Ding v. Canam Super Vacation Inc.

2024 BCCA 102 Ding v. Canam Super Vacation Inc.

COURT OF APPEAL FOR BRITISH COLUMBIA Citation: Ding v. Canam Super Vacation Inc., 2024 BCCA 102 Date: 20240314 Docket: CA48150 Between: Jie Ding Appellant (Plaintiff) And Canam Super Vacation Inc. d.b.a. Super Vacation, Prévost, a Division of Volvo Group Canada Inc., Universal Coach Line Ltd. Brian Spittal, Western Bus Lines Ltd., Mark Yu, Laurels Tak Lau, and Paul Tao Way Chan Respondents (Defendants) Before: The Honourable Chief Justice Marchand The Honourable Madam Justice Fisher The Honourable Mr. Justice Skolrood On appeal from: Orders of the Supreme Court of British Columbia, dated February 10, 2022 and April 3, 2023 (Ding v. Prévost, A Division of Volvo Group Canada Inc., 2022 BCSC 215 and 2023 BCSC 518, Vancouver Dockets M166513 and M166164). Counsel for the Appellant: C. Dennis, K.C. G. Rincon E. Aitken Counsel for the Respondent, Canam Super Vacation Inc. d.b.a. Super Vacation: T. Davies J.E. McGregor Counsel for the Respondent, Prévost, a Division of Volvo Group Canada Inc.: G.P. Brown, K.C. N.L. Trevethan Counsel for the Respondent, Universal Coach Line Ltd. (appeared on Nov 1): H.J.S. Harris Counsel for the Respondents, Brian Spittal and Western Bus Lines: A. Mersey, K.C. E.J. Segal Place and Dates of Hearing: Vancouver, British Columbia November 1-2, 2023 February 12, 2024 Place and Date of Judgment: Vancouver, British Columbia March 14, 2024 Written Reasons

by: The Honourable Madam Justice Fisher Concurred in by: The Honourable Chief Justice Marchand The Honourable Justice Skolrood Summary: This is an appeal from (1) the dismissal of the appellant's claims for negligent design and failure to warn, and (2) an order for costs. The appellant's claims arose from a single vehicle accident involving a tour bus that was not equipped with passenger seatbelts. The bus crashed and rolled on its side. The appellant, who was riding as a passenger, was partially ejected through a side window and seriously injured. The driver of the bus and the company operating it admitted liability. The trial proceeded against the bus manufacturer, the tour operator, another bus company who sub-contracted the bus charter, and the tour guide. After a 72-day trial, the trial judge dismissed the action. He found the bus manufacturer was not negligent in designing and manufacturing the bus in 1998 without seatbelts and he dismissed the action for failure to warn against the bus manufacturer and the tour operator on the basis that causation had not been proven against either defendant. The judge awarded the appellant only her costs to trial, ordered her and the two defendants who had admitted liability to bear their own costs of the trial, refused to make a Sanderson or Bullock order requiring the unsuccessful defendants to pay the costs of the successful

defendants, and awarded the bus manufacturer uplift costs. The appellant asserts various legal and factual errors in the trial judge's analysis of the claim against the bus manufacturer for negligent design and against the manufacturer and tour operator for failure to warn. She also asserts three errors in principle in the judge's costs award. Held: Substantive appeal dismissed; costs appeal allowed in part. The judge did not err in his negligent design analysis. He stated and applied the correct legal test, properly weighed the evidence and made no palpable and overriding errors of fact. In respect of failure to warn, the judge erred in his causation analysis by failing to properly articulate the nature of the warning that was required. However, despite this error, there is no basis to interfere with his conclusion that the evidence was insufficient to establish factual causation. In respect of costs, there is no reversible error in the judge's exercise of his discretion in refusing to make a Sanderson or Bullock order or awarding uplift costs to the bus manufacturer. However, the judge erred in principle in depriving the appellant of her costs of trial against the unsuccessful defendants without a sound basis. The judge's order to that effect is set aside and substituted with an order that the appellant is entitled to her costs of certain portions of the trial. Table of

Contents Paragraph Range A. Negligent design [7] - [115] Background facts [7] - [13] The decision below [14] - [24] On appeal [25] - [28] Legal principles - negligent design [29] - [115] 1. Applying the wrong legal test: failing to consider the gravity of harm [36] - [48] Submissions [38] - [40] Analysis [41] - [48] 2. Placing over-weighted reliance on the absence of a regulatory requirement for seatbelts [49] - [70] Submissions [50] - [55] Analysis [56] - [70] 3. Finding that it was not feasible in 1998 to install seatbelts [71] - [89] Submissions [71] - [76] Analysis [77] - [89] 4. Forgetting, ignoring or misconceiving critical evidence [90] - [115] Submissions [90] - [93] Analysis [94] - [115] B. Failure to warn [116] - [160] The decision below [116] - [123] On appeal [124] - [126] Legal principles - failure to warn [127] - [160] 1. The risk was "too remote" [131] - [137] 2. The nature of the warning - factual causation [138] - [160] Submissions [138] - [141] Analysis [142] - [160] C. Costs [161] - [219] The decision below [164] - [176] 1. Western's liability for costs [165] - [170] 2. Sanderson or Bullock order [171] - [173] 3. Scale of costs and uplift costs [174] - [176] On appeal [177] - [219] 1. Failing to make a Sanderson or Bullock order [181] - [195] 2. Failing to award trial costs against Western [196] - [211] 3. Uplift costs [212] - [219] D. Conclusion and

disposition [220] Reasons for Judgment of the Honourable Madam Justice Fisher: [1] This appeal arises from a single vehicle accident on August 28, 2014, when a tour bus with 55 passengers on board rolled on its side and crashed on the Coquihalla Highway. The bus was not equipped with passenger seat belts. The appellant, Jie Ding, was partially ejected through a side window and suffered serious injuries. [2] Actions were commenced by each of the passengers but only two of the claims proceeded to a single trial on liability to act as "bellwether" cases. The appellant's case was selected as being representative of passengers who had been ejected from the bus. [3] The driver and the owner of the bus, the respondents Brian Spittal and Western Bus Lines Ltd. (Western), admitted liability on the eve of trial. The appellant proceeded to trial with respect to the liability of the respondents Prévost, A Division of Volvo Group Canada Inc. (Prévost), Universal Coach Line Ltd. (Universal) and Canam Super Vacation Inc. (Canam). Prévost manufactured the bus, Canam organized the tour and contracted with Universal to provide a bus and driver, and Universal sub-contracted (or in industry terms, "farmed-out") the bus charter to Western. The appellant claimed against Prévost for negligent design in failing to incorporate seat belts into the design of the bus and against Universal and Canam for

failing to exercise reasonable care in selecting Western (negligent sub-contracting). She also claimed against Western, Universal, Canam and Prévost for failure to warn of the dangers of riding in a bus without seatbelts. [4] The trial judge dismissed all claims and made several orders for costs, mostly in favour of the defendants. The appellant appeals from the dismissal of her claim against Prévost for negligent design and the dismissal of her claims against Prévost and Canam for failure to warn.[1] She also appeals several aspects of the judge's order for costs. [5] For the reasons that follow, I see no basis on which this Court can interfere with the trial judge's dismissal of the appellant's claims in both negligent design and failure to warn. I would, however, allow the appeal of the costs order to a limited extent. I would set aside the order that each of Western, Mr. Spittal and the appellant bear their own costs of the trial and substitute an order that the appellant is entitled to her costs of the trial as against Western and Mr. Spittal, other than the time taken to address the discrete issues related to the claims for negligent sub‑contracting. [6] I will first address the negligent design and failure to warn claims separately. I will then address the costs appeal. A. Negligent design Background facts [7] Most of the evidence reviewed by the trial judge, Justice

Myers, is not disputed. His reasons for judgment are indexed as 2022 BCSC 215. [8] The bus that was involved in this accident was manufactured by Prévost in 1998. The design for this model, H3-45, began in 1992 and was introduced in 1993 as a 1994 model, with some design changes in 1995 and 1998 that included a change to the side windows from laminate to tempered glass. [9] The bus was a monocoque or unibody design, which has crumple zones meant to absorb energy in a crash. Passenger protection was provided by "compartmentalization" where the passenger is seated in a "soft shell" surrounded by smooth or energy absorbing material. The seats have high backs, arm rests and foot steps and the top rack is covered with carpet. The concept is to keep the passenger in the area of the compartment. Although this was the industry standard, approved by Canadian and American regulators, and thought to have an exceedingly safe record, it was recognized that rollover accidents were the most serious type of accident due to the risk of passenger ejection, but such accidents were rare. [10] The change to the side windows had been a matter of debate within Prévost but it was eventually decided that tempered glass was the preferred choice. Prior to 1997, Prévost had used laminate glass in aluminum frames. The design change to tempered glass in frameless windows gave passengers a better view. Very

simply, laminate glass has at least two layers of glass bonded to a middle sheet of plastic material. It breaks into shards, which are held together by the inner plastic layer. Tempered glass is strengthened, non-treated (annealed) glass that breaks into little pieces or kernels. Both types of glass met regulatory standards for passenger windows in Canada and the United States. [11] As the judge noted, the motor transport industry is heavily regulated. In Canada, Transport Canada administers safety standards under the Motor Vehicle Safety Act, S.C. 1993, c. 16, and the Motor Vehicle Safety Regulations, C.R.C., c.

1038. In the United States, the National Transportation Safety Board (NTSB) investigates transportation accidents, studies transportation safety issues and evaluates the effectiveness of government agencies. The National Highway Transportation Safety Agency (NHTSA) develops and implements safety standards and regulations for motor vehicles. There is frequent consultation and communication between the industry and the authorities in both countries. [12] When the H3-45 bus was designed and manufactured, there was no regulatory requirement for passenger seat belts in motor coaches in Canada and the United States. Canada made seat belts mandatory in newly manufactured motor coaches in 2020, the United States in 2016. Neither country required existing coaches to be retro-fitted with seat belts. Prior to this, only Australia (in 1994) and the European Union (in 1996) required the installation of passenger seat belts in new motor coaches. However, motor coaches in Australia and the EU were body-on-chassis designed, which uses a bus body attached to a ladder-type frame. This is different from the monocoque design in North America as it does not incorporate energy absorbing zones. [13] Prévost began to manufacture the H3-45 coach with seatbelts in 2009. It sold this model initially as standard equipment with an option to opt-out but removed this option a few years later. In 2010,

Prévost offered a retrofit program for existing buses at a cost of $30,000-$50,000. The judge found that a major impetus for Prévost's addition of seatbelts was a crash test done in December 2007 by the NHTSA. The decision below [14] The central issue at trial was whether Prévost, in designing and manufacturing the bus without seatbelts, breached the standard of care. The trial judge noted at the outset of his reasons that the advisability of seat belts in motor coaches was not a simple issue: [17] It was not a matter of "seatbelts good; no seatbelts bad". The technology for seatbelts in cars is not directly transposable to coaches. Coaches perform differently from cars and have different safety considerations. There was a concern as to whether coach passengers would use seatbelts; if they did not, they could be more severely injured by the hardened seats that seatbelt installations would require. [18] The advisability of seatbelts in motor coaches was one of ongoing investigation within the industry and the regulators in Canada and the United States. [15] The judge considered the regulatory and industry context to be a major factor in the evidence and arguments before him. He reviewed the history of the regulator's and industry's consideration of seatbelts in motor coaches as relevant to the regulatory environment, industry practice and state of knowledge of Prévost and the

other industry players. The details of this evidence are carefully set out at paras. 57-103 of his reasons, which I will not repeat here. He also reviewed the evidence as to why Prévost decided not to install seatbelts until 2009 as well as the expert evidence. He found the expert evidence adduced by the plaintiffs to be problematic in many respects but this aspect of his reasons is not an issue before us. [16] Before assessing this evidence, the judge set out the legal principles applicable to a claim for product liability: [205] Product liability is negligence-based and is not a tort of strict liability. To succeed, a plaintiff must show: a) the defendant owed a legal duty of care to the plaintiff in respect of the product; b) the product was defective or dangerous; c) the defendant was negligent in failing to meet the requisite standard of care; d) the breach of the standard of care caused the plaintiff's injuries; and e) the plaintiff suffered damage because of the defendant's negligence. Harrington v. Dow Corning Corp., 2000 BCCA 605 at para. 122. [206] As in any case of negligence, the analysis hinges on reasonableness. I think it fair to say--as Prévost has argued--that the question here is not whether Prévost was right or wrong in not installing seatbelts in 1998. Rather, the issue is whether it acted reasonably. As stated in Phillips v. Ford Motor Motor Co. of Canada,

[1971] 2 O.R. 637, 1971 CanLII 389 (O.N.C.A.) at para. 49: While the scope of M'Alister (or Donoghue) v. Stevenson, [1932] A.C. 562, has been greatly extended and is no longer limited to articles of food and drink, but has been applied to underwear, tombstones, motor-cars, elevators and hair dye, and more recently to house property and articles installed therein, 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. [207] A useful statement of the centrality of reasonableness is the following from Daishowa-Marubeni International Ltd. v. Toshiba International Corporation, 2010 ABQB 627: [38] The onus is on the plaintiff to show that the item as designed was not reasonably safe as there was a substantial likelihood of harm and it was feasible to design the product in a safer manner. The duty of reasonable care in design rests on the principle that the manufacturer should use reasonable care to eliminate any unreasonable risk or foreseeable harm. Tabrizi v. Whallon Machine Inc. (1996), 29 C.C.L.T. (2d) 176 (B.C.S.C.) quoting Rentway Canada Ltd. v. Laidlaw Transport Ltd. The requirement is not the safest design possible,

but rather one that is reasonable in the circumstances. . . . [40] The law does not impose strict liability on manufacturers, the onus does not require that they produce items that are accident proof or incapable of doing harm. The manufacturer is not the insurer of anyone who suffers injury while using or misusing a product. [Emphasis added.] [17] As noted, the judge considered the regulatory and industry standards to be central issues. He approached these issues in the following manner: [209] Because of the interaction of the industry and the government regulators, they are factually almost one and the same. [210] Compliance with regulatory standards does not give a manufacturer a "free pass" but it is a relevant factor in the reasonableness analysis and sometimes a very weighty one. As stated by Major J. in Ryan v. Victoria (City), [1999] 1 S.C.R. 201 [Ryan]: 29 Legislative standards are relevant to the common law standard of care, but the two are not necessarily co-extensive. The fact that a statute prescribes or prohibits certain activities may constitute evidence of reasonable conduct in a given situation, but it does not extinguish the underlying obligation of reasonableness. See R. in right of Canada v. Saskatchewan Wheat Pool, [1983] 1 S.C.R.

205. Thus, a statutory breach does not automatically give rise to civil liability; it is merely some evidence of negligence. See, e.g., Stewart v. Pettie, [1995] 1 S.C.R. 131, at para. 36, and Saskatchewan Wheat Pool, at p.

225. By the same token, mere compliance with a statute does not, in and of itself, preclude a finding of civil liability. . . . Statutory standards can, however, be highly relevant to the assessment of reasonable conduct in a particular case, and in fact may render reasonable an act or omission which would otherwise appear to be negligent. This allows courts to consider the legislative framework in which people and companies must operate, while at the same time recognizing that one cannot avoid the underlying obligation of reasonable care simply by discharging statutory duties. [Emphasis added.] [211] Similarly, industry standards are relevant, but not determinative. As the Ontario Court of Appeal noted in Zsoldos v. Canadian Pacific Railway, 2009 ONCA 55 at para. 43, conformity with industry practice does not protect a defendant from negligence where the practice itself is negligent. [18] Assessing this based on the state of affairs at the time Prévost manufactured the H3-45 coach, the judge summarized the crucial aspects of the evidence as follows: [213] a. Until 2020, there was no requirement in Canada to manufacture motor coaches with seatbelts. At the time the coach was built in 1998, the advisability of seatbelts in motor coaches was not a simple matter for the regulators or the industry. There was a weighing of the risks arising from different types of accidents,

seatbelt usage rates, and different bus designs, among other considerations. b. It was not a simple matter of installing seatbelts into an existing bus design without any further engineering. As stated by Mr. Osterman of the NTSB at the April 30, 2002 joint NHTSB-Transport Canada meeting: Although seat belts may be the right solution, you have to redesign the compartment with seat belts engineered into the equation rather than simply applying them to the existing seat that was never designed for belts c. Seatbelts had been discussed for many years prior to the manufacture of the coach and prior to the accident. It was recognised that ejections could take place in the event of a roll-over, but rollovers were considered to be, and were, extremely rare. In 1998, rollovers with fatalities constituted .017% of coach crashes in the U.S. d. As late as 1998 (the year the subject coach was manufactured), Transport Canada noted in its Review of Bus Safety Issues that "seat belts would be of potential benefit in only a very few cases". e. Well after the coach was manufactured, seatbelts remained under consideration as an open issue. As late as 2007, NHTSA noted that: [T]he fundamental information that would be necessary to establish adequate performance requirement for seat belts on motorcoaches does not exist. f. No North American bus manufacturer had the capability to do the

sophisticated crash testing using fully instrumented dummies to develop or implement a proper seatbelt standard. NHTSA did not conduct its crash test until 2007. Following that, Prévost was able to finalise its design for seatbelt implementation and the regulators were able to establish the seatbelt requirement with the appropriate standard. g. No North American seat manufacturers offered seatbelts until after 2006. [19] The judge noted that the plaintiffs did not really challenge the evidence that the industry and regulatory standard was not to have seat belts in motor coaches: [214] Rather the plaintiffs' position--without clearly saying as much--is that both the regulations and industry standards were negligent. However, the plaintiffs' expert evidence does not back that up, as is apparent from my review of that evidence [20] He considered the technical nature of this matter to be "something that must be weighed", as the installation of seatbelts was not necessarily "an obvious alternative that any reasonable manufacturer ought to have adopted": at para.

220. He added: [221] For me to conclude that Prévost was negligent in not installing seatbelts, I would have to conclude that NHTSA, Transport Canada, and other bus manufacturers in North America were equally negligent. That is not something to be done in the absence of cogent evidence: Piché v. Lecours Lumber Co. (1993), 41 A.C.W.S. (3d) 653 at para. 466 (Ont. Ct. (Gen. Div.)). [222] In order to conclude that Prévost, the industry and the regulators were negligent, I would have to balance at least the following factors: · The likelihood of rollover crashes; · The likelihood of users wearing seatbelts; and · The likelihood of non-seatbelt users being more severely injured because of the stiffening of the seats requited by seatbelt installations. The plaintiffs' expert evidence--to the extent it is admissible--does not assist in this regard. The totality of that evidence amounts to saying that seatbelts could have provided better protection in a roll-over. But that point is not in dispute ... [Emphasis in original.] [21] The judge found that Prévost acted reasonably in its design and manufacture of the H3-45 coach without seatbelts, adding: [227] It followed industry and regulatory standards. The plaintiffs have not shown that those standards were negligent or unreasonable or that another industry standard existed in North America. [22] He therefore concluded that Prévost was

not negligent in manufacturing the coach in 1998 without seatbelts and the lack of seatbelts did not make the bus unreasonably safe or defective. [23] The judge then addressed the issue of Prévost's use of tempered glass in the side windows. Although this was a central issue at the start of the trial, it was not a focus in final argument, as the appellant's counsel acknowledged that it could not be shown that any type of glass would have retained passengers. In any event, the judge noted the evidence about the initial debate within Prévost in 1996 regarding the proposed change to tempered glass: [239] At a March 1996 meeting, one of the former owners was against the move and expressed concern that one serious accident could be detrimental to the company and that the transition could be used against it in litigation. At the meeting, Mr. Bolduc expressed his view that Prévost's decision should be guided by the main goals of Volvo: quality, environment, and safety. The meeting ended with a decision to do further research. [24] He also noted that after further study, Prévost determined that laminate glass did not meet mechanical strength requirements of a frameless design and decided to use tempered glass. This decision was influenced by an expert analysis of the comparative benefits of each type of glass that found the greater strength of tempered glass prevented some ejections

and avoided laceration injuries that could occur with laminate glass. [25] The judge relied on the expert evidence adduced by Prévost on this issue and found the industry standard was to use tempered glass in the side windows. On this basis he concluded that the H3-45 coach was not negligently designed because it used tempered glass, "whether in and of itself, or in combination with the lack of seatbelts". He further concluded that laminate glass would not have prevented ejection, and "might even have caused further injury": at paras. 302-303. On appeal [26] The appellant submits that the trial judge erred in his analysis of Prévost's liability for negligent design by: 1. applying the wrong legal test by failing to consider the gravity of harm; 2. supplanting the correct test with over-weighted reliance on the absence of a regulatory requirement for seatbelts; 3. making a palpable and overriding error in finding that it was not feasible in 1998 to install seatbelts; and 4. forgetting, ignoring or misconceiving critical evidence in a way that affected his conclusions. [27] She characterizes (1) and (2) as errors of law, (3) as an error of fact, and (4) as either an error of law or of mixed fact and law. [28] Referring to the large body of evidence that was before the court below, Prévost submits that these issues relate to findings of fact, all of which are solidly supported by

the evidence. It contends the appellant is simply repeating arguments rejected by the trial judge after a careful review of the evidence. [29] There is no dispute that the standard of review for errors of fact and for mixed fact and law is palpable and overriding error, and for errors of law, correctness: Housen v. Nikolaisen, 2002 SCC 33. Legal principles - negligent design [30] The legal principles applicable to a claim in negligent design are not disputed. A manufacturer has a duty of care to avoid safety risks and to make products that are reasonably safe for their intended purpose. The mere fact that a manufacturer could have used a safer design does not automatically result in liability. A design must be one that is reasonable in the circumstances; a manufacturer is required to use reasonable care to eliminate any unreasonable risk of foreseeable harm: St Isidore Co-op Limited v. AG Growth International Inc., 2020 ABCA 447 at paras. 20, 22; Daishowa-Marubeni International Ltd. v. Toshiba International Corporation, 2010 ABQB 627 at para. 38. [31] To succeed in a claim for negligent design, a plaintiff must identify a design defect and establish that (1) the defect created a substantial likelihood of harm, and (2) there exists an alternative design that is safer and economically feasible to manufacture: Valeant Canada LP/Valeant Canada S.E.C. v. British Columbia, 2022 BCCA

366 at para. 116; Kreutner v. Waterloo Oxford Co‑Operative Inc. (2000), 50 O.R. (3d) 140 at para. 8 (C.A.). These requirements are the essential ingredients of a claim in negligent design of a product, which guide a determination of whether a design is reasonable in the circumstances, and more particularly, whether a manufacturer has met the standard of care by using reasonable care to avoid an unreasonable risk of foreseeable harm. [32] In determining whether a design is reasonable, the court is to apply a "risk utility" analysis, in which the risks inherent in the product as designed are balanced against the risks inherent in a safer, alternate design, considering the utility and cost of each. Factors to be considered in conducting this exercise include the utility of the product, the nature of the product in terms of the likelihood it will cause injury, the availability of a safer design, the potential for designing and manufacturing a safer product that remains functional and reasonably priced, the ability of the plaintiff to have avoided injury with careful use of the product, the degree of awareness of the potential danger of the product that reasonably can be attributed to the plaintiff, and the manufacturer's ability to spread any costs related to improving the safety of the design: St Isidore at para. 21; Daishowa-Marubeni at para. 39; Burr v. Tecumseh Products of

Canada Limited, 2023 ONCA 135 at paras. 57-58; Tabrizi v. Whallon Machine Inc., 1996 CanLII 3532 at para. 36 (B.C.S.C.). [33] The harm must be reasonably foreseeable. A manufacturer can only be held liable if the product in question had a design defect based on a safety risk the manufacturer either knew or ought to have known about at the time of manufacture (or a risk that when later discovered, was not addressed): St Isidore at para. 23; Burr at para.

59. In making this assessment, the court will consider the state of knowledge and technology at the time the product was manufactured and should hold the manufacturer to the same level of knowledge as an expert in its field: St Isidore at para. 23. [34] Finally, whether a manufacturer has created an unreasonable risk of harm is measured by the facts of each case and includes consideration of the likelihood of a known or foreseeable harm, the gravity of the harm, and the burden or cost that would be incurred to prevent the harm. External factors such as industry and regulatory standards may also be relevant, but are neither determinative nor co‑extensive: Ryan v. Victoria (City), [1999] 1 S.C.R. 201 at paras. 28-29. As Justice Major stated in Ryan: [29] Legislative standards are relevant to the common law standard of care, but the two are not necessarily co-extensive. The fact that a statute prescribes or prohibits certain activities may constitute evidence of reasonable conduct in a given situation, but it does not extinguish the underlying obligation of reasonableness Thus, a statutory breach does not automatically give rise to civil liability; it is merely some evidence of negligence. By the same token, mere compliance with a statute does not, in and of itself, preclude a finding of civil liability Statutory standards can, however, be highly relevant to the assessment of

reasonable conduct in a particular case, and in fact may render reasonable an act or omission which would otherwise appear to be negligent. This allows courts to consider the legislative framework in which people and companies must operate, while at the same time recognizing that one cannot avoid the underlying obligation of reasonable care simply by discharging statutory duties. [Emphasis added.] [35] Ryan did not involve a claim of negligent design but these principles apply equally to such claims. Prévost, as a manufacturer of motor coaches, has a duty to minimize the harm that may result from accidents and to design its vehicles to make them reasonably crashworthy. I would endorse this statement made by Justice Linden in Gallant v. Beitz (1983), 148 D.L.R. (3d) 522 at 525 (Ont. H.C.J.): Since motor vehicle manufacturers know or should know that many of their vehicles will be involved in collisions and that many people will be injured in those crashes, they must turn their minds to this matter during the process of planning the designs of their vehicles and they must employ reasonable efforts to reduce any risk to life and limb that may be inherent in the design of their products. [36] With these principles in mind, I turn to the four issues raised by the appellant in respect of her claim in negligent design.

1. Applying the wrong legal test: failing to consider the gravity of harm [37] The trial judge stated the correct legal test for negligent design. The appellant accepts that he did so but contends he failed to apply it. In essence, she says the judge altered the test by failing to consider a necessary factor, the gravity of harm. [38] I accept that this alleged failure to apply the correct legal test raises a question of law. Stating a test correctly but failing to apply it is an error of law: Teal Cedar Products Ltd. v. British Columbia, 2017 SCC 32 at para. 44; Clayworth v. Octaform Systems Inc., 2020 BCCA 117 at para. 47; Canada (Director of Investigation and Research) v. Southam Inc., [1997] 1 S.C.R. 748 at para.

39. Submissions [39] The appellant submits that the precautions to be taken must be commensurate with the danger, such that if the potential loss is great, creating even a slight risk may give rise to liability, citing Edmondson et al. v. Edmondson et al., 2022 NBCA 4 at para. 61, Allen M. Linden et al., Canadian Tort Law, 12th ed. (Toronto: LexisNexis Canada Inc., 2022), and Ryan. She says the judge failed to assess the magnitude of harm and instead equated Prévost's duty of care to the regulator's, thereby allowing Prévost "to shield its design choice to introduce tempered frameless glass without adding seatbelts behind the absence of a regulatory requirement for seatbelts". [40] The appellant relies on evidence of Prévost's knowledge at the relevant time that (a) compartmentalization was not an effective safety strategy in rollover accidents, (b) rollover accidents are the most serious type of bus accident due to the risk of passenger ejection, (c) seatbelts are highly effective in preventing passenger ejections in rollover accidents, and (d) the change in glass design reduced the effectiveness of compartmentalization as a safety measure to retain passengers during a rollover. She asserts that Prévost's design choice in 1998 was not reasonable when assessed against the magnitude of the harm. [41] Prévost submits that the trial judge made no legal error in his articulation

and application of the legal principles on negligent design. It says the judge did consider the magnitude of harm by emphasizing the importance of manufacturers and regulators balancing the risk of harm that could result from replacing the compartmentalization system with a new restraint system, as well as the need for reliable research and testing to evaluate alternatives and establish engineering standards before an effective seatbelt restraint system for motor coaches could be developed. Prévost also challenges the appellant's reliance on selected testimony to support her argument that Prévost made certain admissions regarding the effect of the change in glass design, emphasizing the extensive evidence on these issues and the need to assess it all in context. Analysis [42] In my view, the appellant's submission places undue emphasis on one factor, albeit an important one -- the gravity of the potential harm. In the context of this claim in negligent design, the judge is required to consider a myriad of factors in assessing whether Prévost's design was reasonable, as discussed above. This is not the type of case where an unreasonable risk of harm is measured only by balancing the danger created by a defendant's conduct against the utility of that conduct, or, more narrowly, by considering the gravity of the potential harm to outweigh all other factors, including the risk of

that harm occurring. Even in Edmondson, which is not a product liability case and involved a very different kind of dangerous conduct, the Court weighed the risk of harm and the severity of the potential harm against the utility of the conduct: see paras. 58, 61-63. [43] Moreover, it is apparent throughout the judge's reasons that he did consider the gravity of the potential harm resulting from rollover accidents and passenger ejections. He did not expressly refer to this in his final analysis of whether Prévost created an unreasonable risk of harm in designing the H3-45 bus without passenger seatbelts, but he was well aware of the fact that rollover accidents were the most serious type of accident due to the risk of passenger ejection and resulted in a disproportionate number of deaths in bus crashes. There was ample evidence on this point, which, as the judge noted, was not disputed. For example, he referred to the following evidence: · A 1997 Transport Canada internal review of the most serious motor coach collisions concluded that many fatalities resulted from ejection from the bus, often during a rollover event: at para. 63; · A 1999 NTSB report, Highway Special Investigation Report: Bus Crashworthiness Issues, noted the risk of ejection and injury in rollovers: at paras. 73-74; · At a 2002 joint public meeting with Transport Canada and NHTSA, Prévost and other industry

representatives recognized that ejections resulting in death could take place with rollovers: at para. 86; · A 2002 Transport Canada Report, Evaluation of Occupant Protection in Buses, prepared by RONA Kinetics and Associates Ltd. (the RONA Report), concluded that rollovers and ejections were the major causes of serious and fatal injuries: at paras. 91, 196; and · A 2009 Motor Coach Safety Action Plan released by the U.S. Department of Transportation stated the data indicated that ejection due to a rollover crash causes the highest percentage of motor coach passenger fatalities: at para. 106. [44] The judge expressly noted that the industry and the regulators recognized that rollover accidents were the most serious type of accident that could lead to death or serious injury due to the risk of passenger ejection. As this was not in dispute, he focused on weighing this serious harm against the safety record of the bus industry and the relative risk of the occurrence of rollover accidents. For example: [40] The concept of compartmentalization and the belief in its effectiveness was not an invention of Prévost. Rather, as will be seen, it was the industry standard, known and approved by the Canadian and US regulators. All thought the safety record of the bus industry in terms of overall crashes and serious injury to passengers was exceedingly safe. That said, it was recognised that

rollover accidents--as the one here was--were the most serious type of accident because it could lead to passenger ejection. However, rollover accidents were by far the minority of bus accidents and were considered rare events. [52] Again at its most general level, Prévost says that when the coach was designed and manufactured in 1998, the risk of ejection in a rollover was known, but that rollover incidents were exceedingly rare. It was not known until 2002 that they were responsible for a disproportionate number of deaths in motor coach crashes. Rollovers were, and remain, rare events, and crashes involving death or serious injury were rare. In 1998, government regulators were investigating whether seatbelts were the most effective way of enhancing safety, given that they could have a detrimental effect in other types of crashes. In 1998, the industry standard in North America--in collaboration with regulators in Canada and the United States--was not to have coaches with seatbelts. Prévost, it argues, therefore acted reasonably in manufacturing its coach without seatbelts. [53] I point out here that at the time the bus was designed, the regulators and industry knew of the existence of bus crashes, that some of those crashes were rollover events, and that passengers could be ejected from the bus in a rollover. I mention this at the outset because counsel for Ms. Ding

approached the matter as if these were denied by Prévost, but they were not. Rather, what Prévost puts at issue here is the balancing of risks. [Emphasis added.] [45] The appellant takes particular issue with this passage of the reasons: [222] In order to conclude that Prévost, the industry and the regulators were negligent, I would have to balance at least the following factors: · The likelihood of rollover crashes; · The likelihood of users wearing seatbelts; and · The likelihood of non-seatbelt users being more severely injured because of the stiffening of the seats requited by seatbelt installations. [Emphasis in original.] [46] She submits that the judge did not conduct a proper balancing because he did not include the gravity of the harm as a factor. Respectfully, I disagree. In the context of the reasons as a whole, it is implicit in this passage that the judge was balancing "at least" the three factors noted against the gravity of the potential harm -- i.e., death or serious injury by ejection in rollover accidents. This was clearly central to the entire case and there is no basis to suggest that the judge did not apprehend the seriousness of the potential harm to passengers without seatbelts. [47] This was not a simple case in which the gravity of harm moves the balance the other way. As the judge emphasized right from the start, the advisability of seatbelts in motor

coaches was not a simple issue. He was required to consider many factors given the large body of evidence before him. In my view, he made no reviewable error in the way in which he weighed that evidence and balanced the risks. [48] I do not consider the different nuances the appellant places on some of Prévost's evidence to be material. The judge accepted the expert evidence adduced by Prévost that the design change to tempered glass was reasonable and in keeping with industry standards. He also found that laminate glass would not have prevented ejection in this case and may have caused further injury, a point conceded by the plaintiffs at trial. I will address these evidentiary issues further in relation to the fourth ground of appeal.

2. Placing over-weighted reliance on the absence of a regulatory requirement for seatbelts [49] This ground of appeal raises a similar question of law as the first ground, as it alleges an incorrect application of correctly stated legal principles. Submissions [50] The appellant does not take issue with the trial judge's articulation of the legal principles in respect of the relevance of regulatory standards to the common law standard of care. He correctly stated that compliance with regulatory standards "does not give a manufacturer a 'free pass' but is a relevant factor in the reasonableness analysis and sometimes a very weighty one". He relied on Ryan at para. 29 (reproduced above), clearly recognizing that the underlying obligation of reasonable care cannot be avoided simply by discharging statutory duties. [51] Rather, the appellant contends that the judge, despite correctly stating the law, effectively conflated a manufacturer's duty of care with that of a regulator, and by doing so, added an element to the test for negligent design that the law does not require. She submits it is legal error to equate Prévost's design decision with an assessment of the conduct of the regulators, as the development of safety regulations and the imposition of civil liability on a manufacturer serve different purposes. She further submits the judge incorrectly allowed the absence of a

regulatory standard to absolve Prévost from liability for its design choice by impermissibly equating a finding of negligence against Prévost with a finding that the Canadian and American regulators were also negligent. [52] Although the appellant focuses on regulatory standards, she also contends that the judge equated the industry with the regulators and impermissibly equated a finding of negligence against Prévost with a finding that the industry at large was also negligent. She says the judge misdirected himself as to the import of statutory and industry compliance, giving it an unjustified importance. Her burden was to prove only that Prévost was negligent, not that the industry and the regulators were also negligent. [53] The appellant points in particular to this passage in the reasons for judgment: [221] For me to conclude that Prévost was negligent in not installing seatbelts, I would have to conclude that NHTSA, Transport Canada, and other bus manufacturers in North America were equally negligent. That is not something to be done in the absence of cogent evidence: Piché v. Lecours Lumber Co. (1993), 41 A.C.W.S. (3d) 653 at para. 466 (Ont. Ct. (Gen. Div.)). [54] In Prévost's submission, the trial judge did not equate the manufacturer's duty of care with the regulators', pointing to his express recognition that compliance with regulatory and industry standards was a

relevant, but not determinative, consideration. It submits that the judge made no error in placing substantial weight on industry and regulatory standards given the absence of expert evidence on the standard of care and the technical nature of the issues. [55] Prévost suggests that the judge was simply pointing out the obvious: if Prévost was negligent in not installing seatbelts in its coaches in 1998, then the regulators would have been equally negligent in not mandating seatbelts until more than 21 years later, particularly given the close collaboration between bus manufacturers and regulators in developing safety standards. Analysis [56] There is no question that the role of a regulator in developing safety standards is very different from the role of a manufacturer in designing a safe product. A manufacturer who complies with what can be described as minimum safety standards is nonetheless under the obligation of reasonableness: Ryan at para.

29. In this case, there was no regulatory requirement for passenger seatbelts in motor coaches at the relevant times and bus manufacturers had considerable discretion in their design decisions. Mere compliance, therefore, could not exhaust Prévost's standard of care: see Ryan at para. 40. [57] However, I do not agree with the appellant that the trial judge allowed the absence of a regulatory standard to absolve Prévost from liability for its design choice. The issues were far more complex than that, and both regulatory and industry standards were highly relevant. Given the amount of interaction between government regulators and the motor coach industry, the judge found the two "almost one and the same": at para.

209. Industry standards, especially when they involve highly technical knowledge, are important considerations in determining the standard of care. The weight to be given to these standards will of course depend on all the circumstances. For example, in Zsoldos v. Canadian Pacific Railway Co., 2009 ONCA 55 at para. 27, Justice Rosenberg observed that the application of industry standards should be approached with caution where the defendant was "virtually" the industry. [58] The judge's comments at para. 221 must be read in the context of his reasons as a whole, which in my view, demonstrate that he was alive to his proper task: assessing the reasonableness of Prévost's design choice. [59] The judge recognized that the concept of compartmentalization was the industry standard at the time Prévost designed the H3-45 bus given the overall safety record of the bus industry and despite the knowledge about the serious consequences of rollover accidents, which were considered rare. He carefully reviewed the large body of evidence in relation to the consideration of seatbelts by the regulators, the industry and by Prévost. The key evidence on these issues includes the following: · A 1998 Transport Canada report, Review of Bus Safety Issues, found travel by bus to be generally very safe. It recognized that seatbelts could help to prevent passenger ejection but concluded they "would be

of potential benefit in only a very few cases. They would need management by bus operators and reliable use by passengers to achieve effectiveness. The benefit is too uncertain to impose seat belts without a clear demand for a standard from the public and the motor carrier industry". The report also recognized the need for portions of the bus interior to be redesigned "to reflect passenger crash dynamics with specific kinds of seat belts". · A 1999 NTSB report, Highway Special Investigation Report: Bus Crashworthiness Issues, recommended that performance standards for motor coach occupant protection systems that account for front, side and rear impact collisions and rollovers, based on actual crash testing, be developed in two years. · A 2001 Transport Canada report, Bus Safety Consultations - Final Report, noted: There are very few passenger injuries that would potentially be prevented by seat belts and there are potential hazards involved with the use of seat belts by bus passengers, especially children. The most successful safety solution for car occupants is not necessarily the best for bus passengers. · The 2002 RONA Report concluded that rollovers and ejections were the major causes of serious and fatal injuries. · In 2003 or 2004, Prévost began the engineering planning for seatbelts. · A 2007 NHTSA report on motor coach safety compared the Australian and European

standards and concluded that the fundamental information necessary to establish adequate performance requirements for seat belts on motor coaches did not exist. It recommended that crash and sled testing be done and performance requirements be developed based on the results. · In December 2007, NHTSA conducted a crash test of a 2000 motor coach using crash dummies and reported the results to the industry in January 2008. · In 2009, Prévost began delivering motor coaches with seatbelts, with an opt‑out option. · In 2010, Prévost offered its customers a seatbelt retrofit. · In 2013, NHTSA published its final rule requiring three-point seatbelts for coaches manufactured after November 2016, when the requirement came into force. · In June 2018, Canada promulgated regulations requiring three-point seat belts on motor coaches, with no retrofit requirement, which came into force on September 1, 2020. [60] The judge accepted Prévost's evidence about the reasons it did not install seatbelts until 2009, making these findings: · A spike in serious motor coach accidents in 1998 and 1999 led the industry to question whether compartmentalization was sufficient. Prévost then began testing the seat tracks to which the seats were attached to determine whether they could be reinforced: at paras. 119-121. · In 2006 or 2007, Prévost learned from its seat suppliers that they did not have seats with

seatbelts: at para. 122. · After the data from the 2007 NHTSA crash test was released, Prévost had the data required to properly design a seat with seatbelts: at paras. 123-124, 213(f). · Prévost's decision not to install seatbelts in the 1998 H3-45 bus was not motivated by a desire to stay competitive and to save costs: at para. 125. · Crash testing was complex and only the regulators could undertake and understand it using fully instrumented dummies. The regulators were content to do the testing in collaboration with the industry: at para. 126. · Seats with belts were not available to Prévost until after 2006. Seats from Europe would not work for North American buses because buses in North America are bigger and wider than those in Europe, are not chassis-on-frame and have different anchorage points: at paras. 129-130, 144, 213(g). [61] I read the judge's summary of the crucial aspects of the evidence to incorporate the following findings: · Prior to 1998, when the H3-45 coach was manufactured, the regulators and the industry recognized that ejections could occur in rollover accidents but this kind of accident was considered to be extremely rare. · In 1998, installing passenger seatbelts in motor coaches was not a simple matter because the compartment had to be re-designed and re‑engineered and properly tested to performance standards. The regulators and the industry

considered the risks arising from things like different types of accidents, seatbelt usage rates and different bus designs. A Transport Canada report on bus safety issues found that travel by bus generally was very safe and although seatbelts could help to prevent passenger ejection, they would be of potential benefit in only a very few cases. · After 1998, the regulators and the industry continued to consider seatbelts. Adequate performance standards could not be established until sophisticated crash testing was done by one of the regulators, as the North American bus manufacturers did not have the capability of doing so. After the NHTSA conducted crash testing in 2007, Prévost was able to finalize its design to add seatbelts and the regulators were able to establish the appropriate standard for a seatbelt requirement. [62] Importantly, the judge found that the industry and regulatory "standard" when Prévost manufactured the bus in issue was not to have seatbelts in motor coaches. This was not challenged. As the judge observed: [214] The evidence adduced by Prévost clearly shows that the industry and regulatory standard was not to have seatbelts in motor coaches. The plaintiffs have mounted no real challenge to that. Rather the plaintiffs' position--without clearly saying as much--is that both the regulations and industry standards were negligent. However, the plaintiffs'

expert evidence does not back that up, as is apparent from my review of that evidence. [Italic emphasis in original; underline emphasis added.] [63] In my view, the underlined portion explains and anchors the trial judge's analysis. Because the appellant seemed to have accepted that Prévost's decision not to install seatbelts in 1998 aligned with the industry and regulatory standard, he understood her position to be, by necessary implication, that this standard was negligent (and so too, therefore, was Prévost's conduct in following it). I do not accept the appellant's submission that the judge impermissibly equated a finding of negligence against Prévost with a finding that both the industry at large and the regulators were also negligent. Nor do I accept her submission that he placed undue weight on the industry standard, or gave it "unjustified importance". [64] Conformity with an industry standard or practice may answer an allegation of negligence, as it tends to show what the industry considered sufficient, that most probably no other practical precautions could have been taken, and that an adverse outcome will negatively affect the entire industry: see ter Neuzen v. Korn, [1995] 3 S.C.R. 674 at para. 39, citing John G. Fleming, The Law of Torts, 7th ed. (Sydney: Law Book Co., 1987). However, an industry standard or practice may itself be found negligent if it is "fraught

with obvious risks". This usually applies to practices that do not require technical or specialized knowledge, such that anyone is capable of finding an impugned practice negligent through the exercise of ordinary common sense: see ter Neuzen at paras. 41, 44. These principles were applied in Zsoldos, where the industry practice that required only a passive warning system at rural railway crossings was found to have been fraught with obvious risks. [65] The trial judge was well aware of these principles, having referred to both ter Neuzen and Zsoldos. He also noted that the plaintiffs had not adduced expert evidence as to the requisite standard of care. While he did not consider expert evidence legally necessary, he said this: [219] the courts defer to industry standards where the practice is the result of a balancing of factors and the matter is technical. The point is also made (in the negative) by L.G. Theall et al, in Product Liability: Canadian Law and Practice (Toronto: Thompson Reuters, 2021) (loose-leaf updated 2021, release 1), s. 2.18: Evidence of custom will receive less weight where assessing the manufacturing method or standard does not require technical knowledge, and the trier of fact can therefore determine what was reasonable. Moreover, evidence that safer methods existed can diminish the probity of industry practice. [220] Apart from anything else, the

multiple studies and considerations of the issue of seatbelts by regulators in both Canada and the United States show that the matter here is a technical one the technical nature of this matter is something that must be weighed. In the face of the evidence, it is facile to say that the installation of seatbelts was an obvious alternative that any reasonable bus manufacturer ought to have adopted. [Italic emphasis in original; underline emphasis added.] [66] Applying these legal principles, the underlined conclusion is unassailable on the evidence. The standard of care question was a highly technical one and the judge made no error in finding the regulatory and industry standard to be highly probative, especially in the absence of expert evidence. As the judge found, the admissible expert evidence adduced by the plaintiffs simply stated that seatbelts could have provided better protection in a rollover, but that was not disputed. The judge properly considered the matter in the larger context. [67] As for Prévost's position in the motor coach industry, it is clearly a major player, being one of two large suppliers in North America. However, the evidence shows that other entities, such as bus and tour operators, were also involved in consultations with the regulators on these safety issues. The evidence also shows a considerable level of discussion in those consultations, far beyond the superficial. As the judge observed, "the relationship between the industry and the regulators was not a 'nudge-nudge-wink-wink' situation where industry and regulators ignore or defer consideration of known risks": at para.

223. In these circumstances, I see no basis on which to question the weight the judge assigned to the industry standard. [68] It was in this context that the judge made the statement impugned by the appellant (reproduced here for convenience): [221] For me to conclude that Prévost was negligent in not installing seatbelts, I would have to conclude that NHTSA, Transport Canada, and other bus manufacturers in North America were equally negligent. [69] In my opinion, it is simplistic to suggest that the judge approached this case as one that required the appellant to prove that all of Prévost, the industry at large and the Canadian and American regulators were negligent. He correctly focused on whether Prévost was negligent in manufacturing the H3-45 coach without seatbelts in 1998 and properly considered the regulatory and industry standards to be evidence probative of this question. In doing so, he applied the correct legal principles and placed appropriate weight on the complex technical evidence relevant to whether Prévost met the requisite standard of care.

3. Finding that it was not feasible in 1998 to install seatbelts Submissions [70] The appellant submits that the trial judge made a palpable and overriding error in finding it was not feasible in 1998 to install seatbelts. She says his analysis on this point was limited to a conclusion that seats with seatbelts were not commercially available to Prévost until after 2006 but Prévost's own evidence established that Prévost could have incorporated passenger seatbelts into its bus in 1998 had it elected to do so. [71] The appellant relies on the following evidence (in cross-examination) from William Gardner, a former head of Transport Canada's Crashworthiness division, who gave evidence on behalf of Prévost as both a fact and expert witness: Q And in 1998 you would agree that there were opportunities - and when I say "opportunities" I mean the ability and feasibility - of installing 3-point seatbelts in that 1998 bus that Prévost designed and manufactured, correct? A Yeah. [72] She asserts that this is consistent with two other pieces of evidence: (1) a conclusion in the RONA Report that "[f]easible and cost effective seat systems are now available which allow the fitting of lap/torso belts for bus occupants", and (2) the evidence of Gaetan Bolduc, Prévost's former president, that the consensus of the Prévost product committee in 1996 was to "not push the safety issues not

requested by legislations" but had the consensus been to go in a different direction, there was access to manufacturers and suppliers to get the technology in place to implement seatbelts in North America "to some sort of standard". [73] The appellant says this evidence established that Prévost had the ability to incorporate passenger seatbelts in 1998, and this meets the second part of the test for negligent design described in Valeant (whether there exists an alternative design that is safer and economically feasible to manufacture). [74] In Prévost's submission, the trial judge's finding that seats with seatbelts were not available to Prévost until after 2006 is well-supported in the evidence. Prévost emphasizes that it obtains seats from seat manufacturers and points to (1) email correspondence with seat suppliers indicating that these suppliers did not offer seat belts on their passenger seats and testimony from Mr. Bolduc to that same effect, and (2) an internal Prévost project document indicating that, although there was consumer demand for seats with seat belts, "[o]ur seat suppliers are not ready yet to supply us with three point seat belts". [75] Prévost also submits that the appellant's argument ignores the evidence of Mr. Bolduc and Mr. Gardner, neither of whom suggested it would have been safe, sensible or practical to install seats from Europe or Australia, the

only ones available, onto a Prévost coach in 1998. It says their evidence was "strongly to the contrary, because the passenger restraint systems in both regions had not been adequately tested, and the seats were not designed for use in monocoque construction and were not sized for the larger Prévost buses". Analysis [76] As I understand the appellant's argument, she does not challenge the judge's finding that seats with seatbelts were not commercially available until 2006, but says he ought to have assessed whether Prévost could have incorporated passenger seatbelts into its bus in 1998 had it elected to do so. In my view, this argument cannot succeed, as the evidence the appellant relies on does not establish that there existed an alternative design that was safer and economically feasible to manufacture in 1998. [77] Although the appellant correctly observes that the trial judge did not conduct a "dedicated analysis of the feasibility of designing and installing passenger seatbelts in 1998", he made several pertinent findings throughout his reasons in relation to this issue, all of which are amply supported by the evidence. Moreover, his finding on this point was more nuanced than simply stating that seats with seatbelts were not available at that time, but rather seats with three-point restraints were not available for North American buses (or no North American seat manufacturers offered seat belts) until 2006: at paras. 144, 213(g). [78] This finding is consistent with the evidence as to the differences between the monocoque design for buses in North America and the chassis-on-frame design for buses in Europe. As the judge found, seats from European seat manufacturers would not work for North American buses: at para.

129. It is also consistent with the evidence about the need to develop performance standards after proper testing to ensure seatbelts actually conferred safety benefits. The judge accepted that Prévost could not have simply adopted the standards developed in Europe and Australia: [224] The plaintiffs point to the European and Australian standards and argue that Prévost could have simply adopted them. However, the regulators did not consider the testing to have been adequate. Moreover, the tests were for a different bus construction using a chassis-on-frame construction. In fact, the European standard was found to be too low, being 6.6g compared to the 20g that was eventually adopted. [79] In essence, the judge reasoned that until adequate testing was conducted to inform performance standards for North American buses, there were no available seats with three-point seatbelts available that could reasonably be expected to render the Prévost bus safer. [80] In my opinion, none of the evidence relied upon by the appellant suggests otherwise. [81] Mr. Gardner's affirmative response to the suggestion that there were "opportunities" for installing seatbelts in 1998 is speculative at best and not probative of whether a safer alternative design was available. Even if Prévost could have installed some type of seatbelt that year, it does not necessarily follow that its buses would have

been made safer by doing so. Indeed, Mr. Gardner's evidence as to why Transport Canada had not recommended the installation of seatbelts in 1998, which the judge apparently accepted, highlighted engineering issues like the need for further testing beyond that which informed the European or Australian standards, and by implication, the risks associated with installing seat systems based on those standards into North American buses. The judge referred to this evidence in these passages: [201] Mr. Gardner was asked why Transport Canada had not recommended the installation of seatbelts by 1998. He stated that significantly more information was required and could only be obtained from dynamic crash testing which had not been done. [202] Mr. Gardner testified that he was well aware of the seatbelt regulations in Australia and Europe. However, it was not transposable to Canada without further testing. He stated: I think the most general conclusion would be that neither ECE 80 nor ADR 68 could be just accepted into the Canadian regulations. They were -- neither one of them were suitable for our type of sled testing, and neither one of them -- and we weren't really able to determine from our testing whether either one of them was completely appropriate or not. We would have had to do some more of our own testing. Ignoring the requirements of ADR 68 and ECE 80, we would have had

establish our own criteria based on a sled test pulse that we felt was correct and that was achievable on our sled. So a lot of work would have to be done before we could actually reapply any of these standards. [82] Mr. Bolduc's evidence is similarly limited to the following question and answer read-in from his examination for discovery: Q and reading this document [June 5, 1996 report of the product committee meeting], would you agree with me it seems clear that if the consensus had been to go a different direction that there was access to manufacturers and suppliers, be they Plaxton [a U.K. seat manufacturer] or others, to get the technology into place to implement seatbelts in North America? MR. TREVETHAN: The European standard, right? MR. BUCKLEY: Q To some sort of standard? A Well, yeah. Well, yes. Yes. [83] In my view, Mr. Bolduc does nothing more than agree to the vague proposition put to him that if Prévost had pursued it, Prévost (together with seat manufacturers and suppliers) could have put the "technology into place" to install seatbelts in North American buses built to "some sort of standard". This evidence is entirely speculative and also not probative of whether a safer alternative design was available. [84] In fact, the judge apparently accepted Mr. Bolduc's explanation as to why Prévost did not simply accept the European or Australian standards while waiting

for further testing to be conducted by NHTSA. That explanation was, in part, that: "we were not in a position to follow something that was not properly done. That's not our way of doing things." The view of American and Canadian regulators, the judge added, "was in line with that approach": at para. 131. [85] Finally, the passage in the RONA Report the appellant relies on cannot be assessed in isolation. The RONA Report described the results of a review of bus occupant protection research and regulatory practices in Canada, the United States, Australia and Europe. When considered in context, it is apparent that the "feasible" seat systems to which the report refers were those designed to meet specific performance standards then in place (European and Australian), standards which the evidence showed were not transposable to North American buses. This evidence does not suggest that cost-effective seat systems that would make North American buses safer existed. [86] To this end, I agree with Prévost that the appellant's argument ignores the important question of whether a safer alternative design existed: Neither Mr. Bolduc nor Mr. Gardner suggested that it would have been safe, sensible or practical to install seats from either Europe or Australia (the only seats available) onto a Prévost coach in 1998 - their evidence was in fact strongly to the contrary, because the passenger restraint systems in both regions had not been adequately tested, and the seats were not designed for use in monocoque construction and were not sized for the larger Prévost buses. [87] This evidence was uncontroverted.

4. Forgetting, ignoring or misconceiving critical evidence Submissions [88] The appellant submits that the trial judge erred by failing to address critical evidence of Prévost's witnesses that assisted her case. Because a judge is obligated to "consider all of the evidence, whoever brings it" (relying on Lemieux v. British Columbia (Superintendent of Motor Vehicles), 2019 BCCA 230 at para. 55), she says that Prévost's own evidence required careful consideration in assessing whether Prévost's design of the bus met the standard of a reasonably safe design having regard to the risk of harm and the feasibility of a safer alternative (i.e., a design with seatbelts). [89] The appellant says the judge failed to address the following evidence, all of which was helpful to her case: a. the evidence of Mr. Gardner and Mr. Bolduc to the effect that Prévost could have incorporated three-point seatbelts into its design of the bus in 1998 had it chosen to do so; b. the large body of uncontested evidence from Prévost's witnesses which established that (1) compartmentalization is not an effective safety strategy in rollover accidents; (2) rollover accidents are the most serious type of bus accident due to the risk of passenger ejection; (3) seatbelts are highly effective in preventing ejections in rollover accidents; and (4) Prévost knew that changing its window design from laminate glass to

frameless tempered glass reduced the effectiveness of compartmentalization as a safety measure to retain passengers in the bus during a rollover; and c. Mr. Bolduc's evidence that he did not know why Prévost did not install seatbelts at the time it changed the window design from laminate to tempered glass. [90] The appellant contends that while the judge seemingly accepted the evidence of Mr. Bolduc and Prévost's expert witnesses at trial, he did not refer to these aspects of their evidence, all of which weighed in favour of her case. She says the judge's failure to consider this large body of evidence in his negligence analysis resulted in a material error, as it gives rise to a "reasoned belief that the trial judge must have forgotten, ignored, or misconceived the evidence in a way that affected his conclusion": Davis v. Jeyaratnam, 2022 BCCA 273 at para. 52. [91] Prévost takes issue with what it considers to be a strained focus by the appellant on certain statements made by Mr. Bolduc and Mr. Gardner. Prévost notes that there "was a similar emphasis on these statements in closing submissions at trial, and it is notable that the trial judge did not consider any of these statements warranted a reference in his reasons for judgment". It says the appellant ascribes undue importance to evidence that was properly not given significant weight by the judge in his

analysis. Analysis [92] It is well recognized that trial judges are not required to address every piece of evidence in their reasons for judgment, nor are they bound to accept uncontradicted and cogent evidence. However, it is incumbent on judges to provide an explanation where uncontradicted and cogent evidence is rejected, especially on key or critical issues: Jampolsky v. Insurance Corporation of British Columbia, 2015 BCCA 87 at para. 40, citing Savinkoff v. Seggewiss (1996), 25 B.C.L.R. (3d) 1 at paras. 17-21 (C.A.); British Columbia v. Canadian Forest Products Ltd., 2018 BCCA 124 at para.

168. In the absence of an explanation, the reasons must still be read as a whole to properly assess whether the basis on which the impugned evidence was not relied on is evident: Jampolsky at para. 40. [93] Moreover, intervention on appeal will generally be available only where the lack of an explanation leads to a palpable and overriding error (see Savinkoff at para. 20), or where the failure to discuss a relevant factor gives rise to a "reasoned belief that the trial judge must have forgotten, ignored or misconceived the evidence in a way that affected his conclusion": Housen at para. 39, citing Van de Perre v. Edwards, 2001 SCC 60 at para. 15; Davis at para.

52. In the absence of either type of error, an appellate court cannot reconsider the evidence heard at trial. [94] In Davis, this Court found reversible error in the judge's failure to conduct a proper analysis of the credibility of competing witnesses as to whether an accident had in fact occurred, the main issue at trial. In that circumstance, Justice Bennett, for the Court, found a failure to consider a large body of important evidence to have resulted in a material error. She stated it was incumbent on the judge to consider evidence that would or could support one version of events or the other before reaching a conclusion on which version to accept: [63] In my view, the trial judge committed reversible error in failing to conduct the proper analysis and overlooking a significant body of evidence that could have weighed in favour of [the appellant's] version of events and his ultimate credibility. Whether it is an error of law or a palpable or overriding error of mixed fact and law, the result is the same. Had the judge considered the evidence, she may have come to a different conclusion regarding [the appellant's] evidence, the question of whether he collided with the [respondents'] vehicle at all and the issue of whether he was riding his bicycle on the sidewalk. [95] Respectfully, I would not characterize the evidence relied on by the appellant as "a significant body of

evidence", nor would I characterize it as cogent. I will address the three "bodies" of evidence the appellant relies on in turn. [96] I have already assessed the lack of cogency of the evidence of Mr. Gardner and Mr. Bolduc as to whether Prévost could have incorporated three-point seatbelts into its design of the bus in 1998. This evidence could not have weighed in the appellant's favour given its speculative nature. [97] The second body of evidence deals largely with evidence that was uncontested: compartmentalization was not effective in rollover accidents, rollover accidents are the most serious type of bus accident due to the risk of passenger ejection, and seatbelts are highly effective in preventing ejections. None of this was ignored or misconceived by the judge. For example, he observed: [53] I point out here that at the time the bus was designed, the regulators and industry knew of the existence of bus crashes, that some of those crashes were rollover events, and that passengers could be ejected from the bus in a rollover. I mention this at the outset because counsel for Ms. Ding approached the matter as if these were denied by Prévost, but they were not. Rather, what Prévost puts at issue here is the balancing of risks. This will become clearer as this judgment unfolds. [140] As I indicated above, the issue here is not whether there was an absolute risk: the matter of

is one of relative risk and a balancing of multiple factors. It is an issue that needs to be seen in its broader context. [222] The totality of the [plaintiffs' expert evidence] amounts to saying that seatbelts could have provided better protection in a roll-over. But that point is not in dispute. As I said several times, the plaintiffs' experts did not consider the matter in the larger context. [98] This point, made repeatedly by the trial judge, can be stated rather simply: that Prévost knew rollover accidents occurred and that passengers could be ejected during those accidents was not in dispute. Nor was the proposition that seatbelts could have provided better passenger protection in a rollover accident. Neither of these facts was determinative of Prévost's liability. Rather, liability turned on relative risks and a balancing of multiple factors to answer the governing question of whether Prévost's design choice was unreasonable in the circumstances. [99] This point clearly influenced the trial judge's treatment of the evidence. For example, he referred to the evidence of one of the plaintiffs' experts that: The design of the seat system used in the subject Bus poses a risk to the safety of all passengers because none of the passenger seating is equipped with seatbelts. Therefore, passengers would not be adequately restrained during collisions, especially those involving

significant 'lateral' (sideways) and 'vertical (up and down) accelerations, such as rollovers and high severity side-impacts. This would have passengers at high risk of serious injury and death from forceful impacts during such events. [100] The trial judge found this opinion unhelpful because the issue was not whether the absence of seatbelts created an absolute risk, but whether the bus design was reasonable, a question which had to be situated in the broader context: at para. 140. [101] The only evidence of some controversy was the question of Prévost's change to the window design from laminate to tempered glass. [102] To support her argument on this point, the appellant relies on minutes from several of Prévost's product committee meetings in 1996 in which the change to the glass design was discussed, and portions of Mr. Bolduc's evidence at trial where he agreed with these propositions put to him in cross examination: (1) the change to tempered glass was a significant change to the compartment of the bus; (2) Prévost recognized the risk that tempered glass could shatter in a rollover leaving a complete opening; (3) the design change increased the risk for passengers in a rollover collision; and (4) he did not know why Prévost did not shift to installing seatbelts into their "H" series bus when they shifted to tempered glass. [103] Prévost counters this argument by

referring to other passages of Mr. Bolduc's evidence that address the question of "risk" more fully and why his "I don't know" answer to why Prévost did not also shift to installing seatbelts was not a full response. It submits that Mr. Bolduc's evidence must be taken in the context of three days of discovery and six days of testimony at trial. [104] I have reviewed all of the evidence referred to by both parties and I agree with Prévost that the appellant's selected excerpts of both the documents and testimony cannot be assessed in isolation. They do not give a fair picture of the disputed evidence. Nor do they take into account the expert evidence the judge relied on to make his findings on the glass design issue. [105] For example, the appellant points to the following evidence of Mr. Bolduc, the first read-in from his examination for discovery and the second from his cross-examination at trial: Q Okay. So, by changing the glass, and again we're dealing primarily with the accident - rollover accidents in particular - by changing the glass from laminated to tempered your - that risk for the passenger in the rollover collision has now gone up if the window breaks; correct? A Correct. If they are not belted, yes. Q And at the time you were aware that tempered glass in the event of a tip-over or rollover had the risk of creating ejection portals for passengers on the bus,

right? A Well, that's correct. But plastic and laminated also to a certain point. But what you are saying is correct. [106] Prévost points to this evidence a little later in Mr. Bolduc's cross-examination: Q We know from our discussion this morning that in the design of the 1998 bus, with the change to the windows, there was an increased exposure to ejection and injury if the bus tipped over with the tempered glass; right? A When you say "increased", let's be clear. There's accident, there's rollover, there's flip-overs where the laminates disappeared. It depends on the conditions of the bus Q The effect of the change to the glass in an accident like the one that we're here about today, the tip-over accident, was to increase the risk of the passengers of ejection and injury? A I don't agree with you because the crash we're talking about today, there is nothing that tells me that laminated glass would have survived that. The windshields were out of the bus and there was still tempered glass on that side of the bus. Again, I was not there. I didn't look at the bus I'm not an expert in all that. But just what I saw tells me that I cannot agree with what you're saying. Q you changed the design of your H-series bus in a way that is going to enhance the risk to the passenger in a rollover accident by changing from laminated glass to a tempered glass; agreed? A Yes. MR. TREVETHAN: You

mean the risk of ejection. MR. BUCKLEY: Yes. MR. TREVETHAN: Not the risk of shards penetrating. MR. BUCKLEY: The risk of ejection, yes. Q Do you agree with that? A Yeah. [107] These examples demonstrate that the question of the precise risk was a nuanced one, pinpointed later in the cross examination to a risk of ejection, rather than an overall risk to passengers. Mr. Bolduc was not qualified to assess overall risk, as he acknowledged. [108] As for Mr. Bolduc's "I don't know" answer, I do not see this evidence as material. Throughout his reasons, the trial judge made ample references to Mr. Bolduc's testimony, including his evidence as to why Prévost did not install seatbelts until 2009, and the impetus for the decision to make the design change when it did: see paras. 115-131. I agree with Prévost that a single statement of this nature made by Mr. Bolduc on discovery must be taken in the context of his extensive testimony at trial, which provided comprehensive reasons why Prévost installed seatbelts when it did and explained why he did not give a full response at discovery. [109] As for the concerns expressed in the minutes of Prévost's product committee, the judge did not ignore this evidence. He expressly noted the initial debate within Prévost, including the concerns about potential litigation, but evidently considered that the decision to use tempered glass was made after

further study: [241] The decision was influenced, in part, by a peer-reviewed paper by a preeminent glazing expert, Professor Lawrence Patrick, who was a professor at Wayne State University, a leading university for vehicle design in Detroit. One of the other experts who testified at the trial, Mr. Richards, referred to Professor Patrick as "one of the grandfathers of biomechanics" whose papers are frequently referred to by other engineers and in graduate courses. Another expert, Mr. Ridenour, said that Professor Patrick was a leading pioneer in human impact tolerance and safety research and that his work formed the basis for many current automotive safety standards and features. [242] In his paper, Professor Patrick analysed the comparative benefits of laminated and tempered glass in several different combinations and concluded that tempered glass was the preferred choice. He found that the greater strength of tempered glass prevented some ejections and avoided the laceration injuries that could take place with laminate glass. He also noted the savings in weight, which resulted in the benefit of a lower centre of gravity and therefore more stable vehicle. The decreased weight also yields better gas mileage. [L.M. Patrick, "Glazing for Motor Vehicles", presented at the 39th Stapp Car Crash Conference, Nov. 1995.] [243] Although the person at Prévost who consulted Professor

Patrick was not a witness at the trial, records from a June 1997 project committee meeting notes that the glass project and proposed glass type were presented to him "and accepted". [110] The judge reviewed the expert evidence adduced by both parties. He found some of the opinions of the plaintiff's experts inadmissible and gave little to no weight to those that were. He preferred the opinions of Prévost's two experts. Importantly, he grounded his finding that the industry standard was to use tempered glass in side windows on the reasons given by Jack Ridenour, Prévost's glazing expert. Namely: a. U.S. and Canadian standards in place since 1967 identified tempered glass as appropriate for use in side-windows of vehicles; there was never a requirement for use of laminate glass: at para. 284; b. For side-windows, the use of tempered glass is the preferred option and the one which is invariably used in the industry because laminated glass has two serious disadvantages: it is weaker than tempered glass and creates an increased risk of laceration to passengers if the glass fractures: at para. 285; c. There are three serious types of laceration risk associated with laminated glass, each of which could cause severe injury: at para. 286; d. Virtually all motor vehicles made in 1998 to the present use tempered glass for the side windows: at para. 287; e. In the face of the research by

Professor Patrick, it would have been irresponsible for Prévost to have used laminate glass in the side windows: at para. 289; and f. No glass would have survived the crash given the multiple forces the windows were subjected to: at paras. 290, 294. [111] Mr. Ridenour's summary of his opinion, apparently accepted in full by the trial judge, was that: the decision by Prévost to install tempered glass in its 1998 H3-45 was entirely reasonable and in keeping with standards in the motor vehicle industry. The use of laminated glass in motor vehicles is fraught with perils that did not make it suitable for installation as side windows in 1998. I am also of the opinion that if laminated glass was installed in the 1998 H3-45 that glass would also have likely failed and led to ejection of passengers. The laminated glass may well have also caused more severe laceration injuries. [112] The judge's review of the evidence throughout his reasons demonstrates that he was very much aware of what was material and what was not, and what he accepted and what he did not. He made no error in not expressly rejecting the aspects of Mr. Bolduc's evidence the appellant relies on. None of it, when assessed in the proper context, was sufficiently cogent to have affected the judge's conclusion that the H3-45 bus was not negligently designed by using tempered glass, whether in and of itself, or in

combination with a lack of seatbelts. [113] For all of these reasons, I find no basis to interfere with the trial judge's dismissal of the claim against Prévost for negligent design. B. Failure to warn The decision below [114] The Second Amended Consolidated Notice of Civil Claim (NOCC) included claims for failure to warn as against Western, Universal, Canam and Prévost. The claims against Western, Universal and Canam made the same allegation: each failed to warn users of the bus "that the bus was not equipped with seatbelts and/or the lack of seatbelts could cause harm" to users of the bus. The claim against Prévost was more precise: that Prévost failed to warn purchasers, owners or users of the bus of the risk of damage or injury, "particularly involving" things that included rollover incidents and ejection. [115] The trial judge's reasons regarding these claims are relatively brief. He outlined the legal principles, which were not in dispute, as follows: [387] A manufacturer has a duty to warn users of its product of the dangers that it knows or ought to know are inherent in its use. This duty exists even if there are no manufacturing or design defects as long as there are dangers to using the product: Hollis v. Dow Corning Corp., [1995] 4 S.C.R. 634 at paras. 20-22. [388] However there is no duty to warn where an ordinarily prudent person would be aware of the risk: Schulz

v. Leeside Developments Ltd., 90 D.L.R. (3d) 98 (B.C.C.A.); (1984) Krysta v. Funland Enterprises Ltd., 57 B.C.L.R. 32 (S.C.). [389] In determining whether a warning need be given, the materiality of the subject matter is relevant both in terms of the risk of the event occurring and the potential severity of the injury: Double Bar L Ranching Ltd. v. Bayvet Corp., [1996] 10 W.W.R. 673 (Sask. C.A); leave ref'd [1996] S.C.C.A. No. 623. [390] Causation is an element of a claim for failure to warn, as it is with any other tort claim. The plaintiffs must show that the warning would have altered their behaviour; in other words, that they would not have booked the tour or got on the bus if a warning had been provided. [116] The judge then noted that the plaintiffs had not differentiated between the different defendants despite each being in a different position with respect to the warnings they could have given: · Prévost and Western had no relationship with the plaintiffs prior to them boarding the bus. The only warning they could have given was a warning label or plaque on the bus. · Universal had no relationship with the plaintiffs at all and could not have provided the plaintiffs with a warning on the bus because it was not a Universal bus. That is sufficient to dismiss the action for failure to warn against it. · CanAm was the only defendant that could have warned the plaintiffs

before the tour began, whether in its sales brochures, on its website, or its booking form. [At para. 391.] [117] He considered these differences to be "of potential significance" to the issue of causation. He addressed causation first as he considered it to be determinative of these claims. [118] The judge applied a subjective test (as set out in Hollis v. Dow Corning Corp., [1995] 4 S.C.R. 634 at paras. 44-46) in assessing whether the plaintiffs would have heeded a warning. He described the appellant's evidence as follows: [397] Ms. Ding testified that she wanted to do a tour with her son before he started school so she tried to find a tour to fit her schedule. She said that if she had an option between buses with or without seatbelts, she would have taken the one with belts because of safety, including the risk of ejection. She wears seatbelts in cars, but she has taken taxis in China which did not have seatbelts because they were not required by the government. She was fine with doing that because she assumed the government did not think it was a safety issue. [119] He noted that the appellant did not say she would not have booked the tour had she known the bus did not have seatbelts, but rather she would have preferred a bus with seatbelts, she did not make any advance inquiry about seatbelts, and she was aware of the risk of lack of seatbelts. He found this evidence

insufficient to ground liability against any of the defendants. [120] With respect to Prévost and Western, the judge found that posting a warning sticker or plaque on the bus would not have made a difference because the plaintiffs said they were aware of the risk of lack of seatbelts, the lack of seatbelts "was not a hidden defect" but rather something they "would have noticed immediately on getting on the bus" and they continued with the tour anyway. [121] With respect to Canam, which could have provided an advance warning, the judge found that any warning given would not have made a difference to the plaintiffs and they would have booked the tour anyway. He concluded: [403] Causation has not been proven by either plaintiff against any defendant and the failure to warn claim must be dismissed on that basis. [404] But beyond that, I do not think it was necessary to provide a warning here. The risk of a roll-over was too remote and the lack of seatbelts was obvious to anyone getting on the bus. On appeal [122] The appellant submits that the trial judge erred in law and principle in dismissing her claims against Prévost and Canam for failure to warn by: 1. concluding that the risk of a rollover was too remote to require a warning; and 2. mischaracterizing the nature of the warning asserted and required by law, which led him to dismiss the claim on the basis that causation had not

been established. [123] Prévost's position is that it had no duty to warn potential passengers that the bus did not have seatbelts given the judge's finding that buses manufactured without seatbelts were not hazardous or defective, and in any event the claim fails on the issue of causation given the judge's finding that the appellant would not have been deterred from riding on the bus had Prévost installed a warning plaque. It says the judge made no palpable and overriding error in making these findings. [124] Canam contends that it did not have a duty to warn the appellant about the absence of seatbelts and in any event, causation was not proven. Legal principles - failure to warn [125] Again, the legal principles applicable to a claim in failure to warn are not disputed. [126] It is well settled at common law that manufacturers have a duty to warn consumers of dangers inherent in the use of their products of which they have knowledge or ought to have knowledge. This duty is a continuing one, requiring manufacturers to warn not only of dangers known at the time of sale, but also of dangers discovered after the product has been sold and delivered. Warnings must clearly describe any specific dangers that arise from the use of the product: Hollis at para. 20; Buchan v. Ortho Pharmaceutical (Canada) Ltd. (1986), 25 D.L.R. (4th) 658 at 667 (C.A.); Adam v. Ledesma-Cadhit, 2021 ONCA 828 at para. 19. [127] As consumers have far less knowledge than manufacturers concerning the dangers inherent in the use of their products, the duty to warn serves to correct this imbalance by alerting consumers to any dangers and allowing them to make informed decisions about the safe use of a product: Hollis at para.

21. Liability for failure to warn is not based merely on knowledge imbalance, however, but on the reliance consumers reasonably place on manufacturers of products. In this context, the duty to warn may be negated where the consumer has full knowledge of the risk, sufficient to negate reasonable reliance on the manufacturer: see Bow Valley Husky (Bermuda) Ltd. v. Saint John Shipbuilding Ltd., [1997] 3 S.C.R. 1210 at para. 22 per McLachlin J. (dissenting, but not on this point); St Isidore at para. 34. [128] The duty to warn encompasses all dangers inherent in the use of a product that are reasonably foreseeable: Andrew Botterell et al, Fridman's The Law of Torts in Canada, 4th ed. (Toronto: Thomson Reuters, 2020) at 676. A danger that is statistically small may nevertheless be reasonably foreseeable, in particular, where the risk posed by the potential danger is a serious one: see Hollis at para. 41; St Isidore at paras. 35-36.

1. The risk was "too remote" [129] The appellant contends that the judge misapplied these legal principles in concluding that no warning was required because the risk of a rollover was "remote". She interprets the judge's use of the word "remote" as indicating that the risk of a rollover was low and says that the jurisprudence confirms that even if a risk is rare, manufacturers and suppliers have an obligation to warn users of their products of known risks that may cause grave personal harm. The appellant submits that dismissing the need for a warning by characterizing the risk as "remote" fails to apply the correct legal standard, and in this case, the gravity of the potential harm to unbelted passengers in rollover collisions "mandates that steps be taken to warn passengers of the specific risks of riding in a motor coach without seatbelts". [130] In my view, the judge's description of the risk of a rollover being "too remote" is problematic for several reasons. First, while I accept that the judge was simply referring to a low risk, he was doing so in the context of addressing, albeit briefly, whether there was a duty to warn. His use of the word "remote" appears to confuse the remoteness inquiry relevant to a legal causation analysis with the reasonable foreseeability inquiry relevant to a duty of care analysis. Second, insofar as the judge decided that none of the

defendants owed the plaintiffs a duty to warn, I agree with the appellant that he erred in relying only on the low risk of a rollover occurring to negate a duty, as this is inconsistent with the principles outlined above. A low risk with potentially grave consequences may indeed be reasonably foreseeable, and a duty to warn may be required, depending, of course, on the circumstances. [131] That said, the appellant's reliance on the gravity of harm to mandate a duty to warn does not distinguish between Prévost (a manufacturer) and Canam (a tour operator) as defendants, nor does it address the proximity analysis that is essential to a determination of whether a duty of care exists and to whom. The NOCC asserts a failure by Prévost to warn "purchasers, owners and/or users of the bus" and failure by Canam to warn "users of the bus". As the judge noted, each defendant had a different relationship with the plaintiffs. [132] Canam's argument before this Court focuses on whether it owed the appellant a duty to warn and distinguishes its role and knowledge of the risk from that of Prévost as the manufacturer of the bus. Prévost simply asserts that it had no duty to warn because the bus was neither defective nor dangerous. [133] The jurisprudence supports a duty to warn on the part of manufacturers in respect of known dangers inherent in the use of their products. Although the judge

found that buses without seatbelts were not defective or dangerous products, he also found that Prévost had particular knowledge about the limitations of compartmentalization in rollover accidents and the fact that rollovers were the most serious type of accident due to the risk of ejection. While the judge found that the bus without seatbelts itself was not a dangerous product, there is some merit to the appellant's argument that Prévost, as the manufacturer, had a duty to warn of the known danger inherent in its use given the extent of its knowledge of the rare but serious risk of injury to unbelted passengers in the event of a rollover accident. [134] The jurisprudence is not so clear as to the application of the principles to different classes of defendants, such as a tour operator in Canam's position. Canam submits that it does not have a legally recognized duty to warn and disputes the appellant's contention that it had sufficient knowledge of the specific dangers to justify imposing a duty in this case. In a short reply, the appellant contends that the duty to warn in this context is not restricted to manufacturers or distributors of a product, but if it is, a duty arose on an application of the Anns/Cooper test. [135] Importantly, other than his "too remote" comment, the trial judge did not conduct a duty of care analysis in respect of either Prévost or Canam. While he

made findings regarding Prévost's knowledge of the risk, he made none about Canam's knowledge, and the inferences to be drawn from Canam's evidence are very much in dispute. The appellant did not make comprehensive submissions on the legal basis for imposing a duty to warn on either Prévost or Canam, and neither the appellant nor Canam did so on the application of the Anns/Cooper test. In these circumstances, as well as the judge's focus on factual causation, I do not consider it appropriate or necessary to conduct a fresh duty of care analysis of these claims. The appeal can be determined on the question of causation.

2. The nature of the warning - factual causation Submissions [136] The appellant accepts that the trial judge stated the correct legal principles applicable to claims for failure to warn but submits that his analysis of factual causation was flawed because he mischaracterized the nature of the required warning. She says the judge asked the wrong question: The issue was not simply whether an advance warning that the bus had no seatbelts would have altered Ms. Ding's behaviour. The correct inquiry was evaluating the impact of a proper warning to Ms. Ding describing the specific dangers to unbelted passengers resulting from the lack of seatbelts. [137] She contends the respondents, having knowledge of the significant risks to unbelted passengers in rollover accidents, were required to warn passengers not only that the bus had no seatbelts, but also of the specific risks arising from the lack of seatbelts: in particular, that passengers were at an increased risk of serious injury due to passenger ejection in the event of a rollover accident due to the lack of seatbelts, and the bus's safety features were not designed to protect passengers in such an accident. She says a fair inference to draw from her evidence is that she would not have taken this tour had she been properly warned. [138] Prévost submits that the trial judge made no reversible error in finding that the appellant was

aware of the risks, including the risk of ejection, and her stated preference for seatbelts was insufficient to establish causation. It says the result is the same regardless of the specific content of the warning. [139] Canam also relies on the judge's conclusion that the appellant failed to establish causation. It says further that the nature of the warning the appellant seeks from Canam has become more specific than that sought at trial, yet the evidence did not establish that Canam had anywhere near the level of knowledge of Prévost about the increased risk of serious injury in rollover accidents. Analysis [140] I will start by saying that trial judges need to be cautious about dismissing claims for lack of causation without first undertaking some duty of care analysis -- at least to identify the standard of care. Factual causation requires a plaintiff to show on a balance of probabilities that the harm would not have occurred but for the defendant's negligent act or omission: Nelson (City) v. Marchi, 2021 SCC 41 at para. 96; Resurfice Corp. v. Hanke, 2007 SCC 7 at paras. 21-22. Since it is the defendant's negligent act or omission that must be shown to have caused the harm, it is necessary to assess causation in the context of the defendant's breach of the standard of care: Chasczewski v. 528089 Ontario Inc. (Whitby Ambulance Service), 2012 ONCA 97 at para. 15. [141]

Therefore, trial judges should generally decide whether a defendant has breached the standard of care before resolving issues of causation: Armstrong v. Royal Victoria Hospital, 2019 ONCA 963 at paras. 60, 138, rev'd 2021 SCC 1 (aff'g dissent of van Rensburg J.A.). This does not mean that judges must do so, but where they proceed to assess causation assuming there is a duty that has been breached, it is important to articulate what the standard of care required a defendant to do or not do: Sheoran v. Interior Health Authority, 2023 BCCA 318 at paras. 58-60. [142] Where the alleged breach of the duty of care is a failure to warn, the precise nature of the warning that ought to have been given must be properly defined before causation can be properly analyzed. For example, in Revelstoke (City) v. Gelowitz, 2023 BCCA 139, a case of failure to warn of shallow water, the defendant was found to have failed to meet the standard of care by not placing or maintaining "no diving" warning signs at specific locations. This specificity was critical in the assessment of factual causation in the context there, where the judge found that the plaintiff would not have attempted a shallow dive had he seen a warning sign at either of the required locations. [143] In this case, it is not clear what the judge considered the required warning to be for the purposes of his causation analysis. He referred to Prévost warning "potential passengers that the bus did not have seatbelts": at para.

307. He also cited the appellant's position that all defendants owed a duty to warn "of the dangers of the bus not having seatbelts" and more particularly "the specific design of the Bus without seatbelts and tempered glass that would break in a rollover increasing the risk of ejection and injury or death": at paras. 385-386. [144] The judge proceeded to specify how each of Prévost and Canam could have provided warnings to bus users, assuming they each had a duty to warn: Prévost by providing a warning label or plaque on the bus itself, and Canam by doing so in its sales brochure, on its website or in its booking form. However, he did not further articulate the necessary content of the warning. [145] I agree with the appellant that the judge's causation analysis was flawed because he did not properly articulate the nature of the warning that was required by each of Prévost and Canam. However, assuming the existence of a duty to warn, I do not agree that the specific nature of the warning would necessarily be the same for both Prévost and Canam. Nor do I agree that the appellant's evidence was sufficient to establish factual causation even with a more specific warning describing the dangers to unbelted passengers resulting from the lack of seatbelts. [146] First, as to the content of a warning. I accept that a simple warning from either Prévost or Canam that the bus did not have

seatbelts would be insufficient in light of the evidence as to the gravity of the potential danger of ejection in the event of a rollover accident. Warnings must clearly describe the nature of the risk and the specific dangers that arise from the ordinary use of a product so that users are able to make informed decisions about using the product: Buchan at 667; Hollis at paras. 20-22. [147] Given the judge's findings as to Prévost's knowledge of the risk, I also accept that Prévost would be required to warn its purchasers and the ultimate users of the bus of the specific risks to passengers which it knew resulted from the lack of seatbelts, to this effect: in the event of a rollover collision, while a rare occurrence, a passenger would be at risk of being ejected from the bus and seriously injured. [148] Canam, on the other hand, was in a very different position as a tour operator from Prévost as a manufacturer. This appears to have been recognized by the plaintiffs, who alleged in the NOCC that Canam was required to warn bus users only that the bus "was not equipped with seatbelts" or "the lack of seatbelts could cause harm". On appeal, the appellant suggested a more detailed warning outlining the dangers resulting from the lack of seatbelts, although her counsel did concede that a warning by Canam would be different in detail from that of Prévost, calibrated to its level of

knowledge. [149] The absence of findings as to Canam's level of knowledge does not permit me to define the precise content of a required warning from Canam. For the purpose of a factual causation analysis, I will assume that a warning given by Canam would be similar in substance to that of Prévost. [150] Therefore, the alleged breach of the standard of care is that both Prévost and Canam failed to warn the appellant as a user of the bus, that in the event of a rollover collision, while a rare occurrence, a passenger would be at risk of being ejected from the bus and seriously injured. [151] Second, as to whether the harm would not have occurred "but for" the failure to provide this warning, the trial judge's finding on factual causation is clearly subject to a standard of palpable and overriding error: Revelstoke (City) at para.

62. The appellant's evidence on this point was limited to the following: Q And had you been in a position where you had an option to take one of two buses, one with seatbelts and one without seatbelts, which one would you pick? A Seatbelt bus. Q Why? A Safety. And when driving I know it's the law always put the seatbelt on. Everybody - in the car when I drive, I make sure everybody put a seatbelt on. [In direct examination.] Q When you told Mr. Slater that you would prefer a bus with seatbelts, is that because you were of the thought that having a seatbelt reduced the risk of injury to you? A Yes. Q And reduced the risk of you being ejected from a vehicle if there was an accident? A Yes. Q And those were your thoughts before this accident, correct? A Well, so before the accident or before the - I booked a tour. Q Before this accident I can clarify, ma'am. Before this accident - A Yes. Q Before this accident, it was your thought that a vehicle with seatbelts would prevent or reduce the risk of injury or ejection from a vehicle, and that was something you were aware of before this accident? A Yes. Yes. [In cross-examination.] [152] In assessing causation, the judge applied a subjective test, recognizing that a plaintiff's evidence does not have to be taken at face value. As Justice La Forest noted in Hollis, the concern that plaintiffs, with the benefit of hindsight, will always

claim that they would not have used the product had they been properly warned, can be adequately addressed at trial through cross-examination and a proper weighing of the relevant testimony: at paras. 45-46. It is recognized that hindsight evidence, which occurs in a variety of circumstances, is inherently self serving and is to be approached with "a healthy skepticism": see Revelstoke (City) at para. 73, citing Century Services Corp. v. LeRoy, 2018 BCCA 279 at para. 74. [153] The judge found that a warning plaque installed on the bus by Prévost would not have made a difference to the appellant. After noting that she stated only a preference for a bus with seatbelts, he found she was aware of the risk of lack of seatbelts and this was something she would have noticed immediately upon getting on the bus. The testimony cited above supports not only that the appellant was aware of the risk in general but also that she was specifically aware of the risk of passenger ejection. While this does not necessarily establish a full appreciation of the risk of ejection in rollover accidents specifically, it does establish a level of knowledge that seatbelts would prevent or reduce the risk of injury or ejection more broadly. The judge also found that an advance warning by Canam would not have made a difference. He noted in particular the appellant's testimony that she would have preferred a

bus with seatbelts and the lack of evidence that Canam offered tours with seat belts. [154] Whether or not the appellant's evidence as to her knowledge of the risk of injury or ejection is sufficient in itself to negate reasonable reliance on Prévost as a manufacturer, her evidence of a simple preference for a bus with seatbelts does not, in my view, meet the "but/for" test of causation. [155] I do not accept the appellant's submission that a fair inference to be drawn from her evidence is that she would not have taken the tour had she received the proper warnings. The causation inquiry here is not concerned with a hypothetical choice between a bus with seatbelts and a bus without seatbelts. Rather, the inquiry is whether the specific warning (described above) would have resulted in different conduct; namely, that the appellant would not have booked the tour had she received the warning from Canam and she would not have continued the tour had she received the warning from Prévost. [156] The onus was on the appellant to prove the causative relationship between the lack of the specific warning and her injuries. In my view, the lack of evidence on this critical issue was fatal to her claim for the failure to warn, notwithstanding the judge's flawed analysis. There was a similar lack of evidence about the sufficiency of warnings regarding the use of a hockey helmet, which was fatal

to the plaintiff's claim in More v. Bauer Nike Hockey Inc., 2011 BCCA 419: see paras. 50-51. [157] Therefore, despite the trial judge's failure to articulate the nature and scope of the warning, I see no basis to interfere with his conclusion that causation had not been proven with respect to the claims against both Prévost and Canam. C. Costs [158] The issue of costs was complex in light of the distinct roles of the various defendants and the allegations made against them, as well as the admission of liability made by Western and Mr. Spittal on the eve of trial. That admission was made on the basis that Western was vicariously liable for the negligence of Mr. Spittal, who admitted that the accident was caused by inattention, not fatigue. The negligent sub-contracting claim against Canam (and Universal) was that they were negligent in selecting Western because the bus did not have seatbelts and Western had no training programme in place that allowed driver fatigue to be identified. The liability of Prévost for negligent design did not depend on what caused the accident as the claim was that the injuries were made worse due to the lack of seatbelts. Only the failure to warn claims, which also did not depend on proof of the cause of the accident, were similar as against each of Western, Universal, Canam and Prévost. [159] In their amended response to the NOCC, Western and Mr.

Spittal admitted that the accident was caused by Mr. Spittal's negligence but denied the particular allegations of negligence (notably that he fell asleep) or that the plaintiffs suffered the injury, loss or damage as alleged. The plaintiffs nonetheless pursued the allegation that the accident was caused by lack of training and safety procedures resulting in driver fatigue, as well as the failure to warn claims, all of which were dismissed. [160] Western and Mr. Spittal sought an order for costs of the trial despite their admission of liability on the basis that the plaintiffs pursued allegations against them that they did not admit, which were dismissed. The plaintiffs sought costs against Western and Mr. Spittal and a Sanderson or Bullock order that Western bear the costs of the other defendants. Prévost, Canam and Universal sought costs at Scale C as well as uplift costs. The decision below [161] The reasons for judgment on costs are indexed as 2023 BCSC 518. The judge addressed the following issues: (1) Western's liability for costs, or vice versa; (2) whether a Sanderson or Bullock order should be made against Western and Mr. Spittal; and (3) the scale of costs and uplift costs in favour of the other defendants.

1. Western's liability for costs [162] The trial judge recognized the general rule that the successful party is entitled to costs and the substantial onus on the party urging departure from this general rule. He also recognized that normally if liability is found against a party for negligence, the court will not parse out which basis of a claim succeeded and apportion costs between the issues that succeeded and those that failed. He cited Loft v. Nat, 2014 BCCA 108 at para. 46, where this Court described the successful party as a plaintiff who establishes liability under a cause of action and obtains a remedy, or a defendant who obtains a dismissal of the plaintiff's case. [163] In addressing Western's argument that it should be awarded costs of the trial, the judge considered the governing rule to be Rule 14-1(15) of the Supreme Court Civil Rules, B.C. Reg. 168/2009, which permits the court to award costs to an unsuccessful party or deny costs to a successful party with respect to an identifiable issue or part of a proceeding. A major consideration was whether it was reasonable for Western to have actively participated in the trial after it admitted liability. The judge did not accept several of Western's arguments but did accept that Western was justified in defending the safety and fatigue issues because of the potential effect on apportionment of damages. On that basis, he

concluded it was reasonable for Western to participate in the trial when those matters were being addressed. [164] The judge went on the assess the impact of Western's admission: [27] This case does not fit into the "normal" paradigm, because of the admission of liability. Once that admission was made, there was only one theory left for trial regarding the cause of the accident. This was that Spittal had fallen asleep, which in turn was based on lack of training and safety measures. It was not proved that Spittal fell asleep nor that there were insufficient safety measures or training at Western. Moreover, it was not shown what further measures ought to have been taken and whether the accident would have been avoided if those measures or training were in place. Therefore, the only theory of liability that was alive when the trial started was unsuccessful. [165] The judge referred to the appellant's submission that she was substantially successful against Western and Mr. Spittal but was required to prove certain facts related to them in order to pursue her case against the other defendants, and that Western and Mr. Spittal were not required to defend specific particulars of negligence given that they had already admitted liability. He then stated: [28] In some ways, this makes Western's point. It may have been necessary for the plaintiffs to try to prove the fatigue argument

vis-à-vis their claim against the other defendants, but that does not mean that Western should bear the costs of that. [166] Finally, the judge considered the lateness of the admission, which denied the plaintiffs the opportunity to plan accordingly. [167] He then concluded: [30] In the unique circumstances of this case, I conclude that Western, Spittal and the plaintiffs should bear their own costs of the trial. There is no question that the plaintiffs are entitled to their costs against Western up to the time of trial.

2. Sanderson or Bullock order [168] The judge noted that the Supreme Court Civil Rules permit judges to make Sanderson or Bullock orders if the costs of one defendant against a plaintiff ought to be paid by another defendant.[2] He also noted the two-part test established in case law that guides the exercise of the discretion to make such an order: (1) whether it was reasonable for the plaintiff to have joined the successful defendant in the action (a threshold issue), and (2) whether it is just and fair in the circumstances to make the order. He cited Provost v. Dueck Downtown Chevrolet Buick GMC Limited, 2021 BCCA 15 at paras. 17-19. He held that the second part of the test required a "highly contextualised and fact-based" analysis that involved consideration of the strength of the case against the successful defendant and whether the unsuccessful defendant did something to justify a Sanderson or Bullock order. [169] The judge found it was reasonable in this case to have all the claims joined in one action and therefore the plaintiffs met the threshold test. He did not, however, find it just or fair to make the order under the second part of the test. He found the following factors relevant to this conclusion: · The claims against both the non-successful defendants (Western and Mr. Spittal) and the successful defendants (Canam, Universal and Prévost) were not strong; ·

Western did not do something to merit responsibility for the costs of the other defendants as its response to the NOCC was a pro forma pleading allowing for contribution from the other defendants if the plaintiffs made out their case against them; · In respect of the failure to warn claims, it should be open for a defendant to deny it had the duty of care alleged but the other defendants did, without exposing itself to bear the costs of the other defendants: "Put colloquially, the plaintiff called the party and decided who to invite, without input from Western"; and · The fact that these were bellwether cases did not justify Western bearing the costs. [170] The judge rejected the plaintiffs' argument that any pleading by the liable defendants alleging responsibility of the non-liable defendants was enough to merit, or even mandate, a Bullock order, as this would remove the court's discretion to assess whether such an order was just and fair in the circumstances. He also rejected their argument that this Court's decision in Provost supported the imposition of a Sanderson or Bullock order: [51] In that case the court was influenced by the fact that the liable defendant had stolen a car from the non-liable defendant and rammed it into the plaintiffs' (at para. 22). Further, the Court noted that in one of the three actions concerned, the liable defendant had filed a third-party

notice against the other defendants. And finally, in another one of the actions the liable defendant denied that he had caused the accident. That is not the case here: Western did not allege that the accident was caused by any other defendants; rather its position was that if they were liable, they contributed to the damages. Western was following the plaintiff's theory of the case: that the lack of seatbelts and choice of glass made the injuries more severe than they would otherwise have been. It is to be recalled there was never any issue that the accident was caused by the non-Western/Spittal defendants. [Emphasis in original.] 3. Scale of costs and uplift costs [171] The judge found this case to warrant Scale C costs in favour of the successful defendants due to its complexity with respect to the facts and the legal issues, and this is not challenged on appeal. What is challenged is the judge's imposition of uplift costs at 1.5 times the value of the Scale C costs in favour of Prévost. [172] He referred first to s. 2(5) and (6) of Appendix B of the Supreme Court Civil Rules. Section 2(5) gives the court discretion to order increased costs of 1.5 times the value where it finds an award to be "grossly inadequate or unjust" as a result of "unusual circumstances". Section 2(6) provides that an award "is not grossly inadequate or unjust merely because there is a difference

between the actual legal expenses of a party" and the costs to which that party would be entitled. [173] The judge then referred to Shen v. West Continent Development Inc. (BC0844848), 2022 BCSC 462, which summarized the case law regarding the factors relevant to assessing uplift costs. He found three factors justified such an award to Prévost: the importance of the case to Prévost, the significance of the case to the coach industry in Canada, and the complexity of the case, which justified having several counsel. He observed that Prévost did the "heavy lifting" on whether buses ought to have been equipped with seatbelts and was the only defendant who had to defend the glass issue. He found that none of the factors justified an uplift award to the other defendants. On appeal [174] The appellant challenges three aspects of the judge's costs award. She contends he erred by (1) failing to grant a Sanderson or Bullock order making Western and Mr. Spittal responsible for the costs of the other defendants; (2) failing to award her costs of the trial against Western and Mr. Spittal; and (3) awarding Prévost uplift costs. Western, Mr. Spittal and Prévost support the judge's costs award as reflecting a proper exercise of his discretion. [175] It is undisputed that the scope of appellate review of a trial judge's order for costs is limited, as such orders engage a broad exercise of

discretion by trial judges best placed to assess costs-related issues. This Court may interfere with a costs order only where the judge misdirected himself as to the applicable law, made an error in principle, made a palpable error in assessing the facts, or otherwise made an award that is plainly wrong: Briante v. Vancouver Island Health Authority, 2017 BCCA 148 at para. 197; AM Gold Inc. v. Kaizen Discovery Inc., 2022BCCA 284 at para. 63, and the cases cited therein. [176] It bears repeating that the broad discretion accorded to trial judges in assessing costs must be exercised judicially. What this means may vary according to the nature of the question. In British Columbia v. Worthington (Canada) Inc. et al. (1988), 29 B.C.L.R. (2d) 145 at 154 (C.A.), Justice Lambert (dissenting in the result) described the difference between some questions and others: In many questions of discretion there is no right answer and no wrong answer, and those questions are questions of true discretion. But, there are other questions of discretion where there is a right answer and a wrong answer, although the right answer is not dictated by positive and settled law. A question of discretion is one which requires more flexibility than that. Nonetheless, the administration of justice would be brought into disrepute if one judge could decide such a question one way and another judge, faced with

exactly the same situation, could decide it another way. Judicial comity, common sense, and a regard for the seemly administration of justice, require that the same question should be decided in the same way each time it arises. That way, the matter becomes predictable. Guidelines for the consistent exercise of the discretion flow from the mode of exercise of the discretion by one's predecessors. In the end, a failure to exercise such a discretion in accordance with its consistent past exercise becomes a failure to exercise the discretion judicially. The discretion has become a principled discretion, as opposed to a true discretion. [177] The reasoning of Lambert J.A. is consistent with this Court's jurisprudence before and after Worthington, as observed by Justice Fitch in Briante at para. 202. [178] This distinction is important in this case. As the discussion below will demonstrate, the discretion exercised in determining whether to make a Sanderson or Bullock order is different from the discretion exercised in determining whether to deprive a successful party of her costs.

1. Failing to make a Sanderson or Bullock order [179] Rule 14-1(18) of the Supreme Court Civil Rules permits a court to order an unsuccessful defendant to pay the costs of a successful defendant, either directly (a Sanderson order) or indirectly to the plaintiff as a disbursement (a Bullock order). As the trial judge noted, a two-part test is applied in assessing whether such an order should be made. This was confirmed by this Court in Provost: [17] The threshold test for a Sanderson or Bullock order is whether it is reasonable for the plaintiff to have joined the successful defendant in the action. If this is satisfied, the question becomes whether it "would be just and fair in the circumstances" for the unsuccessful defendant to pay the successful defendant's costs: Davidson at paras. 53-56. In Grassi v. WIC Radio Ltd., 2001 BCCA 376 at para. 32, Justice Southin described the secondary consideration as whether the unsuccessful defendant "ought" to pay the costs of the successful defendant. The decision to award a Sanderson or Bullock order is a matter of discretion, which must be exercised judicially based on the judge's assessment of the circumstances of the case: Robertson v. North Island College Technical and Vocational Institute (1980), 119 D.L.R. (3d) 17 (B.C.C.A.) at para. 24. [180] The threshold test (whether it was reasonable for the successful defendant to have been

joined in the action) must be answered affirmatively but the second part of the test (whether a Sanderson or Bullock order is just and fair in the circumstances) involves what Lambert J.A. described in Robertson v. North Island College Technical and Vocational Institute (1980), 119 D.L.R. (3d) 17 at 24 (B.C.C.A), as a "true discretion": Once the threshold question is answered affirmatively then the discretion of the trial judge arises. Of course, he may exercise it either way. It is a true discretion. Whether he grants a Bullock order, or not, must depend on his assessment of the circumstances of the case. In my opinion it is inappropriate to trammel that discretion by endeavouring to extract principles from those cases where the discretion was exercised and from those cases where it was refused. The threshold question must be answered affirmatively; the discretion must be exercised judicially; and that is all. [181] The issue here relates to the second part of the test, as the trial judge found the threshold test to have been met. In this regard, the unsuccessful defendant must have done something to warrant an order requiring it to be responsible for a successful defendant's costs. This would include an assertion that blames the successful defendant, as well as an act that caused the successful defendant to be brought into the litigation: Provost at para. 18, citing Grassi v.

WIC Radio Ltd., 2001 BCCA 376 at paras. 33-34. In some cases, a meritless claim may militate against making the order: see, for example, Davidson v. Tahtsa Timber Ltd., 2010 BCCA 528 at para. 56, where the Court did not consider it just and fair to make the order where there was "no credible evidence" that supported the claim against the successful defendant. [182] The appellant submits that in declining to grant a Sanderson or Bullock order, the judge failed to exercise his discretion in accordance with past practice by improperly distinguishing this Court's guidance in Provost, and thereby erred in principle. She says a Sanderson or Bullock order is often appropriate where the unsuccessful defendant takes the position that another defendant was responsible for the plaintiff's loss or "was the culprit in the case", as described in Grassi at para. 33, and contends that the factual differences between this case and Provost are distinctions without a difference. The appellant argues that the judge's decision undermines the goals of consistency and predictability. [183] In Provost, the defendant Bolton stole a truck from the defendant Dueck (who had carelessly stored the vehicle) and caused a series of accidents when police pursued the stolen vehicle at high speed, colliding with three different vehicles (two being RCMP vehicles). The Attorney General of Canada and Constable

Provost (who had been driving one of the RCMP vehicles) commenced actions for damages against Bolton and Dueck. The third driver commenced an action against Bolton, Dueck and the Minister of Justice for British Columbia (for vicarious liability for the conduct of the RCMP officers). The three actions were heard together. The trial judge found the defendants liable and, in each action, apportioned most of the fault to Bolton and 15% of the fault to Dueck. Dueck successfully appealed the finding of liability against it in all three actions on the basis that it did not owe a duty of care to the plaintiffs. It was therefore entitled to costs of the appeal and the proceedings below. At issue was whether Dueck's costs of defending the three actions were payable by the unsuccessful defendants, primarily Bolton, by way of a Sanderson or Bullock order. [184] The Court made Sanderson orders in favour of all three plaintiffs. It concluded that the threshold question was easily satisfied as "it was eminently reasonable for the plaintiffs to join Dueck as a defendant". At the second stage, the Court considered the following factors to favour a Sanderson order: · The actions of Bolton caused Dueck to be brought into the litigation as his theft of the truck and his actions in evading the police were central to the liability considerations involving Dueck and the police officers; · In each of

the actions, Bolton's pleadings attempted to shift blame for the accident onto Dueck and others, although at trial he did not actively seek to blame Dueck; · Although Bolton did not make submissions about Dueck's liability at trial or on appeal, he did not have to do so in order to obtain the benefit of a finding that Dueck was liable because his pleadings alleged Dueck to be at fault. Dueck's liability was always going to be a live issue at trial in all three actions; · The liability contest between Bolton and Dueck was of no practical significance to the plaintiffs, who should not have their judgment funds eroded by this contest between the defendants; · Dueck was not blameless in the events leading to the accidents, having carelessly stored its vehicle; and · In the third action, all three defendants filed third party notices against each other: Bolton blamed Dueck, Dueck blamed Bolton and the Minister, and the Minister blamed Bolton and Dueck. [185] The appellant submits that the same considerations are at play in this case, as it was Mr. Spittal's negligent operation of the bus that caused the lawsuit in the first place, and there is no principled basis to distinguish the assertions in the pleadings of Western and Mr. Spittal that the other defendants caused the plaintiffs' losses. She says their amended response to the NOCC was not a "pro forma" pleading, as they did not

simply allow for contribution from the other defendants "if the plaintiff made out its case against them", as the judge stated, but also adopted the plaintiffs' allegations of negligence against the other defendants. Moreover, Western went further at trial and submitted in closing that Prévost failed to satisfy a duty to warn. [186] There is merit to some of the appellant's submission. I agree that the response of Western and Mr. Spittal cannot be accurately described as "pro forma" because their admission of liability was limited: they continued to deny the "particular allegations of negligence" as to how the collision occurred and they did "repeat and adopt" the allegations of negligence against the other defendants as set out in the NOCC: 1. The Defendants Western and Spittal admit for this proceeding only that the accident referred to in the Second Amended Consolidated Notice of Civil Claim was caused by the negligence of Spittal but deny the particular allegations of negligence or that the Plaintiff suffered the injury, loss or damage as alleged.

2. In further answer to the Second Amended Consolidated Notice of Civil Claim the Defendants Western and Spittal say if the collision occurred as alleged and that if the Plaintiff has suffered injury, loss or damage as alleged or at all, all of which is not admitted but specifically denied, then the Defendants Western and Spittal say that at all material times the injury, loss or damage suffered by the Plaintiff was caused or contributed to by the negligence of the other named Defendants herein, jointly and/or severally, and the Defendants Western and Spittal repeat and adopt the allegations of negligence against the other named Defendants as alleged in Part 1 and Part 3 of the Second Amended Consolidated Notice of Civil Claim. [Emphasis added.] [187] I do not, as the judge did, understand the appellant's argument to go so far as to suggest that any pleading filed by an unsuccessful defendant alleging responsibility of other defendants is enough to justify a Sanderson or Bullock order. All will depend on the particular circumstances. I agree with the judge to the extent that generally, a simple pleading allowing for contribution from other defendants, should the plaintiff succeed against them, would not in itself justify this kind of order. [188] Here, however, Western and Mr. Spittal limited their admission of liability, not by blaming others for the accident itself, but by

disputing the specific allegations of how it occurred -- i.e., the fatigue issue -- which allegations were interconnected to the negligence claims against Canam and Universal. And although Western had interests that were not aligned with the plaintiffs in respect of those claims against Canam and Universal, it stood to gain significantly if Prévost were to be found liable. [189] Therefore, the fact that Western and Mr. Spittal adopted the plaintiffs' allegations of negligence, and further argued that Prévost should be liable for failure to warn, could be considered as factors favouring an order requiring them to pay the successful defendants' costs. [190] All that said, I do not agree with the appellant that the judge erred in principle in not following this Court's guidance in Provost. That case involved a rather complex interplay among three actions and there were numerous factors specific to those actions that cumulatively supported the Sanderson orders. While the trial judge may have given Provost short shrift to some extent, he did not err in distinguishing it at least in part on the basis that Bolton, as the unsuccessful defendant in that case, had denied causing the accident itself, which was not the case here. Nor did his refusal to make the order sought offend the principles that guide the exercise of discretion under the second part of the applicable test. [191] This

was a case where the plaintiffs, notwithstanding the admission of liability on the part of the defendants who were primarily responsible for the bus accident, chose to proceed to a lengthy trial against numerous other defendants in respect of different causes of action arising from the fact of the accident. The trial judge considered these claims to be weak, particularly the negligent sub-contracting claims against Canam and Universal: [42] While the accident happened due to lack of attention, there was no evidence that the lack of attention was caused by fatigue. There was no evidence that Spittal drove more than the allowable hours. There was no evidence of insufficient training by Western. There was no evidence that Western was negligent in hiring Mr. Spittal. There was no case law to support the proposition that Universal or Canam had a duty to go behind the license of Western and make inquiries as to its safety programme. Moreover, even if there was such a duty it flows from what I have said that the inquiry would not have discovered anything untoward. [43] With respect to failure to warn, the standard in Canada was not to have seatbelts in highway coaches. There was no industry or other standard with respect to providing warnings regarding seatbelts. [44] Even assuming that there was a duty to warn, the only defendant that could have given a meaningful warning - namely a

warning in advance of the tour - was Canam, with whom the plaintiffs booked the tour. Universal had no relationship with the passengers and did not own the bus. Prévost and Western, not having any relationship with the plaintiffs, could only have placed a warning sign in the bus itself, by which time the passengers would have been on it or boarding. [45] The case against Prévost with respect to lack of seatbelts was marginally stronger. [46] There was no case to be made against Prévost with respect to glass, because the only credible engineering evidence was that laminated glass instead of tempered glass would have made no difference to the injuries of the passengers. [192] The trial also focused largely on the liability of Prévost, which raised discrete issues that did not involve Western and the other defendants. [193] These circumstances were rather unique, and the trial judge was familiar with the entire context. It is plain that he did not consider it just and fair that Western and Mr. Spittal bear the costs of all the other defendants. Therefore, despite the nature of Western's pleadings, I see no reversible error in the judge's exercise of his discretion in refusing to make a Sanderson or Bullock order.

2. Failing to award trial costs against Western [194] As the trial judge clearly recognized in his reasons, the usual rule is that costs follow the event, absent special considerations. As provided in Rule 14-1(9), costs must be awarded to the successful party unless the court orders otherwise. The successful party "is the plaintiff who establishes liability under a cause of action and obtains a remedy, or a defendant who obtains a dismissal of the plaintiff's case": Loft at para.

46. There is a substantial onus on a party who seeks to have the court depart from the usual rule, as a successful litigant has a reasonable expectation of being awarded her costs: Briante at para. 198; Sutherland v. The Attorney General of Canada, 2008 BCCA 27 at para. 26. [195] The exercise of discretion in this context is somewhat narrower than the "true discretion" exercised in assessing the fairness of a Sanderson or Bullock order. There must be a sound basis on which to deny a successful plaintiff her costs. Here, the plaintiffs were successful in their negligence claims against Western and Mr. Spittal. The admission of liability cannot detract from that success, especially given Western's position that its liability was never seriously in question. Nor can the failure to prove the precise nature of Mr. Spittal's negligence detract from the impact of the admission. The appellant was a plaintiff who established liability against Western and Mr. Spittal in negligence and she is entitled to a remedy. [196] The appellant submits that the trial judge's failure to give proper effect to the assertion of Western and Mr. Spittal attributing fault for her losses to the other defendants also tainted his decision to deprive her of her costs of trial. She says the fact that Western and Mr. Spittal maintained this assertion required the trial to proceed against the remaining

defendants: Ms. Ding could not simply have accepted Western and Mr. Spittal's admission and abandoned her claims against the other defendants (especially where, as here, Western and Mr. Spittal continued to allege contributory negligence, which would sever joint liability if established). [197] The appellant contends the judge erred in failing to consider Western and Mr. Spittal's position as a relevant factor, as well as the fact that she was not found to have prolonged the case unnecessarily in respect of the fatigue and safety issues. She also contends that the precise cause of Mr. Spittal's negligence is not relevant given the admission of liability, and in any event, the judge gave no reason for denying her trial costs for the trial time unrelated to the fatigue and safety issues. [198] Western submits that the trial judge's decision to order the appellant and Western to bear their own trial costs was a reasonable balancing of Western's success in defending the fatigue and safety issues with its late admission of liability. [199] It is difficult to discern from the trial judge's reasons why he considered it appropriate to depart from the usual rule to award trial costs to the plaintiffs as the successful parties vis-à-vis Western and Mr. Spittal. His analysis focuses on Western's application to be awarded costs despite its admission of liability. He found it was reasonable

for Western to have participated in the trial, noted the plaintiffs' lack of success on the fatigue and safety issues and accepted that the plaintiffs were required to prove those issues as against the other defendants, but he did not consider this to mean that Western should bear the costs of that. He also considered the lateness of the admission of liability, which denied the plaintiffs the opportunity to plan accordingly. Then, rather than deciding whether Western was entitled to costs (the question he apparently sought to answer), the judge simply concluded that each of these parties should bear their own costs of the trial. [200] Implicit in this is a conclusion that none of these parties should be entitled to their trial costs. However, absent from the analysis is a clear reason why, despite Western's last-minute admission that still required the plaintiffs to proceed to trial against the other defendants on the fatigue and safety issues, and Western's continued participation in the trial in its own interest, the successful plaintiffs should be denied their costs of the trial as against these unsuccessful defendants. The appellant suggests the judge did so "on the apparent basis" that he found against the plaintiffs on the fatigue and safety issues. [201] If that is the case, the judge failed to recognize that an order under Rule 14‑1(15) to deny a successful party their costs on discrete matters is not a regular part of litigation and should be confined to relatively rare cases: Sutherland at para.

43. Such cases may certainly include those where the court rules against the successful party on an issue that took a discrete amount of time at trial (Loft at para. 49) as well as cases involving some kind of misconduct in the litigation, such as where the successful party prolonged the case unnecessarily on an issue on which they were unsuccessful: Sutherland at paras. 34-36. If the judge was applying the first example, he failed to address why the lack of success on the one issue justified denying all of the plaintiff's trial costs. [202] I agree with the appellant that the judge erred in principle by failing to consider the position of Western and Mr. Spittal as set out in their pleadings as well as the fact that the plaintiffs were not found to have unnecessarily prolonged the trial. These were relevant factors in the analysis. [203] Just as it is an error in principle to deny a successful defendant its costs where some elements of a negligence claim were proven (as in Briante and Brito (Guardian ad litem of) v. Woolley, 2007 BCCA 1), it is an error to deny a successful plaintiff her costs where some aspects of her claim were not proven, absent a sound basis to do so in either case. [204] It follows that I would set aside this order. The question now is whether a different order should be substituted. Is there a sound basis on which to deny the appellant, as the successful

plaintiff, her costs of trial?[3] [205] The plaintiffs succeeded in their negligence action against Western and Mr. Spittal as a result of their admission of liability. They did not succeed in establishing that Mr. Spittal's negligence was due to fatigue, and therefore they did not succeed in their negligence actions against Canam and Universal on the safety issues arising from the allegation of fatigue. However, the plaintiffs' success against Western and Mr. Spittal was not divided. As the trial judge noted, if liability is found against a party for negligence, the court will not parse out which basis of the claim succeeded and apportion costs between it and the unsuccessful bases. On the face of this, the appellant should be entitled to her trial costs against Western and Mr. Spittal. [206] However, as the trial judge quite properly observed, this case did not fit into the normal paradigm. Once Western and Mr. Spittal made the admission of liability, the plaintiffs had to decide whether to proceed to trial against the other defendants. This was a difficult choice to make, especially on the eve of trial. Again, the judge properly observed that the plaintiffs did not have the opportunity to plan their trial strategy given the lateness of the admission. In these circumstances, one cannot fault them for continuing. [207] It is quite clear in the judge's reasons that he

considered the negligent sub-contracting claims against Canam and Universal arising from the fatigue and safety issues to be exceedingly weak. In this circumstance, there is a sound basis for denying the appellant her trial costs in relation to these issues. I see no sound basis, however, to deny the appellant her trial costs for the remainder of the trial. Western considered it in its interest to attend the trial and as I indicated above, it stood to gain if the plaintiffs succeeded in their claims against Prévost due to the claim for contribution. [208] The correct order, in my view, is that the appellant is entitled to her costs of the portion of the trial unrelated to the fatigue and safety issues. The discrete amount of time taken at trial on these issues can be determined by examining the record. [209] I note that the trial judge accepted Western's estimate that 48 of the 72 days of trial concerned Western's liability, but he did so without comment from the plaintiffs and without counting the days himself. The appellant challenges this "finding" as a palpable and overriding error, as she says the great majority of the trial was devoted to the negligent design and failure to warn issues. I would direct the parties to determine the correct amount of time, and failing that, would remit this narrow issue to the trial judge.

3. Uplift costs [210] Increased, or uplift costs are governed by s. 2(5) and (6) of Appendix B of the Supreme Court Civil Rules: (5) If, after it fixes the scale of costs applicable to a proceeding under subsection (1) or (4), the court finds that, as a result of unusual circumstances, an award of costs on that scale would be grossly inadequate or unjust, the court may order that the value for each unit allowed for that proceeding, or for any step in that proceeding, be 1.5 times the value that would otherwise apply to a unit in that scale under section 3 (1). (6) For the purposes of subsection (5) of this section, an award of costs is not grossly inadequate or unjust merely because there is a difference between the actual legal expenses of a party and the costs to which that party would be entitled under the scale of costs fixed under subsection (1) or (4). [211] The exercise of discretion to award uplift costs requires two things: (1) unusual circumstances, that would result in (2) a grossly inadequate or unjust award of costs at the fixed scale. This is necessarily a fact-based inquiry driven by the nature of the litigation and the conduct of the parties: Herbison v. Canada (Attorney General), 2014 BCCA 461 at para. 42; Shen at para.

31. An award of uplift costs is a discretionary costs order that attracts a high degree of deference. [212] A party's conduct can constitute unusual circumstances if it is deserving of some form of rebuke (less than that required for special costs), but there are numerous circumstances that may be "unusual" within the meaning of s. 2(5). These include (a) misconduct by the unsuccessful party; (b) the serious nature of the allegations; (c) the complexity or difficulty of the issues in the litigation; and (d) the importance of the litigation to the parties or to the development of the law. These last two factors overlap with some of the factors that are considered in determining the appropriate scale of costs: British Columbia v. Adamson, 2017 BCSC 168 at para. 53, citing International Energy and Mineral Resources Investment (Hong Kong) Company Limited v. Mosquito Consolidated Gold Mines Limited, 2012 BCSC 1475 at paras. 23-24; Shen at paras. 32-35. [213] Whatever constitutes the unusual circumstances, uplift costs are not intended to punish the unsuccessful party but rather to indemnify the successful party: Shen at para. 30. [214] The appellant does not challenge the trial judge's finding that the importance of the case to Prévost and the coach industry, and the complexity of the case against Prévost, constituted unusual circumstances. She says the judge made no finding that these circumstances would result in a grossly inadequate or unjust award at the fixed scale. She submits that he thus failed to consider a required element of the test, thereby committing an error of law and principle. [215] Prévost submits that the judge referred to the correct law and expressly found that three circumstances "justify an award of uplift costs to Prévost": at para.

65. It says that inherent in this statement is a finding that an award of ordinary costs would be grossly inadequate or unjust. [216] The judge's reasons are brief. He did not expressly find that the unusual circumstances would result in a grossly inadequate or unjust award. However, the record before us, which includes affidavit evidence and extensive written submissions by both Prévost and the plaintiffs on both Scale C and uplift costs, demonstrates that these elements were squarely before the judge. The affidavit evidence filed on behalf of Prévost provided considerable detail about the work involved in preparing and running the trial, and included a statement that the full amount of a costs award at Scale C was "only a fraction of the actual costs that Prévost incurred" in doing so. [217] Although the judge ought to have provided clearer reasons, in light of the record, there is no question he was fully aware that increased costs may only be awarded where the unusual circumstances result in a grossly inadequate or unjust award of costs at the fixed scale. I am therefore satisfied that he applied the correct principles in concluding that the circumstances were sufficient to "justify an award of uplift costs to Prévost" and I would not accede to this ground of appeal. D. Conclusion and disposition [218] For all of these reasons, I would dismiss the appeal of the judge's

order dismissing the action against Prévost and Canam. I would allow the appeal of the costs order only to the extent of setting aside the order that each of Western, Mr. Spittal and the appellant bear their own costs of the trial. I would substitute an order that the appellant is entitled to her costs of the trial other than the time taken to address the fatigue and safety issues. "The Honourable Madam Justice Fisher" I agree: "The Honourable Chief Justice Marchand" I agree: "The Honourable Justice Skolrood" [1] The appeal from the dismissal of the claim for failure to warn included Universal but the appellant abandoned her appeal against Universal after these parties reached a settlement. [2] In a Sanderson order, the court orders an unsuccessful defendant to pay the costs of a successful defendant directly; in a Bullock order, the court orders the costs to be paid indirectly to the plaintiff as a disbursement. [3] Although the Ruling on Costs was made in respect of both plaintiffs, the order under appeal applies only to the appellant, Ms. Ding.