General Motors of Canada Ltd. - 2012 NSCA 120 - 2012-12-04 - Library Sheet Decision Content NOVA SCOTIA COURT OF APPEAL Citation: Johansson v. General Motors of Canada Ltd., 2012 NSCA 120 Date: 20121204 Docket: CA 357820 Registry: Halifax Between: Maria Johansson, Steven Johansson and Jody Johansson Appellants v. General Motors of Canada Limited Respondent Judges: MacDonald, C.J.N.S., Saunders and Fichaud, JJ.A. Appeal Heard: September 12, 2012, in Halifax, Nova Scotia Held: Appeal allowed, non-suit ruling overturned, costs order vitiated, motion for non-suit dismissed, and new jury trial ordered, with costs of $8,000 plus disbursements payable by the respondent forthwith in any event of the cause, per reasons for judgment of Fichaud, J.A., MacDonald, C.J.N.S., and Saunders, J.A., concurring Counsel: Jamie F. MacGillivray and Nicolle A. Snow, for the appellants Michelle C. Awad, Q.C., Jeff Aucoin and Michael Blades for the respondent Reasons for judgment: CONTENTS Paragraph 1. Background 4 2. Issues 14 3. Court of Appeal’s Standard of Review to a Non-suit Ruling 18 4. The Non-suit Test 23 5. The Evidence 32 6. Analysis of Principal Issue - The Non-suit 49 (a) The Farro Case and the Effect of Fontaine 50 (b) The Plaintiff’s “Tactical Decision” to Sue in Tort 75 (c) Inferring Negligence versus Strict Liability 77 (d) The Need for Evidence of Regulatory Benchmarks or Industry Practice 90 First: The Scope of Jury Inferences 99 Second: The Evidence of “Regulatory Benchmarks” 111 Third: The Need for Expert Evidence 112 Fourth: Is This a “Technical” Issue? 121 Conclusion on the Need for Evidence 124 (e) Could a Jury Reasonably Infer Negligence in this Case? 125 7. Analysis of Secondary Issue - The Roadside Reports 131 8. Conclusion 136 __________________________________________________________________ [1] Mrs. Johansson’s Chevrolet Lumina swerved off the road for no apparent reason. She suffered a serious brain injury. A few years later General Motors Canada issued a recall notice, citing a steering “defect” in the Lumina. Mrs. Johansson sued General Motors Canada for negligence. It was a jury trial. After Mrs. Johansson closed her case, the trial judge granted General Motors Canada’s non-suit motion and dismissed her action. The judge said he was “satisfied that a prima facie case has been established that the plaintiff’s Lumina was defective, that the defect caused the accident”. But he ruled there was no evidence to establish General Motors Canada’s standard of care or breach of that standard. [2] The judge viewed the Supreme Court of Canada’s treatment of res ipsa loquitur in Fontaine v. British Columbia (Official Administrator), [1998] 1 S.C.R. 424 as making it “substantially more difficult” for a plaintiff to resist a non-suit motion. According to the judge, allowing a jury to decide whether or not negligence may be inferred from evidence related to the “defect” is “reasoning [that] falls precisely within the definition of strict liability”. He said either evidence of a “regulatory benchmark” or expert evidence was essential for a prima facie case. These views informed the judge’s non-suit of Mrs. Johansson’s claim. [3] The issue is whether the ruling contains an appealable error in the application of the non-suit test to a negligence action. 1. Background [4] On October 25, 1998, Maria Johansson was driving her 1997 Chevrolet Lumina on the Golden Grove Road near Saint John, New Brunswick. Her seatbelt was buckled. Her son and his girlfriend were in the back seats. It was Sunday afternoon, and the mall in the city had a flea market. It was sunny and clear. The road was near her home, and she had driven it often. On a left bend in the road, her car veered right-ward, struck a culvert and ended in a ditch. Nothing unusual had appeared on the road. No other vehicle, person or animal was involved. Mrs. Johansson was left with a permanent brain injury. [5] General Motors of Canada Limited (“GMC”) had assembled and distributed the Lumina. [6] Mrs. Johansson had no idea why the car swerved to the right. She did not sue at that time. [7] In February 2004, GMC sent the Johanssons a Recall Notice for the 1997 Lumina. The Notice arose from a General Motors Corporation Recall campaign. The Recall campaign resulted from a request for an investigation from the United States National Highway and Traffic Safety Administration to General Motors Corporation (“General Motors US”), followed by failure testing of the steering equipment by General Motors US. General Motors US is GMC’s parent. The Recall Notice said: General Motors has decided that a defect, which relates to motor vehicle safety, exists in certain 1996, 1997, and 1998 model year Buick Regal; 1997 and 1998 model year Chevrolet Lumina, Malibu, and Monte Carlo; 1997 and 1998 model year Oldsmobile Cutlass; 1996 and 1997 model year Oldsmobile Cutlass Supreme; 1998 model year Oldsmobile Intrigue; and 1996 model year Pontiac Grand Prix vehicles. Some of these vehicles have a condition where the lower pinion bearing in the power steering gear may separate. Most reports indicate the driver experienced an intermittent loss of power steering assist when making left turns, usually at low speeds. Power assist is normal in right hand turns. When trying to turn left, some drivers could experience higher resistance or, in a few cases, assist towards the right. If this happens while the vehicle is moving, a crash could result. Your GM dealer will inspect the condition of the lower pinion bearing and replace the lower pinion bearing, or in a few cases, replace the rack and pinion steering gear assembly. This service will be performed for you at no charge. I have italicized the wording that was the focus of the submissions. [8] After reading the Recall Notice, it occurred to Mrs. Johansson and her husband George that the circumstances of her 1998 accident matched the risk identified in the Notice - i.e. her Lumina had “assisted” to the right during an attempted left turn. She sued GMC for negligence. The Statement of Claim pleaded that Maria Johansson’s Lumina was “defective” because of a “defective lower pinion bearing in the power steering gear”, that the manufacture and distribution of a vehicle with such a defect was unsafe, being a particular of negligence by GMC, and that this negligence caused the accident and Mrs. Johansson’s injury. [9] The trial began on April 26, 2011 before a jury, with Justice John Murphy of the Supreme Court of Nova Scotia presiding. Mrs. Johansson presented her evidence over eight days in April and early May 2011. Later I will review the evidence. On May 9, 2011, after she closed her case, GMC moved for a non-suit. [10] In his non-suit ruling, the trial judge found that the lower pinion bearing on the rack and pinion mechanism was “defective”. The rack and pinion subassembly was manufactured by Delphi Automotive LLP, a subsidiary of General Motors US, then provided to GMC and assembled by GMC into the Lumina. The judge found: [22] ... The plaintiff’s Lumina was assembled in Canada by the defendant General Motors of Canada Limited, which distributed the vehicle exclusively in this country. The defective lower pinion bearing was manufactured by Delphi Automotive LLP, which at the time, was a subsidiary corporation of General Motors Corporation. The judge said (para 23) “the relationship between the defendant [GMC], its parent corporation, and the defective part supplier is not determinative of this motion”. [11] By an oral decision on May 11, 2011, followed by written reasons on September 23, 2011 (2011 NSSC 352), Justice Murphy granted the motion for a non-suit and dismissed Mrs. Johansson’s action. His Order of January 19, 2012 ordered Mrs. Johansson to pay costs of $47,750 plus disbursements of $17,752.58, for a total of $65,502.58, to GMC. The judge’s written reasons concluded: [5] While Mrs. Johansson was navigating a slight left turn in the road, the Lumina continued straight rather than turning, left the roadway, and crashed into a ditch. The accident caused injuries to all occupants of the vehicle, but Mrs. Johansson’s injuries were particularly severe and permanent. ... [15] With respect to the first two issues, I am satisfied that the plaintiff has established a prima facie case that the 1997 Lumina was defective and that the defect caused the accident. ... [20] In the case at bar, a jury, properly instructed on the law could, on the facts adduced, find that the plaintiff’s Lumina was defective and that the defect caused the accident. The unexplained nature of the accident, combined with its circumstances, and the nature of the Recall Notice, could support a finding that it is more likely than not that the accident was caused by a defective lower pinion bearing rather than any other competing explanation such as driver error; this is sufficient to satisfy the prima facie case threshold on a nonsuit motion. ... [28] As an assembler or as a distributer [sic], I am satisfied that the defendant owed the plaintiff a duty of care to take all reasonable precautions in the assembly or distribution of the Lumina. I am also satisfied that it was reasonably foreseeable that failure to take such precautions could lead to a defective automobile being delivered to a consumer, possibly resulting in an injurious accident. I am further satisfied that a prima facie case has been established that the plaintiff’s Lumina was defective, that the defect caused the accident, and that the plaintiff suffered damages as a result. Where the plaintiff’s case falters is the element of negligence requiring establishment of the relevant standard of care and breach of that standard. [12] The judge granted the non-suit because, in his view, there was no evidence to establish the standard of care or GMC’s breach of the standard. Later I will review the judge’s reasons. [13] Mrs. Johansson appeals the non-suit ruling. 2. Issues [14] The judge’s rulings that (1) “the defendant owed the plaintiff a duty of care to take all reasonable precautions in the assembly or distribution of the Lumina”, (2) “it was reasonably foreseeable that failure to take such precautions could lead to a defective automobile being delivered to a consumer, possibly resulting in an injurious accident”, and (3) “the plaintiff’s Lumina was defective, that the defect caused the accident, and that the plaintiff suffered damages as a result” are not challenged on this appeal. [15] This appeal concerns the judge’s ruling: [28] ... Where the plaintiff’s case falters is the element of negligence requiring establishment of the relevant standard of care and breach of that standard.” [16] I will consolidate the submissions into one question - Did the judge commit an appealable error, in his application of the principles that govern a non-suit in a negligence claim, by ruling that there was no evidence from which a properly instructed jury could infer GMC’s standard of care or GMC’s breach of a standard of care? That is the principal issue. [17] Mrs. Johansson also submits that the judge erred by not considering Roadside Reports of other steering related incidents involving General Motors vehicles. I will discuss that secondary issue near the end of these reasons. 3. Court of Appeal’s Standard of Review to a Non-suit Ruling [18] This Court applies correctness to issues of law on an appeal from a judge: Housen v. Nikolaisen, [2002] 2 S.C.R. 235, paras 8-9. [19] The issue here is legal. In MacDonell v. M & M Developments Ltd., [1998] N.S.J. No. 49 (Q.L.) (C.A.), Justice Hallett said: 40 In Wentzell v. Spidle (1987), 81 N.S.R. (2d) 200 this Court implicitly accepted the opinion expressed in Sopinka, The Law of Evidence, that the assessment of the sufficiency of the evidence before a trial judge on a non-suit motion is a question of law and it is, therefore, subject to appellate review. [20] In Herman v. Woodworth, [1998] N.S.J. No. 38 (Q.L.) (C.A.), Justice Flinn elaborated: 4 In an application for a non-suit, following the close of the plaintiff’s case at trial, the question as to whether the plaintiff has established a prima facie case is a question of law. As such, it is reviewable by this Court. The following passage from The Law of Evidence in Civil Cases, Sopinka and Lederman, 1974 at p. 521 has been cited with approval by this Court ... ... This decision of the judge on the sufficiency of evidence is a question of law; he is not ruling upon the weight or the believability of the evidence which is a question of fact. Because it is a question of law, the judge’s assessment of the probative sufficiency of the plaintiff’s evidence, or the defendant’s evidence on a counterclaim for that matter, is subject to review by the Court of Appeal. [Justice Flinn’s emphasis] [21] The parties’ factums agreed that this Court’s standard is correctness. [22] I will apply correctness as stated in MacDonell and Herman. 4. The Non-suit Test [23] The judge’s decision said: [11] ... After the close of the plaintiffs’ case, the defendant brought this nonsuit motion pursuant to Rule 51.06(1) of the Nova Scotia Civil Procedure Rules. [24] Rule 51.06(1) says: At the close of the plaintiff’s case and before the defendant elects whether to open the defendant’s case and present evidence, the defendant may make a motion for dismissal of the proceeding, or a claim in the proceeding, on the ground that there is no evidence on which a properly instructed jury could find for the plaintiff. [25] Rule 51.06(1) in the 2009 Rules permits a dismissal “of the proceeding, or a claim in the proceeding”. The former Rule 30.08, from the 1972 Rules, permitted only the “dismissal of the proceeding”. That difference is immaterial here, because the judge dismissed the entire action. The test for a non-suit has not changed with the adoption of the new Rule 51.06(1). As expressed in the trial judge’s reasons, with which I agree on this point, the case law under the former Rule 30.08 still frames the test under Rule 51.06(1). [26] In MacDonell v. M & M Developments Ltd., Justice Hallett described the approach: The Law Applicable to Non-Suit Motions 38 On a non-suit motion, the trial judge has to consider all of the circumstances, including the issues of fact and law raised by the pleadings (J.W. Cowie Enrg. Ltd. v. Allen (1982), 52 N.S.R. (2d) 321). 39 The general test for a non-suit motion is whether or not a prima facie case was made out by the plaintiffs. It is sometimes expressed as whether a jury, properly instructed on the law could, on the facts adduced, find in favour of the plaintiff. If not, the motion will succeed. (Turner-Lienaux v. Nova Scotia A.G. (1993), 122 N.S.R. (2d) 119). [27] In Herman v. Woodworth, Justice Flinn (para 4), more expansively, adopted the following from Sopinka’s The Law of Evidence in Civil Cases: ... If such a motion is launched, it is the judge’s function to determine whether any facts have been established by the plaintiff from which liability, if it is in issue, may be inferred. It is the jury’s duty to say whether, from those facts when submitted to it, liability ought to be inferred. The judge, in performing his function, does not decide whether in fact he believes the evidence. He has to decide whether there is enough evidence, if left uncontradicted, to satisfy a reasonable man. He must conclude whether a reasonable jury could find in the plaintiff’s favour if it believed the evidence given in trial up to that point. The judge does not decide whether the jury will accept the evidence, but whether the inference that the plaintiff seeks in his favour could be drawn from the evidence adduced, if the jury chose to accept it. [Justice Flinn’s emphasis] [28] The seminal statement of the demarcation between the functions of judge and jury is Lord Cairns’ passage in Metropolitan Railway Co. v. Jackson (1877), 3 App. Cas. 193 (H.L.), at p. 197: The Judge has a certain duty to discharge, and the jurors have another and a different duty. The Judge has to say whether any facts have been established by evidence from which negligence may be reasonably inferred; the jurors have to say whether, from those facts, when submitted to them, negligence ought to be inferred. It is, in my opinion, of the greatest importance in the administration of justice that these separate functions should be maintained, and should be maintained distinct. [Lord Cairns’ emphasis] The Supreme Court of Canada repeatedly has adopted Lord Cairns’ statement: The King v. Morabito, [1949] S.C.R. 172, at p. 174; Mezzo v. The Queen, [1986] 1 S.C.R. 802, at para 54, per McIntyre, J. for the majority; R. v. Arcuri, [2001] 2 S.C.R. 828, at para 24, per McLachlin, C.J.C. for the Court. [29] As to other provinces, in Prudential Securities Credit Corp., LLC v. Cobrand Foods Ltd., [2007] O.J. No. 2297 (Q.L.) (C.A.), Justice Laskin for the Court said: 35 On a non-suit motion, the trial judge undertakes a limited inquiry. Two relevant principles that guide this inquiry are these. First, if a plaintiff puts forward some evidence on all elements of its claim, the judge must dismiss the motion. Second, in assessing whether a plaintiff has made out a prima facie case, the judge must assume the evidence to be true and must assign “the most favourable meaning” to evidence capable of giving rise to competing inferences. This court discussed this latter principle in Hall et al. v. Pemberton (1974), 5 O.R. (2d) 438, at 438-9, quoting Parfitt v. Lawless (1872), 41 L.J.P. and M. 68 at 71-72: I conceive, therefore, that in judging whether there is in any case evidence for a jury the Judge must weigh the evidence given, must assign what he conceives to be the most favourable meaning which can reasonably be attributed to any ambiguous statements, and determine on the whole what tendency the evidence has to establish the issue. ... From every fact that is proved, legitimate and reasonable inferences may of course be drawn, and all that is fairly deducible from the evidence is as much proved, for the purpose of a prima facie case, as if it had been proved directly. I conceive, therefore, that in discussing whether there is in any case evidence to go to the jury, what the Court has to consider is this, whether, assuming the evidence to be true, and adding to the direct proof all such inferences of fact as in the exercise of a reasonable intelligence the jury would be warranted in drawing from it, there is sufficient to support the issue. 36 In other words, on a non-suit motion the trial judge should not determine whether the competing inferences available to the defendant on the evidence rebut the plaintiff’s prima facie case. The trial judge should make that determination at the end of the trial, not on the non-suit motion. See John Sopinka, Sidney N. Lederman and Alan W. Bryant, The Law of Evidence in Canada, 2d ed. (Toronto: Butterworths Canada, 1999) at 139. [30] In Capital Estate Planning Corp. v. Lynch, [2011] A.J. No. 820, para 19, the Alberta Court of Appeal described Justice Laskin’s statement from Prudential Securities as “[t]he definitive test on a non-suit motion”. [31] In my view, all the above statements are consistent and authoritative, and vary only in the nuances of expressing the governing principle. 5. The Evidence [32] Delphi Automotive LLP, a subsidiary of General Motors US, manufactured the defective rack and pinion steering mechanism. GMC acquired the mechanism and installed it into the Lumina. Then GMC distributed the Lumina for sale by a GMC dealer to Mr. Johansson. There is no direct evidence of GMC’s acts or omissions in the quality control, inspection or testing of the defective rack and pinion mechanism before GMC installed it in the vehicle for retail. Accordingly, whether the evidence establishes a prima facie case that GMC breached its standard of care depends on what inferences reasonably may be drawn by a jury from the circumstantial evidence that is on the record. This means it is essential to examine that evidence, which I summarize as follows. [33] The Agreed Statement of Facts included the following. On February 12, 1997, the Johanssons bought a 1997 Chevrolet Lumina from a Chevrolet dealer in Saint John, New Brunswick. At about 1 PM on Sunday, October 25, 1998, the Lumina was involved in a single vehicle accident while travelling southwest on Golden Grove Road in Saint John, New Brunswick. The Lumina left the travelled part of the road, and struck a culvert on the right side. Maria Johansson was driving, and was wearing a seatbelt. Stephen and Jody Johansson were backseat passengers. The Lumina’s odometer showed about 79,500 kilometers at the time of the accident. In February 2004 GMC issued a Recall Notice for the Lumina. [34] Maria Johansson testified. But she recalled nothing about the accident. She recalled leaving home, driving her car, and waking in the hospital. She had driven that road many times. [35] Stephen Johansson, Maria’s son, testified that he was in the car at the time. The weather was sunny and clear. They were going to the flea market in the City, [a]nd at one point in the road, it sort of ... it comes down into a turn, and it would have been shortly after that I remember hearing gravel, like the tires hitting gravel, and then hitting into the ditch, I guess. He added that no other vehicle, person or animal was involved. He testified: I know there was a lot of speculation of why she would have went into the ditch like that. There was no real apparent reason. [36] Jody Johansson, Stephen’s spouse, testified that she also was in the car. She said that nothing unusual happened before the car left the road. [37] George Johansson testified. He is Maria’s husband. He said that they lived in Rothesay, near Saint John, New Brunswick. He said that he had purchased the 1997 Chevrolet Lumina on February 12, 1997. The car had been a demonstrator with some mileage when he bought it. He referred to the Warranty and Owner Assistance Information, and the Owner’s Manual, that were entered as Exhibits. As to the cause of the accident, Mr. Johansson said: I couldn’t understand anything about how it occurred because one, I wasn’t there with my wife or driving the car with her in the vehicle. It bothered me that she drove that road so many times how come, bang, she’s into a ditch. [38] George Johansson testified that in February 2004, he received a Recall Notice from General Motors Canada. The Notice applied to his 1997 Lumina. The Recall Notice was entered as an Exhibit, in the jointly agreed Exhibit Book. The Notice, from GMC to George Johansson, is quoted above (para 7). George Johansson testified, about the Recall Notice: Q. Okay, and so what did you think when you read that? A. When I read that it described my wife’s accident to a “T”. Q. Okay. A. She was going along Golden Grove Road. She made a left, slight left turn, and she went into the ditch. Q. In what direction did she go into the ditch, what did you understand? A. She went into the right side. [39] After considering the implications of the Recall Notice, Mrs. Johansson started this lawsuit. [40] Over GMC’s objection, the judge qualified Dr. Stuart Smith as an expert engineer in accident investigation and reconstruction. He filed a report and testified. He noted that he had not been able to inspect the Johanssons’ Lumina, because it had been written off and sold to a salvage company. His report said: According to the documentation reviewed, the recalled defect can in a few cases cause assist (i.e. steering force) to be applied to the right while the driver is trying to turn to the left. This unwanted assist to the right is very probably a result of malfunction of the control valve on separation (failure) of the pinion bearing. ... The circumstances of the accident, with the vehicle leaving the road on the right hand side while the driver was attempting to negotiate a left hand curve, are consistent with the power steering failing to provide proper assist and/or applying unwanted assist to the right, as described in the Recall Notice. The possibility of an accident being caused in this way is specifically mentioned in the Recall Notice: “If this happens while the vehicle is moving, a crash could result”. [Dr. Smith’s emphasis] Dr. Smith’s report attached diagrams to show the mechanics of the failure. [41] Dr. Smith’s testimony elaborated on the failure: Q. ... Can you describe in layman terms what that means, what I just read from your report? A. It means that the bearing at the bottom of the steering shaft which we’re going to look at later was prone to failure. The way that the failure occurred was that a retaining ring or clip would fall out. And after that, the ball bearings themselves would fall out. And the end of the steering shaft would then not be properly supported by the bearing, and would be able to shift and flop and move around as the load on the steering changed. Q. All right. And maybe what I’ll do ... what would be the consequences of that in terms of the bearings falling out and ... A. Well, in general it would be erratic steering in a somewhat unpredictable way, although the recall notice does say that the most common symptom is a loss of assist when trying to make left turns. And even in some cases that when the driver attempts to turn the wheel to the left, the steering can go to the right. [42] Using components of the rack and pinion steering assembly that had been entered as demonstrative evidence, Dr. Smith explained to the jury how the failure occurred. I will not quote all this testimony, that occupies several pages of the transcript. He concluded: And the problem with this particular system was that that bearing would tend to fail. And eventually the balls in the ball bearing would fall out, and they would be collected. They wouldn’t go down (on the road ?) [question mark in the transcript] because there’s a dust cap on the bottom which is this piece. So if the bearing had failed and you took it apart, if you took this dust cap off, it would have some of the ball bearing caught in it. And when that happens, the end of the pinion gear shaft ... (inaudible) and moves around. So instead of just being kept (centered in line), this can flop around and, in a somewhat unpredictable way, cause this control valve to move in a somewhat unpredictable way and cause the hydraulic fluid to go into the rack in a somewhat unpredictable way, so the car doesn’t go where you try to steer it. So that was the defect that was recalled. [43] The trial evidence included the discovery transcript of Mr. Bill Davies, the Canadian Manager of Field Performance Assessment for GMC. Mr. Davies was since deceased. Mr. Davies’ transcript included: Q. So after my review of the documents that were submitted on behalf of GM Canada, it would appear to me that there’s some information that is not in dispute, and I’m going to go over that with you and see if I can get your agreement with that. October 2002, there was a recall on approximately 1.5 million cars under Recall No. 02049. A. Correct. Q. Okay. And that those cars contained defective parts or a defective part which relate to the safety of the motor vehicle in question? A. Yes. Q. And that those defective parts included the power steering gear, lower pinion bearing separation - or I guess it was a separation is what was happening with the defective part. Is that correct? A. Correct. Q. Okay. In December 2003, another 824,186 cars, approximately, were added to the recall under Recall No. 03062. A. Correct. Q. And those cars also had defective parts related to the power steering gear, lower pinion bearing. A. Correct. Q. And one of the make or models of cars included in the Recall No. 03062 was the 1997 Chevrolet Lumina? A. Correct. Q. And it is your understanding that the, that the Plaintiffs in this case were in a 1997 Chevrolet Lumina at the time of the accident? A. Yes. ... Q. Okay. So the direction, then, on the recall came from the investigation which occurred in the US, and GM Canada simply followed suit? A. Correct. Q. Okay. Is there any particular reason that you’re aware of to why GM Canada didn’t find it necessary to do their own investigation? A. In these situations, there is usually an investigation which is undertaken by the area, if you will, that’s best able to gather all the data. The General Motors Corporation undertook the investigation at the request of the National Highway and Safety Administration, and it would be redundant for GM of Canada to do exactly the same thing. ... Q. Okay. The source ... A. The racks ... Q. ... being GM. A. The steering racks that were replaced came from Delphi. Q. Okay. So - and I’m just going to outline what I understand to be the result of the failure, what would, what would be happening with the failure. My understanding from the documentation that I have read here is that if the lower pinion bearing in the rack and pinion steering gear separated, that that would then cause the pinion shaft to move. A. Correct. Q. With the pinion shaft moving, the driver would then be experiencing the difficulty, and it seems to be with the left turns. A. Correct. Q. And that this could be followed by the wheel then involuntarily turning to the right. A. (Pause) I, I think the way that the recall was written was that when you got into a situation where there was total failure, meaning all the ball bearings in, and the cage were completely in a position where they weren’t of any use to hold the pinion, that the pinion would move, which allowed redirection of fluid. And when that occurred, if you tried to move the wheel to the left, the vehicle would actually move to the right. And why they said on the recall that this was an extremely rare occurrence was because you had a graduated failure where you would start to see this and it wasn’t until all of the bearing was completely gone that you got into a situation where you actually had to redirect. So these conditions were very rare, though that would be the ultimate problem. Q. The ultimate failure. A. Yeah. [44] Also in evidence were DVDs showing General Motors US’ systematic testing of the lower pinion system in 2002, and the failure of the bearings. In response to an objection by GMC, the trial judge issued a mid-trial ruling that the DVDs were admissible and said “It seems to me that the DVD footage is probative in that it shows that the defect occurs in practice”. The videos were shown to the jury during Dr. Smith’s testimony. Dr. Smith’s testimonial commentary includes, for example: A. The retainer that retains the balls in the bearings is starting to come loose. The retainer has completely come loose, and the balls are now no longer evenly spaced around the bearing. I believe there are eight balls in the bearing if I count properly. It’s going to continue. Now we’re looking at 10:52 p.m., so we’ve jumped to a later time on April 29th, 2002. I think I’m saying the obvious, that the shaft is now wobbling. ... MS. SNOW: I think we had a second one that we were going to show as well. And before I do that, Dr. Smith, would you be able to comment on what was happening with those events near the end, and you’d pointed out when the ball bearings were coming loose? A. The ball bearings were coming out of the bearing assembly. Q. Okay. A. And if they were ... they’re in a test obviously. But if they were in a car, the balls would end up falling into that cup at the bottom of the steering shaft. [45] The jointly submitted book of exhibits included a General Motors safety recall bulletin with a protocol for a dealer’s mechanic to inspect and replace the defective steering mechanism. The agreed Book of Exhibits also contained a list of answers prepared by General Motors to anticipated questions that their dealers might face, concerning the recall, including: Q2: What might occur as a result of this condition? A2: If the lower bearing separates and the pinion shaft moves, the driver may experience an intermittent increase in steering efforts when turning left. If the pinion shaft moves too far, the driver could experience a higher resistance and, in a few cases, unintended power assist to the right. ... Q5: Can we expect that the condition will surface in other 1996-98 GM vehicles and will have to be recalled? A5: No. GM and the supplier have conducted an extensive evaluation and analysis of all vehicle applications. Certain forces on the bearing are highest in the vehicles included in this recall. We do not expect to recall other 1996-98 GM cars due to this particular condition. ... Q7: How does “assist to the right” happen? A7: As the driver steers the vehicle to the left, the pinion shaft moves and power steering fluid is misdirected, so the power assist is to the right instead of to the left. ... Q12: What was the cause of the condition? A12: Some of the lower pinion bearings were not manufactured properly and can separate with vehicle usage over time. When the bearing separates, steering performance may change when the pinion shaft moves upward in the valve housing during some left turns. [46] At the hearing in the Court of Appeal, GMC’s counsel opened her submissions by stating: The fact is this: We know exactly how this defect arose. It’s in evidence and we have demonstrative exhibits that will actually assist the Court, and with your permission, I want to use them for about 5 minutes to show you, put some real visual into the defect that we’ve been talking about and the thing that was mentioned in the Recall Notice and which, my friends say, should allow the Court to infer the standard of care. She then demonstrated, with components of the steering rack and pinion assembly that were in evidence, how the failure occurred. Her explanation was consistent with the Recall Notice, the DVDs and Dr. Smith’s explanation. GMC’s counsel referred to the evidence that GMC had assembled the vehicle with the part from Delphi. She continued: So there’s no need for this Plaintiff to get inside the factory, and figure out who did what and what might have gone wrong. [47] The Exhibits in the agreed joint exhibit book included General Motors’ Chevrolet Lumina 1997 Sales Brochure. The Brochure opened with the statement: For countless families across Canada, choosing a new car comes down to a single word. Trust. They need a car they can depend on, day after day and month after month, in bad weather and in good. A car they can sometimes take for granted, because other things - kids, work, commitments - have to take priority. A car they can count on to protect. For these people - and perhaps for you as well - the new Chevrolet Lumina is everything it should be. Tried, tested and true - and a car you can have confidence in every day. The Brochure then spoke of the Lumina’s “smooth, strong performance” and “full complement of safety features”, said “Lumina is engineered as a Total Safety System from one end to the other”, and mentioned Lumina’s “refined power-assisted rack-and-pinion steering”. [48] The Owner’s Manual for the Johanssons’ Lumina was in evidence. The manual includes: REPORTING SAFETY DEFECTS TO THE UNITED STATES GOVERNMENT If you believe that your vehicle has a defect which could cause a crash or could cause injury or death, you should immediately inform the National Highway Traffic Safety Administration (NHTSA), in addition to notifying General Motors. According to GMC’s Mr. Davies (above, para 43), it was the NHTSA which requested General Motors US to conduct the investigation that led to the Recall Notice. 6. Analysis of Principal Issue - The Non-suit [49] The judge’s analysis followed several intersecting reasoning paths that I will address separately. (a) The Farro Case and the Effect of Fontaine [50] Mrs. Johansson cited to the judge the decision of the Ontario Court of Appeal in Farro v. Nutone Electrical Ltd. (1990), 72 O.R. (2d) 637, [1990] O.J. No. 492 (Q.L.). The Court of Appeal said that, in a products liability claim, the standard of care and its breach may be inferred from the evidence surrounding the defect. The judge (paras 44-47) discussed Farro at length. [51] In Farro the plaintiff had suffered damages from a defective part supplied by a third party to the defendant manufacturer of a ceiling exhaust fan. The defective part had been destroyed and was unavailable for testing. In those respects, Farro resembles Mrs. Johansson’s case. The action was tried before a judge without a jury. In Farro, the trial judge held there was no evidence as to the appropriate standard for the manufacture of the mechanical part, and therefore no proof that the standard had been breached. There was no mid-trial non-suit. After the trial, the judge dismissed the action. The Ontario Court of Appeal reversed the judge and allowed the action for damages. [52] Justice Lacourciere for the Court of Appeal said: 11 A manufacturer has a duty to take reasonable care in the manufacture of his product, including all its component parts, and failure to take such reasonable care can result in liability to the ultimate user or consumer. 12 In Charlesworth on Negligence, 5th ed., at p. 394, paras. 631-2, the following appears: The duty of the manufacturer may be said to be to take reasonable care in the manufacture of his product, and failure to take such care will render him liable to any consumer or user whose person or property is injured by his product, provided (1) the product causing the injury has the same defect as it had when it left the manufacturer; and (2) the manufacturer should have contemplated that the product would be consumed or used in the same condition as it was when it left him. Component parts. A manufacturer’s duty is not limited to those parts of his product which he makes himself. It extends to component parts, supplied by his submanufacturers or others, which he uses in the manufacture of his own products. He must take reasonable care, by inspection or otherwise, to see that those parts can properly be used to put his product in a condition in which it can be safely used or consumed in the contemplated manner by the ultimate user or consumer. The last proposition is based on Macpherson v. Buick Motor Co., 217 N.Y. 382 (1916), referred to with approval by Lords Atkin and MacMillan in Donoghue v. Stevenson, [1932] A.C. 562 (H.L.). ... 15 ... The clear finding, well supported by the evidence, is that the fan motor caused the fire. The appellants not only established a prima facie case of a defect in the exhaust fan’s component but also eliminated all possible extraneous causes of the fire. 16 In Prosser on Law of Torts, 3rd ed. (1964), p. 671, the learned author discussed the difficulty in laying the groundwork leading to an inference of the manufacturer’s negligence: Since the injured plaintiff almost never has any direct proof of what has occurred in the manufacturer’s plant, he usually must resort to circumstantial evidence. In the ordinary case this means that he must rely upon the doctrine of res ipsa loquitur. In order to bring himself within the doctrine, he must make it appear that the cause of the injury was something which lay within the responsibility of the defendant. In other words, he must introduce evidence to exclude the possibility that it was due to his own conduct, or that of intermediate handlers or any meddling third party. ... 18 ... Having found that the fan motor caused the fire, the learned trial judge, with great respect, placed too heavy a burden on the appellants to show how the particular defect occurred. There was evidence of sufficient weight and cogency to support an inference that the fan motor was defective when it left the manufacturer’s plant. 19 It is not necessary, and it is generally impossible for a plaintiff to adduce direct evidence that a defect existed when the manufactured product left the factory. I adopt the language of MacKeigan C.J.N.S. delivering the judgment of the court in Smith v. Inglis Ltd. (1978), 83 D.L.R. (3d) 215 at p. 218, 25 N.S.R. (2d) 38, 3 B.L.R. 153 (C.A.): I am very respectfully of the opinion that the appellant’s case against the respondent should not be judged by a comparison of likelihoods, a test which imposes a burden on the appellant of showing exactly where and how the negligent deed was done, a burden which I do not think the law requires him to assume. The appellant does, however, have the burden of establishing on a preponderance of probability on the evidence as a whole that the respondent was the agency responsible for the defect which caused him injury. This he may discharge by showing circumstantially that the defect must have been there when the refrigerator left the factory. He can in my opinion do this if he can exclude the probability of some other person having created the hazard after the product left the factory. [53] The judge, in Mrs. Johansson’s case, distinguished Farro on two bases. He said: [47] There are two reasons why Farro must be approached with caution in the context of this case. Firstly, Farro is distinguishable from this case. In Farro, the circumstantial evidence excluded all other possible causes of the fire at issue in that case, and it also excluded the possibility that the defect could have arisen after the product left the manufacturer. That is significantly stronger circumstantial evidence than what was presented by the plaintiff in this case. Secondly, and perhaps more importantly, the doctrine of res ipsa loquitur was still accepted in law at the time Farro was decided. It is clear that res ipsa loquitur played a key role in the Court’s willingness to infer a breach of the standard of care once it was established that the defect was present when the product left the manufacturer. Without reliance on res ipsa loquitur it is uncertain whether the Court would have reached the same result. [48] In my view, given that res ipsa loquitur is now to be treated as expired, in a product liability action based on negligence, it is not enough for a plaintiff to establish that the product was defective when it left the manufacturer, assembler or distributor. The plaintiff must establish that the defendant’s actions or inactions, in relation to the part of the process for which it was responsible in bringing the product to the consumer, was below the standard of care expected of that defendant in the circumstances - proof that the product was unsafe and below government safety standards is not sufficient. ... [50] There may be circumstances where it is appropriate to shift the burden of proof. In Snell v. Farrell, [1990] 2 S.C.R. 311 at 321 [Snell], a case dealing with professional malpractice, the Supreme Court of Canada held that while the burden of proof in a civil action is usually with the party who asserts a proposition, “where the subject matter of the allegation lies particularly within the knowledge of one party, that party may be required to prove it.” ... [51] I share the plaintiff’s concern that the extinguishment of the doctrine of res ipsa loquitur may make it quite difficult for product liability plaintiffs to prove their case. Like Sopinka J. [in Snell], if I were convinced that product liability defendants were escaping liability because plaintiffs cannot prove a breach of the standard of care under the traditional approach, I would not hesitate to consider a reverse onus approach. However, in my view, the traditional requirement that the plaintiff prove a breach of the standard of care, in a product liability action for negligence, should not be altered, even considering that the expiry of the doctrine of res ipsa loquitur has made this proof substantially more difficult. [54] I respectfully disagree that Farro is distinguishable on either basis cited by the judge. I will address these two bases in turn. [55] The judge distinguished Farro first because, in Farro, the circumstantial evidence “excluded all other possible causes of the fire at issue in that case, and it also excluded the possibility that the defect could have arisen after the product left the manufacturer”. These comments address causation. Farro’s evidence, the judge said, was “significantly stronger circumstantial evidence than what was presented by the plaintiff in this case”. [56] In my view, this factor does not diminish the application of Farro to Mrs. Johansson’s case. [57] The passages on breach of the standard of care, that were adopted by Farro, derived from authoritative texts and decisions, such as Lord Atkin’s ruling in Donoghue v. Stevenson and Chief Justice MacKeigan’s reasons in the Appeal Division of the Nova Scotia Supreme Court in Smith v. Inglis Limited. Those precedents should carry significant weight in the Supreme Court of Nova Scotia. Yet the judge distanced them because of Farro’s facts that related to a different element of the tort - i.e. causation. [58] The judge correctly recognized that Farro and the other authorities contemplate a permissible inference of negligence only if the defect existed when the product left the defendant’s possession. To that extent an issue of causation may bear on the inference of the standard of care. But, on that point, the judge, in Mrs. Johansson’s case, found: [15] ... I am satisfied that the plaintiff has established a prima facie case that the 1997 Lumina was defective and that the defect caused the accident. ... [20] In the case at bar, a jury, properly instructed on the law could, on the facts adduced, find that the plaintiff’s Lumina was defective and that the defect caused the accident. The unexplained nature of the accident, combined with its circumstances, and the nature of the Recall Notice, could support a finding that it is more likely than not that the accident was caused by a defective lower pinion bearing rather than any other competing explanation such as driver error; this is sufficient to satisfy the prima facie case threshold on a nonsuit motion. [28] ... I am further satisfied that a prima facie case has been established that the plaintiff’s Lumina was defective, that the defect caused the accident, and that the plaintiff suffered damages as a result. These findings are the most that can be expected of a plaintiff’s case on a non-suit motion, respecting causation. All Mrs. Johansson could do, at that stage, is establish a prima facie case. The judge determined that Mrs. Johansson had done so. In Farro, the Court of Appeal entered judgment for the plaintiff. But that was because in Farro the trial was over, and the defendant had no more evidence to offer. [59] I will turn to the evidence for the judge’s prima facie findings of causation in Mrs. Johansson’s case. Earlier I recited (1) the contents of General Motors documents - the Recall Notice and internal documents among GMC staff, (2) the testimony of the GMC witness Mr. Davies, (3) the evidence of General Motors’ testing on the DVDs, (4) the report and testimony of the plaintiff’s expert Dr. Smith, (5) the facts of the accident itself - a vehicle mysteriously veering to the right during a left turn, and (6) the acknowledgement by GMC’s counsel to the Court of Appeal respecting the sufficiency of the trial evidence to show causation of the defect. All the evidence is to the effect that Mrs. Johansson’s accident was caused by the steering defect that existed before the Lumina left GMC’s assembly plant. There is no evidence - and the judge cited none - to suggest either that the accident was caused by something other than the steering defect, or that the steering defect arose after the Lumina left GMC’s possession. [60] Next, the judge’s second basis to distinguish Farro. He noted that Farro predated Fontaine v. British Columbia (Official Administrator), [1998] 1 S.C.R. 424. The judge said (para 51) “the extinguishment of the doctrine of res ipsa loquitur may make it quite difficult for product liability plaintiffs to prove their case” and “the expiry of the doctrine of res ipsa loquitur has made this proof substantially more difficult”. The judge assumed that res ipsa loquitur had shifted a burden to the defendant, which Fontaine returned to the plaintiff. The judge’s ruling made it clear that, in his view, this substantial additional difficulty encumbered Mrs. Johansson’s ability to resist GMC’s non-suit motion. [61] In my opinion the judge misinterpreted Fontaine. [62] In Fontaine, Justice Major for the Court, under the topic “When does res ipsa loquitur apply?”, said: 19 For res ipsa loquitur to arise, the circumstances of the occurrence must permit an inference of negligence attributable to the defendant. The strength or weakness of that inference will depend on the factual circumstances of the case. ... 20 ... It has been held on numerous occasions that evidence of a vehicle leaving the roadway gives rise to an inference of negligence. Whether that will be so in any given case, however, can only be determined after considering the relevant circumstances of the particular case. [emphasis added] [63] Res ipsa loquitur would arise only when the circumstances “permit an inference of negligence attributable to the defendant”. But, if such an inference is permissible, the case should go to the jury anyway under the principles that govern non-suit motions. See Lord Cairns’ statement in Metropolitan Railway (above, para 28). This meant that - certainly in jury trials - res ipsa loquitur was “superfluous”. Justice Major made this point by adopting a passage from Sopinka’s The Law of Evidence in Canada (1992): 23 ... Rather, the effect of the application of res ipsa loquitur is as described in The Law of Evidence in Canada (1992), by John Sopinka, Sidney N. Lederman and Alan W. Bryant, at p. 81: Res ipsa loquitur, correctly understood, means that circumstantial evidence constitutes reasonable evidence of negligence. Accordingly, the plaintiff is able to overcome a motion for a non-suit and the trial judge is required to instruct the jury on the issue of negligence. The jury may, but need not, find negligence: a permissible fact inference. If, at the conclusion of the case, it would be equally reasonable to infer negligence or no negligence, the plaintiff will lose since he or she bears the legal burden on this issue. Under this construction, the maxim is superfluous. It can be treated simply as a case of circumstantial evidence. [64] It followed, according to Justice Major, that res ipsa loquitur should be discarded as “meaningless”: 26 ... Its use has been restricted to cases where the facts permitted an inference of negligence and there was no other reasonable explanation for the accident. Given its limited use it is somewhat meaningless to refer to that use as a doctrine of law. 27 It would appear that the law would be better served if the maxim was treated as expired and no longer used as a separate component in negligence actions. After all, it was nothing more than an attempt to deal with circumstantial evidence. The evidence is more sensibly dealt with by the trier of fact, who should weigh the circumstantial evidence with the direct evidence, if any, to determine whether the plaintiff has established on a balance of probabilities a prima facie case of negligence against the defendant. Once the plaintiff has done so, the defendant must present evidence negating that of the plaintiff or necessarily the plaintiff will succeed. [65] My interpretation of Fontaine is this. Res ipsa loquitur was just an epigram for the process by which circumstantial evidence is assessed by the trier of fact. That process stays the same, with or without the phrase. So the incantation of “res ipsa loquitur” was a “superfluous” and “meaningless” distraction, and the trial should just move directly to that factual process. As Justice Major said, the “evidence is more sensibly dealt with by the trier of fact”, without a Latin recital. The key point for Mrs. Johansson’s case is that Supreme Court of Canada did not tighten, to the plaintiff’s detriment, that underlying process by which circumstantial evidence is to be assessed by the trier of fact. Whatever inferences a jury could reasonably draw before Fontaine are no less reasonable afterward. [66] Before and after Fontaine, when the plaintiff’s evidence “permit[s] an inference of negligence attributable to the defendant”, the non-suit motion should be dismissed and the jury, after any evidence from the defendant, would decide whether or not that inference should be drawn. Before Fontaine, “res ipsa loquitur” was voiced en passant. Since Fontaine, it is not, but that same case remains jury bound. [67] This view is supported by the texts on Canadian negligence law published since Fontaine. [68] Lewis N. Klar, Q.C., Tort Law, 5th ed. (Toronto: Carswell-Thomson Reuters Canada Limited, 2012) at pp. 588-607, the most recent scholarly text to address the subject, reviews the history of res ipsa loquitur, then concludes (pp. 606-7): Aside from the instruction that the res ipsa loquitur is not a separate doctrine or rule that justifies having its own name, Major J.’s judgment in Fontaine v. British Columbia (Official Administrator) did not change the law with respect to res ipsa loquitur or prevent the continued use of circumstantial or indirect evidence in tort cases. Circumstantial or indirect evidence can, in the appropriate case, create an inference that the plaintiff’s injury was caused by the defendant’s negligence. The factors as discussed above such as 1) control; 2) the occurrence would not ordinarily happen without negligence; and 3) the absence of direct evidence, continue to be the relevant ones for courts to consider in deciding what weight is to be given to circumstantial evidence. The procedural effect of circumstantial evidence still remains the same. The advice given by Major J. was simply to remind courts that this is not to be applied in a formulistic way but simply as a matter of a common sense approach to evidence. [69] Similarly, Allen M. Linden and Bruce Feldthusen (Contributors), Halsbury’s Laws of Canada - Torts, under the topic Products Liability - Negligence Theory - Inferring negligence, para HTO-107, “Standard of proof” (QL), states: The Supreme Court of Canada has dispensed with the doctrine of res ipsa loquitur, describing it as “expired”. It is clear that the Latin expression has been abandoned in all situations; but it is also clear that proof of the facts that once led courts to infer negligence will continue to do so, but without the use of the Latin phrase. The process of proof in future products cases will not be markedly different than before, except that the Latin words will be absent from the discussion. [70] To similar effect: G.H.L. Fridman, Q.C., F.R.S.C., The Law of Torts in Canada, 3rd ed. (Toronto: Carswell-Thomson Reuters Canada Limited, 2010) at p. 387; Jamie Cassels and Craig Jones, The Law of Large-Scale Claims - Product Liability, Mass Torts, and Complex Litigation in Canada (Toronto: Irwin Law, 2005) at p. 25; Philip H. Osborne, The Law of Torts, 3rd ed. (Toronto: Irwin Law Inc., 2007) at p. 52; The Honourable Allen M. Linden and Bruce Feldthusen, Canadian Tort Law, 9th ed. (Markham, Ont: LexisNexis Canada Inc., 2011) at p. 256 (see quotation below, para 73). [71] Two appellate decisions illustrate the point. [72] In Marchuk (c.o.b. Gerry Marchuk Contracting) v. Swede Creek Contracting Ltd., [1998] B.C.J. No. 2851 (Q.L.) (C.A.), affirming [1997] B.C.J. No. 1927, the trial judge cited res ipsa loquitur and found that the defendant was negligent for causing a fire. The Supreme Court of Canada’s decision in Fontaine intervened before the appeal. The Court of Appeal dismissed the defendant’s appeal. Justice Mackenzie for the Court of Appeal said: 9 The Supreme Court of Canada has recently said that the Latin maxim res ipsa loquitur should be abandoned as confusing and unhelpful in cases involving circumstantial evidence of negligence: [citing Fontaine]. That case was decided after the judgment at trial in the case at bar. 10 While the Supreme Court was critical of the Latin maxim, the underlying principles governing the use of circumstantial evidence in determining liability for negligence were not modified. ... 12 It is a reasonable inference in the circumstances that activity would not cause a spark unless someone was careless. The effect of the trial judge’s non-finding on cutting and welding left it as one of the potential causes, although one not specifically proved. Accordingly, in my view, it was open to the trial judge to conclude, as he did, that the cause of the fire was a spark, and that the source of the spark was negligent activity by the employees of the defendant. [73] In Newfoundland Light and Power Co. v. Furlong Estate, [2005] N.J. No. 146, 2005 NLCA 25, (Q.L.), the homeowner died in a fire. His estate sued Newfoundland Power, alleging that an electrical malfunction had caused the fire. The trial judge held that Newfoundland Power was liable for negligence. Welsh, J.A. held that the judge was entitled to infer that the electrical malfunction was caused by the defendant’s negligence. She (para 44) quoted the following from Allen M. Linden, Canadian Tort Law, 7th ed. (Toronto: Butterworths, 2001) at p. 236: While the language of res ipsa is no longer to be used, the fact situations that gave rise to res ipsa in the past may offer some aid to courts about when negligence may be inferred. If, in the ordinary course of events, an accident would not occur in the absence of negligence, a court might infer negligence. ... The current edition, Linden and Feldthusen, Canadian Tort Law, 9th ed. (2011), (supra, para 70), p. 256, contains the same passage. In Newfoundland Light and Power, Justice Rowe, concurring, directly rejected the defendant’s submission that resembles the judge’s approach to GMC’s standard of care in Mrs. Johansson’s case: 78 Newfoundland Power emphasized the lack of evidence before the Trial Judge regarding Newfoundland Power’s inspection practices. Counsel for Newfoundland Power argued this lack of evidence was fatal to the Estate’s case. I cannot agree. ... 80 ... it was open to Newfoundland Power to seek to show that they had taken reasonable measures to avoid such an occurrence. Had they shown that, they could not have been found liable in negligence. However, they did not, nor did they seek to do so; in their view, it was for the Estate to prove that Newfoundland Power had failed to take such reasonable measures. 81 The implications of Newfoundland Power’s view of liability can most readily be understood with an example. The purchaser of a new car drives it off the dealer’s lot and at the first intersection is involved in a collision because the brakes fail. Once this is proven at trial, in order to make out his case, must the vehicle purchaser also lead evidence to show that the manufacturer had failed to take reasonable measures to avoid brake failure by its vehicles (e.g. in the design or production of the brakes)? 82 From Newfoundland Power’s perspective, the plaintiff in the foregoing example would have to do more than prove that a malfunction of the defendant’s equipment had caused an injury; rather, the plaintiff would also have to prove that this arose from a failure by the defendant to take reasonable measures to prevent such a malfunction. 83 This view may well have arisen from a misapprehension of the Supreme Court of Canada’s statement that res ipsa loquitur has “expired”. In Fontaine v. British Columbia (Official Administrator), supra, Major, J., for the Court, wrote at para. 27: [quotation of para 27 omitted - see above para 72] 84 The Trial Judge unwisely referred to res ipsa loquitur, but in my view did not rely on it; rather, in effect, he made a finding that the Estate had made out a prima facie case of negligence against Newfoundland Power. In the words of Major, J., in Fontaine, supra: Once the plaintiff has done so, the defendant must present evidence negating that of the plaintiff or necessarily the plaintiff will succeed. 85 That is what happened here. The brakes on new cars are not supposed to fail. And, the operation of an electrical utility’s equipment is not supposed to cause a fire in the dwelling of a customer. If, as here, a plaintiff has proven on the balance of probabilities that the electrical utility’s equipment has caused such a fire, then a prima facie case has been made out against the utility. ... [74] For those reasons, in my opinion the judge misinterpreted the effect of Fontaine. The judge thought the Supreme Court of Canada had altered the law, governing a jury’s inferences from circumstantial evidence, to make it “substantially more difficult” for a plaintiff to withstand a non-suit motion. The judge then burdened Mrs. Johansson’s reply to GMC’s non-suit motion with that mistaken view. That was an error of law which, in turn, led the judge to distinguish Farro based on a mistaken assumption. (b) The Plaintiff’s “Tactical Decision” to Sue in Tort [75] The judge said: [52] There are at least two ways that the plaintiff in this case might have overcome the hurdles posed by the expiry of res ipsa loquitur. Firstly, the plaintiff could have considered joining Maria Johansson’s husband, with whom the defendant communicated as purchaser, and who was the registered owner of the Lumina, as a plaintiff and claimed concurrently in tort and contract. Then later: [56] The plaintiff made a tactical decision to claim only in tort. The judge’s second way to overcome the burden posed by the expiry of res ipsa loquitur was that the plaintiff could conduct a more thorough discovery. Later (paras 93 ff) I will discuss this general topic. [76] With respect, I don’t understand how the judge’s comments about suing in contract pertain to Mrs. Johansson’s claim for her brain injury. She had no privity of contract with GMC, and no option to recover in contract. Mr. Johansson, who bought the car and had privity, was not involved in the accident. The non-suit of Mrs. Johansson’s personal injury claim should not have been characterized as a consequence of a mistaken tactical choice to sue in tort. (c) Inferring Negligence versus Strict Liability [77] Mrs. Johansson cited to the judge two authorities to the effect that the trier of fact could infer a standard of care and its breach from evidence surrounding the defect. One was Farro, discussed earlier. The other was Hollis v. Birch, [1989] B.C.J. No. 2261 (Q.L.) (B.C.S.C.). [78] The judge (para 43) declined to follow Hollis. He said: [42] Nonetheless, the plaintiff argued that the defendant’s recall of the 1997 Chevrolet Lumina, for safety reasons, is evidence of a breach of the standard of care in and of itself. The plaintiff relied principally on two cases in support of this proposition, Hollis v. Birch, [1989] BCJ No 2261 (QL) (BC SC) [Hollis] and Farro v. Nutone Electrical Ltd. (1990), 72 OR (2d) 637, [1990] OJ No 492 (QL) (CA) [Farro]. [43] In Hollis [Hollis v. Birch, [1989] B.C.J. No. 2261 (QL) (B.C.S.C.)]which was a product liability case dealing with the rupture of a silicone breast implant, the trial judge cited American case law for the proposition that negligence could be inferred from the existence of a product that did not function properly. In my view, this reasoning falls precisely within the definition of strict liability: the defendant’s liability flows immediately from the fact that the product was defective, and not from the actions or omissions that the defendant took in bringing the product to market. Strict liability is not the law in Canada. On this basis, I am not prepared to follow the decision in Hollis. [emphasis added] [79] GMC’s submission operates from the same perspective. GMC’s factum to the Court of Appeal says that the dismissal of its non-suit motion would impose “strict liability”. 89. The essence of the Appellant’s submissions on this point is that the fact of the Recall is evidence of breach of the standard of care. As noted above, that submission is akin to saying that there is strict liability when a manufacturer or distributor is involved in a recall and it is well-established that strict liability is not the law in Canada. [80] I don’t behold any hovering spectre of strict liability. The judge and GMC confuse liability without negligence with the jury’s function of deciding whether or not to reasonably infer negligence. [81] Drawing inferences is standard fare for juries. An inference is a finding deduced or induced from a premise without direct evidence of the inferred fact. It is a factual jump on the reasoning path. The judge ensures that the span is not so broad or irrational that a reasonable jury would stumble. Otherwise the system trusts the jury’s common sense and agility to mind the gap and land softly. To resolve the non-suit motion simply because there is no direct evidence of GMC’s standard of care for rack and pinion steering assemblies, is to emasculate the jury’s function of assessing whether or not to reasonably infer the standard’s particulars from appropriate evidence. In Grant v. Australian Knitting Mills Limited, [1936] A.C. 85 (H.L.), pp. 96 and 101, Lord Wright, in well known extracts, discussed the process of inference from circumstantial evidence in a products liability case: ... [The defendant’s counsel] contended that the appellant’s case involved arguing in a circle; his argument, he said, was that the garments must have caused the dermatitis because they contained excess sulphites, and must have contained excess sulphites because they caused the disease: but nought, he said, added to nought still is no more than nought. This, however, does not do justice either to the process of reasoning by way of probable inference which has to do so much in human affairs, or to the nature of circumstantial evidence in law courts. Mathematical, or strict logical, demonstration is generally impossible: juries are in practice told that they must act on such reasonable balance of probabilities as would suffice to determine a reasonable man to take a decision in the grave affairs of life. Pieces of evidence, each by itself insufficient, may together constitute a significant whole, and justify by their combined effect a conclusion. ... But when the position of the manufacturers is considered, different questions arise: there is no privity of contract between the appellant and the manufacturers: between them the liability, if any, must be in tort, and the gist of the cause of action is negligence. The facts set out in the foregoing show, in their Lordships’ judgment, negligence in manufacture. According to the evidence, the method of manufacture was correct: the danger of excess sulphites being left was recognized and was guarded against: the process was intended to be fool proof. If excess sulphites were left in the garment, that could only be because some one was at fault. The appellant is not required to lay his finger on the exact person in all the chain who was responsible, or to specify what he did wrong. Negligence is found as a matter of inference from the existence of the defects taken in connection with all the known circumstances: even if the manufacturers could by apt evidence have rebutted that inference they have not done so. [emphasis added] [82] In Mrs. Johansson’s case there is substantial evidence, and a prima facie finding by the judge, relating to the steering system’s defect when it left GMC’s possession. To itemize: (a) GMC’s Recall Notice said that “General Motors has decided that a defect, which relates to motor vehicle safety, exists” in 1997 Luminas. The defect is a “condition where the lower pinion bearing in the power steering gear may separate”. This meant that “[w]hen trying to turn left” some vehicles could “assist towards the right”, and “a crash could result”. (b) General Motors’ internal memo to its dealers explained the reason for the problem: Some of the lower pinion bearings were not manufactured properly and can separate with vehicle usage over time. As the driver steers the vehicle to the left, the pinion shaft moves and power steering fluid is misdirected, so the power assist is to the right instead of to the left. (c) GMC’s Mr. Davies agreed, in his transcript, that the 1997 Luminas contained “defective parts related to the power steering gear, lower pinion bearing”. (d) Mr. Davies testified the United States National Highway and Traffic Safety Administration (NHTSA) had requested General Motors to investigate the problem. General Motors’ Owners’ Manual for the Johanssons’ Lumina notified owners that the NHTSA was the authority to be notified if the vehicle is thought to have “a defect which could cause a crash or could cause injury or death”. (e) Mr. Davies testified: where there was total failure, meaning all the ball bearings in, and the cage were completely in a position where they weren’t of any use to hold the pinion, that the pinion would move, which allowed redirection of fluid. And when that occurred, if you tried to move the wheel to the left, the vehicle would actually move to the right. (f) The DVD evidence of General Motors’ testing showed, according to the judge’s mid-trial ruling on admissibility, “that the defect occurs in practice”, making the footage “probative”. (g) Dr. Smith, qualified as an expert, commented on the DVD evidence that the retainer that holds the bearings “has completely come loose”, the “balls are now no longer evenly spaced around the bearing”, later “[t]he ball bearings were coming out of the bearing assembly” and “if they were in a car, the balls would end up falling into that cup at the bottom of the steering shaft”. Dr. Smith testified that the bearing “was prone to failure”, meaning the “retaining ring or clip would fall out”, then “the ball bearings themselves would fall out”, then “the steering shaft would then not be properly supported by the bearing, and would be able to shift and flop and move around”. The “wobbling” shaft meant “erratic steering”, because that “instead of just being kept (centered in line), this can flop around and, in a somewhat unpredictable way, cause this control valve to move in a somewhat unpredictable way and cause the hydraulic fluid to go into the rack in a somewhat unpredictable way, so the car doesn’t go where you try to steer it”. (h) The defect was avoidable. According to GMC’s documents, other 1996-98 General Motors’ vehicles did not have the defect, and the recalled vehicles were remediated by a standard repair protocol. [83] Each case turns on its own evidence. Nonetheless, it is instructive to overlay Mrs. Johansson’s circumstances onto the background of other cases and textual authorities that discuss whether inferences reasonably may be drawn from an equivalent evidentiary premise. [84] Farro, Marchuk and Newfoundland Light and Power are examples where three appellate courts acknowledged that similar inferences were available to establish the standard of care and its breach. To this list may be added Chief Justice MacKeigan’s comments in Smith v. Inglis Limited (1978), 25 N.S.R. (2d) 38 (S.C.A.D.), para 8, that were adopted by the Ontario Court of Appeal in Farro, para 19 (quoted above, para 52). Chief Justice MacKeigan (para 9) noted that he was not invoking res ipsa loquitur, then (paras 9 ff) elaborated on the inference of negligence from evidence surrounding a defect that existed in the product when it left the manufacturer. [85] Klar, Tort Law (5th ed. 2012), p. 382, states: As stated by Waddams, and agreed to by the courts, where a product has been manufactured with a defect, and this defect has resulted in the plaintiff’s injuries, “the inference of negligence is practically irresistible”. This inference is predicated upon proof that the defect was in the product when it left the manufacturer. To the same effect: S. M. Waddams, Products Liability, 5th ed. (Toronto: Thomson Reuters Canada Limited, 2011) at pp. 69-70; Linden and Feldthusen, Canadian Tort Law, (9th ed. 2011), pp. 632-33; Cassels and Jones, The Law of Large-Scale Claims, (2005), p. 34; Dean F. Edgell, Product Liability Law in Canada (Markham, Ont.: Butterworths, 2000) at pp 25-26; Lewis W. Klar, Allen M. Linden et al., Remedies in Tort, vol. 3, (Carswell, 2011), para 45, pp. 20-37 and 20-38; and the case authorities cited in these passages. These texts post-date Fontaine. [86] I will return to the judge’s concern about strict liability. Strict liability is defined by Linden and Feldthusen, Canadian Tort Law (9th ed. 2011), p. 539 as: One person may be required to compensate another for injury or damages, even though the loss was neither intentionally nor negligently inflicted. [87] GMC will be liable to Mrs. Johansson only if GMC is negligent - i.e. if, once all the evidence is in, the jury chooses to draw the inferences necessary to find that GMC has breached its standard of care. Strict liability is not in play. [88] As noted by the trial judge, in Hollis v. Birch the judge cited an American case. Justice Bouck said that “[i]n the course of my research, I ran across an American case of striking similarity”. But that alone doesn’t taint the Hollis decision. It is necessary to examine the court’s reasons in Hollis. Justice Bouck dismissed the defendant’s non-suit motion because: I believe there is some evidence before me from which negligence may be inferred on the part of Dow Corning Corporation. ... While this is some evidence from which negligence may be reasonable be inferred, it is still necessary for the plaintiff to show that it is evidence from which negligence ought to be inferred. Metropolitan Railway Co. v. Jackson, (1877), 3 App. Cas. 193 at 197 [H.L.]. Put another way, if I were sitting with a jury I would find there is some evidence which might reasonably satisfy the jury the defendant was negligent. Whether the jury would in fact find such negligence is another matter. This passage merely applies Lord Cairns’ principle (above para 28) on jury inferences. It neither articulates nor applies strict liability. [89] In my respectful view, the judge mistook the process by which a jury may choose whether or not to infer “negligence” from circumstantial evidence, for “strict liability” without negligence. That was an error of law. (d) The Need for Evidence of Regulatory Benchmarks or Industry Practice [90] I’ll summarize the judge’s reasoning on this topic. [91] The judge (para 33) quoted the following from Crits v. Sylvester (1956), 1 D.L.R. (2d) 502 (O.C.A.), p. 508: “Every medical practitioner must bring to his task a reasonable degree of skill and knowledge and must exercise a reasonable degree of care” and “is bound to exercise that degree of care and skill which could reasonably be expected of a normal, prudent practitioner of the same experience and standing, and if he holds himself out as a specialist, a higher degree of skill is required of him than of one who does not profess to be so qualified by special training and ability.” [92] Immediately after this quotation, the judge, in Mrs. Johansson’s case, ruled that evidence of industry practice was mandatory for a plaintiff in a products liability claim: [33] ... From this legal standard, the trier-of-fact must then hear evidence on what a “prudent practitioner” would mean in the particular circumstances, and assess whether the defendant met that standard. [34] In this case, the plaintiff presented no evidence addressing the standard industry practice of an automobile manufacturer, assembler or distributor with respect to parts supplied by other parties. Such evidence might include answers to the following questions: Is it industry practice for a manufacturer, assembler or distributor to test such parts? If so, what kind of testing is normally performed? Does the testing depend on the type of part and the level of risk posed by a failure of the part? Would a standard manufacturer, assembler or distributor be expected to discover an improperly-crimped lower pinion bearing, such as the alleged defect in this case? [93] The next question for the judge was - What type of evidence of industry practice is required of a plaintiff? The judge gave several answers. At one point (para 52), he said that to “overcome the hurdles posed by the expiry of res ipsa loquitur ... the plaintiff could have conducted a more thorough discovery of the defendant”. Presumably this would not be expert evidence. But earlier (paras 35-37) the judge said that Mrs. Johansson needed to adduce either evidence of “a regulatory benchmark” or expert evidence. [94] As to the regulatory benchmark, the judge said: [35] ... The Motor Vehicle Safety Act creates safety standards that all Canadian automobile manufacturers must satisfy. Such standards provide a regulatory benchmark of what is expected of normal manufacturers, assemblers or distributors. Nonetheless, the plaintiff provided no evidence of what the Act required of the defendant, or of how the defendant performed in the circumstances of the case. The judge was referring to the Motor Vehicle Safety Act, S.C. 1993, c. 16. The Act authorizes subordinate legislation dealing with safety issues, and in s. 17 creates an offence for violating the Act or its Regulations. [95] The judge continued (para 36) to say that failure to lead evidence of non-compliance with a regulatory benchmark “is not always fatal to a plaintiff’s claim” because the plaintiff could opt to lead expert evidence. [96] The judge said (para 36) that expert evidence is needed unless “the matters are non-technical” or where the defendant’s conduct is “egregious”. He stated (para 37) that “[t]he defendant’s manufacture, assembly and/or distribution of the 1997 Chevrolet Lumina is not of a non-technical nature” and “no evidence was presented to suggest that the defendant’s conduct was egregious”. [97] From the foregoing analysis, the judge concluded: [38] Without this type of evidence, a jury, properly instructed on the law, would not be able to translate or apply the standard of care to the circumstances of the case; a jury would not be able to determine what “reasonable care” means in the circumstances of a Canadian automobile manufacturer, assembler or distributor. As such, a jury would not be able to determine whether the defendant’s conduct breached that relevant standard; this is determinative of the defendant’s nonsuit motion. [98] In my respectful view, the judge erred in four respects. First, evidence of non-compliance with industry practice or regulatory standards of course might be persuasive in closing argument, and probative in the jury room at the end of the day. But these factors are not stand alone prerequisites, required by law as an element of the plaintiff’s prima facie case in mid-trial. In my respectful view, the judge failed to appreciate that the jury is permitted to consider whether or not to draw an inference from the other evidence, such as the evidence surrounding the defect. Second, the judge misapprehended the evidence respecting the regulatory benchmarks. Third, the judge transferred to a products liability action a principle that has particular significance in professional liability professional negligence claims. Fourth, the judge failed to appreciate that the technical aspects of the steering failure and defect were supported by evidence, including expert evidence. I will address these in turn. [99] First: The scope of jury inferences. On the permissible inference from evidence surrounding the defect, I reiterate the numerous authorities that are cited earlier in these reasons (paras 52, 62-73, 81, 84-85). [100] In Ryan v. Victoria (City), [1999] 1 S.C.R. 201, Justice Major for the Court defined the standard of care: 2. Standard of Care 28 Conduct is negligent if it creates an objectively unreasonable risk of harm. To avoid liability, a person must exercise the standard of care that would be expected of an ordinary, reasonable and prudent person in the same circumstances. To the same effect: Fridman, The Law of Torts in Canada (3rd ed. 2010), pp 363-4. In Mustapha v. Culligan of Canada Ltd., [2008] 2 S.C.R. 114, Chief Justice McLachlin for the Court said, more tersely: [7] The second question in a negligence action is whether the defendant’s behaviour breached the standard of care. A defendant’s conduct is negligent if it creates an unreasonable risk of harm (Linden and Feldthusen, at p. 130). The page reference in the current edition is Linden and Feldthusen, Canadian Tort Law (9th ed. 2011), p. 134. [101] The judge, in Mrs. Johansson’s case determined: [28] As an assembler or as a distributer [sic], I am satisfied that the defendant owed the plaintiff a duty of care to take all reasonable precautions in the assembly or distribution of the Lumina. ... [32] In the context of product liability, it is clear that the trial judge is responsible for informing the trier-of-fact that the applicable legal standard of care on the defendant was “to use reasonable care in the circumstances and nothing more” (Phillips at para 49). The judge referred to Phillips v. Ford Motor Co. of Canada, 1971 CarswellOnt 657, [1971] 2 O.R. 637 (C.A.), para 49; ... our Courts do not, in product liability cases, impose upon manufacturers, distributors or repairers, as is done in some of the States of the American union, what is virtually strict liability. The standard of care exacted of them under our law is the duty to use reasonable care in the circumstances and nothing more. [102] GMC’s factum (para 28) cites these passages and says that GMC “takes no issue with that formulation”. [103] Clearly Mrs. Johansson established that legal standard of care, and the judge so determined in his para 28. [104] The contentious issue relates to what the judge described as the “translation or application of that standard to the facts”. In this respect, the judge said: In my view, it is irrelevant whether the translation or application of that standard to the facts is characterized during trial as a question of mixed fact and law or as a question of fact, because in either case, the responsibility for this translation falls on the trier-of-fact. The trial judge may instruct the trier-of-fact on the appropriate factors that may be considered - for example, the defendant’s expertise, the riskiness of the product, the defendant’s knowledge of that risk, industry standards, et cetera - but it is ultimately for the trier-of-fact to apply the standard of care in the circumstances and determine whether the defendant’s conduct breached that standard. [105] I agree with this passage. But it follows from those principles that it is for the jury to balance those factors and weigh the evidence related to the application of the standard of care. It is for the jury to decide whether in the circumstances of a particular case evidence of industry practice or regulatory benchmarks, on the one hand, or evidence surrounding the defect, on the other, should tip the scale. It is not for the judge to wade into the factual waters with a finding that a particular factor, such as industry practice, on which the judge has heard no evidence, will outweigh the others. [106] In Ryan, after defining the standard of care as quoted above, Justice Major continued to describe the factors that particularize the application of the standard of care: 28. ... The measure of what is reasonable depends on the facts of each case, including the likelihood of a known or foreseeable harm, the gravity of that harm, and the burden or cost which would be incurred to prevent the injury. In addition, one may look to external indicators of reasonable conduct, such as custom, industry practice, and statutory or regulatory standards. [emphasis added] [107] Justice Major said one “may look” to industry practice and regulatory standards as “additional” factors. They are permissible indicators. They are relevant, usually probative and, at the end of the day, may be conclusive. But they are not a legal sine qua non for the plaintiff to escape a non-suit, and a plaintiff may establish her case with other evidence. [108] Fridman, The Law of Torts in Canada (3rd ed. 2010), supra, pp. 363-64, notes the factors cited in Ryan, then examines them in turn. Respecting “Customary Practices”, Fridman states (pp. 378-80): 7. Customary Practices A defendant accused of having fallen short of the standard of care may be able to rely on the fact that his or her conduct was in accordance with customary, general and approved practice in the profession, trade or business in which he or she was engaged. ... Evidence of the practice may be given, and if the practice is established then there is a presumption that adherence to it is sufficient to show that there was no negligence. Nevertheless, it is not for expert witnesses on such practices to determine the issue of negligence. The modern approach is to give considerable weight to compliance with customary practice in areas involving complex, scientific or technical matters. In ter Neuzen v. Korn [(1995), 127 D.L.R. (4th) 577 (S.C.C.)] the Supreme Court of Canada held that the trier of fact could not find a doctor to have breached the standard of care if the doctor complied with the customary practice. The only exception, discussed below, is in cases where an ordinary person without particular expertise could conclude that the practice itself was negligent. There must be adequate evidence that the practice in question actually exists and is observed. Hence, in Lepine v. University Hospital Board [(1964), 50 W.W.R. 709 (Alta. T.D.), aff’d (1965), 54 D.L.R. (2d) 340 (Alta. C.A.), rev’d on other grounds [1966] S.C.R. 561] the defendants could not rely on any such practice to excuse them from liability when the plaintiff, a patient, threw himself out of a window. There was no established practice relating to the treatment of mentally ill patients, as opposed to epileptics and those suffering from automatism, that demonstrated the hospital was acting reasonably in failing to take precautions to prevent what happened. ... Common or customary practice will not operate as a defence to negligence where the activity concerned involves an element of danger that is reasonably foreseeable, such as the conduct of a chemistry experiment in a school. Nor will reliance on such a practice be enough if the defendant’s skill and training should lead him or her to expect that other precautions ought to be taken to prevent the possibility of harm. Moreover, a practice that is careless will not excuse the defendant. No amount of repetition of a careless practice will make it any less careless. As Sopinka J. said in ter Neuzen v. Korn, where the common practice is fraught with danger, a judge or jury may find that the practice is itself negligence. Hence, if the defendant relies on custom or approved practice to defeat an allegation of negligence, the plaintiff will be able to attempt to prove that the practice was one which no reasonable person would follow. [109] Klar, Tort Law (5th ed. 2012), supra, pp. 355-59 also discusses these factors, then (pp. 363-5) comments on industry practice: 5. GENERAL PRACTICE A very useful factor in determining whether a person’s conduct was reasonable or not is the general practice of those engaged in a similar activity. ... Evidence of general practice will be given more or less weight depending on the circumstances. There are no specific requirements to establish a generally approved practice. ... The burden of proving that a custom exists is on the person alleging, and “only in the rarest and most patently obvious cases will the Courts take judicial notice of a custom” [quoting Iacobucci, J. in Waldick v. Malcolm (1991), 8 C.C.L.T. (2d) 1 (S.C.C.), at p. 15]. Under the previous, but now discredited, theory that evidence of general and approved practice was conclusive proof that the defendant acted reasonably, it was of course important to have a fixed idea of what constituted general practice. Under the current, more realistic, approach that what others do is only a relevant factor in deciding what is reasonable, the weight of which varies depending on the circumstances, it is unnecessary to attempt to define the elements of a general practice with precision. ... It is almost universally conceded that evidence of general practice can never, as a matter of law, settle the negligence issue. Thus whether the evidence is that of the plaintiff, i.e., that the defendant failed to adhere to general practice, or of the defendant, i.e., that the defendant was adhering to general practice, the evidentiary effect is the same. This may raise an inference of negligence, or of reasonable care, or may in fact raise a prima facie case, but will not be conclusive proof. [110] From these textual summaries, it is clear that a plaintiff may lead evidence that the defendant failed to comply with industry practice, and that evidence may assist to establish a prima facie case. Or the defendant may lead evidence of compliance with industry practice. Either way, the evidence is to be weighed, with other evidence, by the jury at the conclusion of the trial. But evidence of industry practice does not necessarily trump other evidence and settle the negligence issue in the jury room. So it cannot be a legal prerequisite to the plaintiff’s prima facie case in mid-trial. A products liability plaintiff may establish her prima facie case with other evidence. The more common sequence is that the plaintiff leads with other evidence, and the evidence of industry practice makes its debut in the defendant’s case. [111] Second: The evidence of “regulatory benchmarks”. The judge (para 35) said there was no evidence “of how the defendant performed” against the “regulatory benchmark” of the Motor Vehicle Safety Act of Canada. This is not a private prosecution for an offence under s. 17 of the Motor Vehicle Safety Act. It is a negligence action where the trier of fact is to assess the reasonableness of GMC’s conduct. GMC’s Mr. Davies testified there was no Canadian investigation of the steering defect, because that investigation was underway in the United States. He said “[t]he General Motors Corporation undertook the investigation at the request of the National Highway and Safety Administration, and it would be redundant for GM of Canada to do exactly the same thing” (quoted above, para 43). This United States Government Association is the regulatory body that, according to GMC’s Safety Manual for the Johanssons’ Lumina, is to be notified of “a defect which could cause a crash or could cause injury or death” (quoted above, para 48). So there was evidence that the steering defect incited action - the request for an investigation - by a regulatory authority. What weight this evidence generates is for the jury. But Mrs. Johansson should not be non-suited either because evidence was lacking of a non-existent and “redundant” Canadian investigation, or because any regulations under the Canadian statute may not specifically address rack and pinion steering. [112] Third: The need for expert evidence. The judge, in Mrs. Johansson’s case, referred to Crits, supra, and Krawchuk v. Scherbak, 2011 ONCA 352 as authority that expert evidence is required to establish the standard of care. Crits and Krawchuk discussed professional negligence claims. Neither involved a non-suit. The judge (see above, paras 91-2) moved seamlessly from professional negligence to products liability, and ruled that, absent evidence from Mrs. Johansson on industry practice respecting rack and pinion steering, a non-suit was in order. [113] The leading authority on the need for expert evidence in professional negligence cases is ter Neuzen v. Korn, [1995] 3 S.C.R. 674, a negligence claim against an obstetrician and gynaecologist. The plaintiff contracted HIV from the defendant’s procedure for artificial insemination. The procedure occurred in the early 1980s, before the first published case of HIV transmission from artificial insemination. There was expert evidence that the defendant’s approach complied with gynaecological practices in Canada at the time. The jury nonetheless found the defendant negligent. There was no non-suit. The Court of Appeal overturned the verdict and ordered a new trial. The Supreme Court of Canada dismissed the appeal. The case turned on what knowledge of HIV risk was available in the profession at the time of the procedure. Justice Sopinka said: 44 As was observed in Lapointe, courts should not involve themselves in resolving scientific disputes which require the expertise of the profession. Courts and juries do not have the necessary expertise to assess technical matters relating to the diagnosis or treatment of patients. Where a common and accepted course of conduct is adopted based on the specialized and technical expertise of professionals, it is unsatisfactory for a finder of fact to conclude that such a standard was inherently negligent. On the other hand, matters falling within the ordinary common sense of juries can be judged to be negligent. For example, where there are obvious existing alternatives which any reasonable person would utilize in order to avoid a risk, one could conclude that the failure to adopt such measures is negligent notwithstanding that it is the prevailing practice among practitioners in the area. [114] I disagree with the judge that the principles respecting expert evidence in professional negligence cases support a non-suit of Mrs. Johansson’s claim. [115] The standard of care for each professional is defined with reference to his or her profession. Linden and Feldthusen, Canadian Tort Law, (9th ed. 2011), p 165, under the heading “Professional Negligence”, states: In all these cases, the courts are balancing the interests of the clients or patients in receiving skilled service as well as the interests of professional people in a certain degree of autonomy in their dealings with the community. As always, an uneasy compromise has been reached. Every recognized professional group has its own individual standard - a standard to which all the members of the profession must conform. Chief Justice McLachlin, for example, has asserted that, in conducting an investigation, the standard of care required of the police is that of a “reasonable police officer in all the circumstances”. Surveyors, for an example, must “exercise a reasonable amount of care and a reasonably competent degree of skill and knowledge”. A chiropractor’s diagnosis, for another example, must be “sufficient by chiropractic standards”. A similar obligation, which is said by some to be based on contract, rests on engineers, architects, accountants and others. It follows that evidence related to the standards of the defendant’s particular profession is in order. In ter Neuzen v. Korn, for example, the case pivoted on the awareness, in the gynaecological profession, of HIV risk in the early 1980s, a topic tailored for expert evidence. [116] The same cannot be said in this case. GMC’s standard is to use reasonable care in the circumstances. From Justice Major’s comments in Ryan, those circumstances may, but do not necessarily, include industry standards. Expert evidence of industry standards would be relevant, but is not legally mandatory for a claim such as Mrs. Johansson’s. [117] Appellate courts have recognized this point. [118] G & S Haulage Ltd. v. Park Place Centre Ltd., [2011] N.S.J. 142, 2011 NSCA 29 involved an appeal from a ruling by a judge without a jury, after trial, that apportioned negligence and contributory negligence respecting an oil leak on a commercial property. Justice Beveridge said: [103] Park Place submits legal error occurred because there was no specific evidence of the appropriate standard of care for an ordinary, reasonable and prudent operator of a hotel and office complex. ... [105] The presence of expert evidence in an action against a professional such as a physician is the norm. Conduct that accords with well recognized and acceptable professional practice will usually, but not necessarily, defeat an allegation of negligence (see ter Neuzen v. Korn, [1995] 3 S.C.R. 674). I see no basis to elevate an operator of a hotel and commercial complex to the level of a professional exercising his or her skill in a specialized or technical area. There was nothing technical or specialized about the conduct in this case. Justice Beveridge (paras 106-7) cited with approval the companion cases of Burbank v. Bolton, [2007] B.C.J. No. 752, leave denied [2007] S.C.C.A. No. 316, and Radke v. M.S. (Litigation guardian of), 2007 BCCA 216, and in particular the following from Justice Lowry’s decision in Burbank: [57] It is first important to recognize that in a negligence action it is not usually necessary to adduce evidence, much less expert evidence, to prove the standard of care. It is generally a matter to be determined by the trier of fact based on common experience having due regard for what may be taken from any applicable legislation or policies governing the activity in question; in some instances, evidence of custom associated with any particular conduct may also be germane. It is only where the subject matter of the inquiry is beyond the common understanding of judge and jury that expert evidence may be adduced to assist the court in determining the appropriate standard of care. [119] In Holland Carriers Ltd.v. MacDonald, 2012 NSCA 47, a part dislodged from a moving truck, injuring the plaintiff. The trial judge, without a jury, allowed the action for negligence against the truck’s owner/operator. On appeal, the appellant/defendant submitted that, absent evidence respecting an industry standard for maintenance and inspection of a truck, negligence had not been proven. The Court of Appeal rejected that submission. Chief Justice MacDonald (paras 16-21) adopted Justice Beveridge’s comments from G & S Haulage, and held that the trial judge was entitled to apply common knowledge and experience to apply the legal standard of reasonable care. [120] In Moreau v. Roman Catholic Bishop of Edmundston, [2011] N.B.J. No. 85, another negligence case where the defendant was not a professional, Chief Justice Drapeau for the Court said: 17 ... Although expert testimony on the accepted norms may be beneficial in many instances, it is essential only where the context is so esoteric that a person with ordinary judgment and experience could not form a valid opinion on the reasonableness of the conduct in issue (see Beshara v. Dysart (1998), 207 N.B.R. (2d) 14, [1998] N.B.J. No. 480 (N.B. C.A.)). ... [121] Fourth: Is this a “technical” issue? The judge observed (para 37) that expert evidence was required of Mrs. Johansson because: The defendant’s manufacture, assembly and/or distribution of the 1997 Chevrolet Lumina is not of a non-technical nature. [122] The technical aspects of the manufacture and assembly of the Lumina’s defective steering component were fully explained during the plaintiff’s evidence. That evidence includes (1) the expert report and testimony of Dr. Smith, (2) components of the rack and pinion assembly entered as demonstrative evidence, and discussed by Dr. Smith’s testimony, (3) the testimony of Mr. Davies, GMC’s Canadian Manager of Field Performance Assessment, (4) the DVDs displaying the technical investigation by General Motors US, and (5) the documents sourced from GMC or General Motors US that explain the technicalities of the defect, its reasons, how it occurs in practice and how to repair it. I have extracted this evidence earlier. This evidence was sufficiently comprehensive that GMC’s counsel told the Court of Appeal: The fact is this: We know exactly how this defect arose. It’s in evidence ... Then after elaborating, with the assistance of the demonstrative exhibits, GMC’s counsel stated: So there’s no need for this Plaintiff to get inside the factory, and figure out who did what and what might have gone wrong. [123] What remains, then, is whether a reasonable person in GMC’s circumstances would have taken precautions, for instance respecting inspection, testing or some other aspect of quality control, to identify and rectify such a defect before the Lumina retailed to the Johanssons. That is not esoteric “technical” territory. It is a topic well within the comprehension and appropriate function of a jury in a negligence trial. [124] Conclusion on the Need for Evidence: For the above reasons, in my view it is not essential that Mrs. Johansson’s case contain evidence of regulatory benchmarks or expert evidence of industry practice, provided that the evidence she has adduced is sufficient to permit a jury to infer negligence. Whether her evidence is sufficient in that regard is the point I will address next. (e) Could a Jury Reasonably Infer Negligence in this Case? [125] Instead of dwelling on res ipsa loquitur, strict liability and the absence of expert evidence, the judge should have just determined whether GMC’s negligence could be reasonably inferred from the evidence that was adduced. That was his task at hand under the non-suit test set out earlier (paras 24-31). This task entails a review of the evidence, and an analysis of what inferences may be reasonably inferred from that evidence. I have summarized that evidence (paras 32-48). The judge’s decision neither reviewed that evidence, other than reciting the facts of the accident and Recall Notice, nor analysed what inferences may reasonably be drawn from that evidence. In my respectful view, these fundamental omissions were an error of law. [126] Turning to that evidence, could a jury reasonably infer GMC’s breach of the standard of care? [127] I refer to the key aspects of the evidence summarized earlier (para 82). In my opinion, from that evidence, a jury reasonably could, if it wishes, infer that the low end of the standard of care governing quality control, investigation or testing for an assembler of a new vehicle does not sanction: a “defect” that impugns “safety” where “a crash could result”, because steering veers opposite to the intended direction, caused by components that are “not manufactured properly”, so bearings would suffer “failure” and fall out making them useless for their purpose, leading to “misdirected” power steering fluid, and a “wobbling” steering shaft that can “flop around” unpredictably, causing “erratic steering”. The quoted words are extracted from the evidence set out earlier. It seems obvious that when the driver steers left, the car shouldn’t turn right. This is inferable from the meaning of “steering”, apart from the rest of the evidence in this record. That a jury might infer a standard of quality control to govern this circumstance is neither impermissible nor unreasonable. That the defective rack and pinion steering system was manufactured by another General Motors US subsidiary, then assembled by GMC does not alter my view of how, on this evidence, a jury may be permitted to apply the assembler’s standard of reasonable care by inferring particulars governing quality control, investigation or testing. [128] Then there is GMC’s Chevrolet Lumina 1997 Sales Brochure, which said: For countless families across Canada, choosing a new car comes down to a single word. Trust. They need a car they can depend on, day after day and month after month, in bad weather and in good. A car they can sometimes take for granted, because other things - kids, work, commitments - have to take priority. A car they can count on to protect. For these people - and perhaps for you as well - the new Chevrolet Lumina is everything it should be. Tried, tested and true - and a car you can have confidence in every day. The Brochure recited the Lumina’s “full complement of safety features”, “refined power-assisted rack-and-pinion steering” and said the “Lumina is engineered as a Total Safety System from one end to the other”. This evidence was adduced for the jury. That evidence may reflect on the reasonably foreseeable risk, its gravity and the cost of avoiding risk that GMC attributed to the Lumina, all factors that affect the standard of care under Ryan. Then, as to the breach, the jury might ponder how the “refined” rack and pinion steering took the car in a direction opposite to the turn of the steering wheel. Or the jury might consider how the steering system that was “[t]ried, tested and true” in 1997 left GMC’s assembly plant with the steering defect that General Motors US’ testing identified in 2002, leading to the recall. [129] I am not saying the jury should or would draw any of these inferences. But, given the raw material in this record of evidence, neither can I say the inferences are impermissible. That means, returning to the non-suit motion, it is for the jury to decide whether or not to draw any such inferences that assist the jury’s application of the standard of reasonable care and then assess its breach. [130] In my view, Mrs. Johansson is entitled to a jury verdict. 7. Analysis of Secondary Issue - The Roadside Reports [131] Mrs. Johansson submits that the judge erred by not considering, in his non-suit analysis, some 121 pages of Roadside Reports that record approximately one thousand reports of steering failures. [132] The Roadside Reports are records of complaints to General Motors from the driving public respecting the steering system. In response to an objection by GMC, the judge made a mid-trial ruling on their admissibility. He said they were relevant, authentic and in the possession of the defendant, but they were hearsay. The judge determined that they were not admissible as business records for the truth of their contents, but they were admissible to show that “the defendant received complaints regarding steering failures, and these complaints form part of the investigation and the basis for the recall”. [133] After the judge’s ruling, the Roadside Reports were not tendered as Exhibits or entered into evidence. It appears that the plaintiff intended to use the Roadside Reports to cross examine GM Canada’s witnesses later in the trial. The submissions on the non-suit motion included the following exchange between the judge and Mrs. Johansson’s counsel: MS. SNOW: They go back to 1996, complaints related to this vehicle. And, again, that’s information that I expect will be brought out through the defendant. And so whether they knew ... I suppose they did know that there were problems with the steering prior to this accident occurring. THE COURT: I haven’t seen those reports. I haven’t ... they haven’t been ... they’re not in evidence yet. MS. SNOW: They’re not yet. THE COURT: I ruled that they would be admissible ... MS. SNOW: Yes. THE COURT: ... but not for the truth of their contents. They’d be admissible ... MS. SNOW: Yes. THE COURT: ... not as hearsay documents, but as documents which were compiled based on information ... MS. SNOW: Yes, and ... THE COURT: ... but I haven’t seen ... but they’re not in evidence yet. MS. SNOW: They’re not in evidence yet. That’s right. [134] The Roadside Reports were not in evidence, even for the limited purpose that the judge had sanctioned in his mid-trial ruling. The judge did not err in law by declining to consider documents that were not in evidence. I would dismiss this ground of appeal. [135] The Roadside Reports have played no role in my consideration whether the judge erred by granting GMC’s motion for a non-suit. 8. Conclusion [136] I would allow the appeal, overturn the judge’s dismissal of Mrs. Johansson’s claim and dismiss GMC’s motion for a non-suit. [137] GMC urges that, if the appeal is allowed, the trial should resume with the defendant’s case before Justice Murphy without a jury. GMC’s counsel submits that it would be unfair that the plaintiff’s witnesses have a preview of their cross-examination questions. I respectfully disagree. A defendant’s failed motion for a non-suit should not deprive a plaintiff of her right to a jury trial. GMC will enjoy a preview of the answers. I would remit the matter to the Supreme Court for a new trial with a new jury. [138] The trial judge ordered Mrs. Johansson to pay GMC costs of $65,502.58, including $17,752.58 disbursements. I would vitiate this order and direct that any amount paid in compliance be returned to Mrs. Johansson. Regarding costs on appeal, I note that the scope of the trial judge’s award included pre-trial and trial effort beyond the non-suit motion. So the normal 40% guideline for appeal costs does not apply. Nonetheless, the issues on the appeal required a detailed analysis of the substantial record of evidence and the legal principles of negligence. With this in mind, I would order GMC forthwith to pay the appellants $8,000 plus reasonable appeal disbursements as costs for the appeal, in any event of the eventual cause after the new trial. The entitlement and quantum of costs of the first trial will be for the consideration of the trial judge after the verdict in the new jury trial. Fichaud, J.A. Concurred: MacDonald, C.J.N.S. Saunders, J.A.