Zsoldos et al. v. Canadian Pacific Railway Company et al.

[Indexed as: Zsoldos v. Canadian Pacific Railway Co.]

93 O.R. (3d) 321

Court of Appeal for Ontario,

Rosenberg, Gillese and Blair JJ.A.

January 21, 2009

Torts -- Negligence -- Contributory negligence -- Plaintiff seriously injured when his motorcycle struck train at rural railway crossing at night -- Plaintiff impaired and driving too fast -- Trial judge finding that accident would have occurred whether or not plaintiff was impaired -- Trial judge erring in relying on seatbelt cases in finding plaintiff to be 25 per cent at fault but finding of contributory negligence based on plaintiff's speed being appropriate.

Torts -- Negligence -- Standard of care -- Railway crossing -- Plaintiff seriously injured when his motorcycle struck train at rural railway crossing at night -- Crossing not having active warning system -- Defendant railway company not meeting standard of care as it failed to carry out nighttime inspections of crossing -- Nighttime inspection would have made it clear that additional protection was necessary -- Technology which would have made crossing safer existing at time of accident and defendant not meeting standard of care in failing to employ it -- Defendant negligent despite its compliance with regulatory framework.

The plaintiff was seriously injured in 1994 when his motorcycle struck a train at a rural railway crossing at night. There was an advance warning sign 290 feet before the crossing, which alerted motorists to the presence of the crossing and advised them to reduce their speed. There was no active warning system such as lights, bells or gates to warn a motorist that a train was in the crossing, and there was no artificial illumination of the crossing by street lights. The road approached the tracks at an acute angle of 45 degrees, and sightlines could be obscured by crops. The plaintiff did not reduce his speed. His blood alcohol content at the time of the collision was just below the legal limit. The plaintiff brought an action against the railway company for damages for negligence. The trial judge noted that the applicable government regulations did not mandate any level of warning beyond the passive system in place at the crossing. She concluded, however, that given the amount of discretion the defendant had in determining what safety measures should be established at level crossings, mere compliance with the regulatory framework did not necessarily relieve the defendant of liability. She found that the plaintiff could not have stopped before the crossing, even had he begun to reduce his speed at the advance warning sign. She found that the defendant was negligent in failing to carry out any inspections of the crossing at night and in making no attempt to determine whether the passive system was sufficient to protect motorists using the crossing at night. She further found that had the defendant conducted a nighttime inspection, it would have concluded that additional protection was necessary, the technology for which clearly existed and the cost of which would not be an impediment. She found the defendant 75 per cent negligent. Although she found that the accident would have occurred even if the plaintiff were not impaired, she concluded that his impairment and the fact that he was driving too quickly given the dark conditions both contributed to the accident. She found the plaintiff 25 per cent contributorily negligent. The defendant appealed.

Held, the appeal should be dismissed. [page322]

It was open to the trial judge to find that the defendant's failure to carry out nighttime inspections of the crossing was negligent. While the defendant did not fall below industry standards, even a common practice may be negligent if it is fraught with obvious risk. The failure to carry out nighttime inspections was fraught with obvious risk. It was also open to the trial judge to find that the defendant's failure to take additional measures constituted negligence. She was not required to deal with each proposed measure individually and consider the feasibility, including cost, of each measure. It was sufficient that she found that measures were available in 1994 that a reasonable railway operator could have employed to meet the standard of care.

The trial judge did not fail to properly address the issue of causation. She found that with additional protection, the plaintiff would have received sufficient warning of the presence of the train in the crossing, thereby allowing him to avoid the collision. There was evidence to support that finding.

The trial judge erred in fixing the amount of contributory negligence on the basis of an analogy to seatbelt cases. She did not identify any precautionary measures that the plaintiff should have taken that would have reduced the severity of his injuries. The only conduct by the plaintiff that contributed to the accident was excessive speed. In those circumstances, the 25 per cent assessment of responsibility to the plaintiff may have been somewhat excessive, but there was no cross-appeal by the plaintiff.

APPEAL by the defendant from the judgment of Rady J., 2007 7583 (ON SC), [2007] O.J. No. 942, 46 C.C.L.I. (4th) 294 (S.C.J.) in an action for damages for negligence. [page323]

Cases referred to Duddle v. Vernon (City), [2004] B.C.J. No. 1430, 2004 BCCA 390, 201 B.C.A.C. 153, 35 B.C.L.R. (4th) 24, 49 M.P.L.R. (3d) 233, 132 A.C.W.S. (3d) 578; Empire Co. v. Sheppard, [2001] N.J. No. 35, 2001 NFCA 10, 198 Nfld. & P.E.I.R. 53, 103 A.C.W.S. (3d) 436; Housen v. Nikolaisen, [2002] 2 S.C.R. 235, [2002] S.C.J. No. 31, 2002 SCC 33, 211 D.L.R. (4th) 577, 286 N.R. 1, [2002] 7 W.W.R. 1, J.E. 2002-617, 219 Sask. R. 1, 10 C.C.L.T. (3d) 157, 30 M.P.L.R. (3d) 1, 112 A.C.W.S. (3d) 991; Kauffman v. Toronto Transit Commission, 1960 4 (SCC), [1960] S.C.R. 251, [1960] S.C.J. No. 5, 22 D.L.R. (2d) 97; Resurfice Corp. v. Hanke, [2007] 1 S.C.R. 333, [2007] S.C.J. No. 7, 2007 SCC 7, 278 D.L.R. (4th) 643, 357 N.R. 175, [2007] 4 W.W.R. 1, J.E. 2007-333, 69 Alta. L.R. (4th) 1, 404 A.R. 333, [2007] R.R.A. 1, 45 C.C.L.T. (3d) 1, 153 A.C.W.S. (3d) 1012, EYB 2007-113553; Ryan v. Victoria (City), 1999 706 (SCC), [1999] 1 S.C.R. 201, [1999] S.C.J. No. 7, 168 D.L.R. (4th) 513, 234 N.R. 201, [1999] 6 W.W.R. 61, J.E. 99-357, 117 B.C.A.C. 103, 59 B.C.L.R. (3d) 81, 44 C.C.L.T. (2d) 1, 50 M.P.L.R. (2d) 1, 40 M.V.R. (3d) 1, 85 A.C.W.S. (3d) 208; Snushall v. Fulsang (2005), 2005 34561 (ON CA), 78 O.R. (3d) 142, [2005] O.J. No. 4069, 258 D.L.R. (4th) 425, 202 O.A.C. 297, 36 C.C.L.T. (3d) 63, 25 M.V.R. (5th) 180, 142 A.C.W.S. (3d) 533 (C.A.); ter Neuzen v. Korn, 1995 72 (SCC), [1995] 3 S.C.R. 674, [1995] S.C.J. No. 79, 127 D.L.R. (4th) 577, 188 N.R. 161, [1995] 10 W.W.R. 1, J.E. 95-1970, 64 B.C.A.C. 241, 11 B.C.L.R. (3d) 201, 58 A.C.W.S. (3d) 336 Authorities referred to Fleming, John G., The Law of Torts, 7th ed. (Sydney: Law Book Co. Ltd., 1987) Osborne, Philip, The Law of Torts, 3rd ed. (Toronto: Irwin Law, 2007) Fridman, Gerald H.L., The Law of Torts in Canada, 2nd ed. (Toronto: Carswell, 2002) Klar, Lewis N., Tort Law, 3rd ed. (Toronto: Carswell, 2003)

John A. Campion, Steven F. Rosenhek and Sophie Vlahakis, for appellants. Jim Virtue, James J. Mays and Emily Foreman, for respondents.

The judgment of the court was delivered by

[1] ROSENBERG J.A.: -- On August 26, 1994, at about 9:15 p.m., the respondent Jason Zsoldos was driving home on his motorcycle when he collided with a Canadian Pacific freight train. While the respondent has no memory of the collision, it is apparent that he did not see or hear the cars of the train until it was too late to avoid a collision. He slid under the train and as a result lost both arms and a leg. At the trial before Rady J., the only issue was liability. The trial judge found the appellants CP and Gordon Gosnell, the engineer, 75 per cent negligent and assessed the respondent's contributory negligence at 25 per cent.

[2] The appellants' principal ground of appeal is that the trial judge erred in finding that CP fell below the standard of care without identifying what that standard of care required it to do. The appellants also argue that this error infected the trial judge's finding on causation. Finally, the appellants submit that the trial judge erred in only finding the respondent 25 per cent contributorily negligent.

[3] For the following reasons, I would dismiss the appeal.

The Facts

[4] The respondent was 22 years of age at the time of the collision. He lived with his family in a rural area near the village of Bothwell in Lambton County, about 100 metres from the CP rail line. He was well familiar with the line, crossing it daily during the day and at night where it crossed McCready Road, a gravel road running north/south. McCready Road is straight and flat and is bordered by agricultural fields. There was a maturing cornfield on the east side of the road and another unknown crop on the west side. The speed limit on McCready Road is 80 kilometres per hour.

[5] The CP level crossing on McCready Road is similar to thousands of rural crossings in Ontario and Canada. The crossing is equipped with two cross bucks that bear reflectorized tape on the front only. The cross buck on the south side, the direction from which the respondent was approaching, is 15 feet south of the south rail. There is also an advance warning sign 290 feet before the crossing. This sign, erected by the municipality, alerts motorists to the presence of the railway crossing and advises motorists to reduce their speed to 20 kilometres per hour as they approach the crossing. This is a passive system. There are no active warning [page324] systems such as lights, bells or gates to warn a motorist that a train is in the crossing. There is no artificial illumination of the crossing by street lights.

[6] At the time of the collision, CP ran six to ten trains daily on this line with four or five running at night. They did not move on a fixed schedule. The train with which the respondent collided consisted of three engines and 58 freight cars. Most of the cars were of United States origin and did not have reflectorized markings on them. The train was over a kilometre long and the respondent collided with the 35th car. Thus, the lead engine with its headlights and whistle had long passed the crossing at the time of the collision. The train had recently left a siding and was therefore only coming up to speed as it reached the crossing. The 35th car was travelling at about 20-25 miles per hour at the time.

[7] The respondent had been drinking beer with some friends before he set off home at approximately 9:10 p.m. The trial judge found that the respondent had a blood alcohol level just below the legal limit of 80 milligrams of alcohol in 100 millilitres of blood at the time of the collision. The respondent was travelling at the 80 km/h speed limit and did not reduce his speed as he approached the crossing. The night was clear and warm. The sun had set by the time the respondent left on his motorcycle. The moon was not yet up. The headlight on the motorcycle was on, and the respondent was probably using the high beams.

[8] As indicated, the Bothwell crossing is a typical rural crossing in the sense that there is no active warning system such as bells, flashing lights or gates. It is unusual, however, in at least one respect: the road approaches the tracks at an acute angle of 45 degrees, rather than head-on at a 90 degree angle. That is not to say that there are not many other rural crossings with this problem, but it was a particular problem at this crossing. It was for that reason that the warning sign was erected advising, but not requiring, motorists to reduce speed to 20 kilometres per hour. Another problem at this particular crossing at the time of year when this accident occurred is, as indicated above, the presence of crops, particularly corn crops that can obscure the sightlines.

[9] The respondent introduced important evidence through Curtis Scherer, an expert in accident reconstruction, as to when a person driving a motorcycle would be able to see a freight train travelling through the Bothwell crossing at night. The respondent also relied upon the evidence of Professor Alison Smiley, an expert in ergonomics, the study of human factors to reduce errors and accidents. The consensus of Mr. Scherer and Professor Smiley seemed to be that a motorcyclist would not have been able to see the train until he was within 80 metres of the crossing even if the [page325] person was expecting to see a train. The distance would increase when other variables were taken into account, especially that most motorists would not be expecting to find a train in the crossing.

[10] The respondent also adduced evidence as to what speed the motorist would have to be travelling in order to avoid the collision, given the time necessary to react (perception reaction time, or PRT) and the time needed to bring the motorcycle to a stop. The respondent's witnesses developed a number of scenarios, using PRTs of 1.5 seconds and 2.5 seconds. A PRT of 1.5 is based on a relatively young person driving in the day. Consumption of alcohol would increase the PRT. A PRT of 2.5 is a conservative figure used by highway designers. In short, using a PRT of 1.5, the cyclist would have to be travelling at between 60 and 66 kilometres per hour or less to avoid an accident. Using a PRT of 2.5, which is the standard used by highway designers, the person would have to be travelling at 57.9 kilometres per hour or less to be able to stop in time. The trial judge found that the respondent's PRT was 2.5 seconds given his consumption of alcohol.

The evidence on standard of care

[11] CP conceded that it owed the respondent a duty of care. The principal issue was the standard of care and whether CP fell below that standard of care in the manner in which it maintained this particular crossing. CP and the respondent called employees and former employees of Transport Canada. CP called its manager of public works. The respondent, as indicated, called Mr. Scherer and Professor Smiley.

[12] This crossing was opened to rail traffic in 1889 and is located on a single-track freight railway line near the village of Bothwell. Prior to the incident involving the respondent, there were no reported accidents at this crossing. A term of art in the industry is the cross-product, being the number of vehicles travelling over the crossing daily multiplied by the number of trains. The cross-product at the Bothwell crossing was 300 in 1994. The trial judge found that there was no regulation or government order that governed when an active warning system had to be installed. That decision was left to the discretion of the railway companies and the municipalities with input from Transport Canada. Transport Canada would not have recommended an active warning system for this type of crossing where the cross-product was less than 1000.

[13] I will have more to say about industry practice and available technology in considering the grounds of appeal. [page326]

The Trial Judge's Reasons

[14] After a review of the evidence, the trial judge turned to the law. She began with the leading case on railway negligence, Ryan v. Victoria (City), 1999 706 (SCC), [1999] 1 S.C.R. 201, [1999] S.C.J. No. 7. She correctly set out the law as established in Ryan, to which I will make further reference. She found that the applicable government regulations did not mandate any level of warning beyond the passive system in place at the Bothwell crossing. She concluded, however, that given the amount of discretion CP had in determining what safety measures should be established at level crossings, mere compliance with the regulatory framework did not necessarily relieve CP of liability.

[15] The trial judge also appears to have concluded that industry practice at the time would not have required anything more than the passive system. While Transport Canada had circulated draft guidelines the year before this accident that might have required enhanced measures, the trial judge appears to have accepted that those guidelines did not reflect industry practice.

[16] The trial judge reviewed at length the expert evidence adduced by the respondent centring around the conditions that the respondent would have faced on the night of August 26, 1994. She also considered the expert evidence concerning braking distances, reaction times, perception and driver behaviour. As I read her reasons, the trial judge accepted the expert evidence that the respondent would have had a low expectation of a train being in the crossing and that the speed advisory sign would have had little or no impact on a driver's choice of speed. Most significantly, the trial judge [at para. 81] appears to have accepted as a fact that "a driver with average perception abilities, traveling faster than 50 kilometres per hour and using low beams could not have stopped in time to avoid a train". Further, "The same result occurred if high beams were used with the motorist travelling at slightly in excess of 60 kilometres per hour."

[17] Accordingly, the trial judge concluded [at para. 83] that "the crossing as it was configured the night of the accident, with only the passive warning and speed advisory alert presented an objective risk of harm to users of the highway". While she could have been clearer, it seems that the trial judge also accepted that the risk was reasonably foreseeable given the number of fatalities and injuries at passive crossings.

[18] The trial judge then reviewed the steps taken by CP to deal with this risk. She noted that the Transport Canada inspector had only inspected the crossing during the day, although he [page327] claimed that he inspected the crossing from the perspective of nighttime motorists. She also noted that a CP employee admitted that the lack of any lighting was a risk factor for nighttime accidents and that there was no record of CP ever having reviewed the crossing to ensure that a motorist could safely determine if a train is in the crossing. She also noted that the appellants' witnesses accepted that the cross buck does not warn a motorist that a train is in or about to enter the crossing.

[19] The trial judge also considered what technology was available in 1994 to enhance the safety at night at the Bothwell crossing. She found that a variety of technology existed, an issue to which I will return below, since this finding lies at the heart of the appellants' appeal.

[20] Finally, the trial judge accepted [at para. 106] the evidence from the appellants' witnesses that "cost is no object in eliminating or reducing danger at a crossing".

[21] The trial judge then made her findings of fact, most of which are reflected in the statement of facts set out above. The critical findings are that the respondent had no expectation that a train would be in the crossing and there were no visual clues that a train was in the crossing, given the darkness of the night, the lack of lighting in the area and that the train cars themselves were dark. As well, the crops in the field, especially the corn crop, obscured the train's presence to motorists. The respondent could not hear the train because he was wearing his motorcycle helmet and was driving on a gravel road. The trial judge also found that the respondent's perception reaction time was 2.5 seconds. Younger persons tend to have a lower PRT, but consumption of alcohol increases PRT. The trial judge then made this critical finding of fact [at para. 115]:

Given my findings of the speed at which Mr. Zsoldos was traveling and his PRT, it is apparent that he could not have stopped before the crossing, even had he begun to reduce his speed at the advance warning sign.

[22] The trial judge then found that CP was negligent in failing to carry out any inspections of the Bothwell crossing at night and in making no attempt to determine whether the passive system was sufficient to protect motorists using the crossing at night. She concluded [at para. 117] that "it is almost impossible to accurately assess the safety of a crossing at night when such an inspection is carried out in the bright light of day". She further found that had CP conducted the inspections that it should have, it would have concluded that additional warnings were required, "given the steep angle at which the [page328] track bisected the roadway, the absence of any ambient lighting and the presence of crops on the side of the roadway, all of which conspired to make the presence of a train in the crossing very difficult to discern". As well, the advance warning sign erected by the municipality was situated too close to the track to allow a motorist sufficient time to react and slow in time to stop and it was advisory only. CP would know that such signs are frequently ignored by motorists. She concluded her finding of negligence with the following summary [at para. 120]:

In summary, therefore, if CP had undertaken an inspection of this crossing at night, it would have concluded that some additional protection was necessary, the technology for which clearly existed and the cost of which would not be an impediment, according to Mr. Churcher and Mr. Lukianow [the appellants' witnesses]. With additional protection, Mr. Zsoldos would have received sufficient warning of the presence of the train in the crossing, thereby allowing him to avoid the collision.

[23] The trial judge turned to the question of contributory negligence and accepted the respondent's expert's evidence as to the respondent's blood alcohol level at the time of the accident. She found that the respondent had a BAC close to but below the legal limit of 80 mg/dl. Thus, there was [at para. 137] "no question. . . that his ability to operate the motorcycle was compromised to some extent". She concluded, however, that the accident would have occurred even if the respondent were not impaired. However, his impairment and the fact that he was driving too quickly given the very dark conditions both contributed to the accident. Using the analogy of contributory negligence where a motorist fails to wear a seatbelt, the trial judge considered the respondent 25 per cent contributorily negligent.

The Issues

[24] The issues on this appeal are as follows: (1) Did the trial judge err in finding that CP fell below the standard of care without identifying the required standard of care and, in particular, in failing to identify what was reasonable in the circumstances? (2) If the trial judge erred in failing to identify the standard of care, did she necessarily err in finding that any failure to act on the part of CP caused the accident? (3) Did the trial judge err in her assessment of contributory negligence and, in particular, in relying upon the seatbelt analogy? [page329]

Analysis

(1) The standard of care

[25] In Ryan, the Supreme Court of Canada abandoned the special rule that placed railway companies in a privileged position within the law of negligence. The special rule was that so long as railway companies complied with government regulations and orders, absent extraordinary circumstances, they were under no further obligation to act in an objectively reasonable manner. Henceforth, railway companies were subject to the same standard of care as other similarly situated defendants. That does not mean that the regulatory framework is irrelevant. As Major J. explained, at para. 40, "Where a statute authorizes certain activities and strictly defines the manner of performance and the precautions to be taken, it is more likely to be found that compliance with the statute constitutes reasonable care and that no additional measures are required." He also held, however, that where the regulatory framework is general or permits discretion as to the manner of performance, mere compliance with the statutory framework "is unlikely to exhaust the standard of care".

[26] The trial judge found that CP had a broad discretion when it came to determining the safety measures at level crossings. This finding was amply supported by the evidence, especially CP's own witnesses who testified that the thrust of pre-1994 deregulation amendments to the statutory framework were to shift responsibility to the railway companies. I agree with the trial judge that CP's compliance with the statutory framework did not exhaust the standard of care.

[27] CP, however, also relies upon industry standards. It notes the trial judge's finding that industry standards did not require anything more than passive measures for a rural crossing like the Bothwell crossing, one of 16,000 similar crossings throughout the country. In my view, the evidence of industry standards and its application to this case must be approached with caution. While there was some evidence of studies and reports from Transport Canada and from the United States, industry standard on this record was little more than an assertion by CP, supported by the Transport Canada witnesses, that given the cross-product and the conditions at rural crossings, passive measures were an adequate response to the risk. As counsel for the respondent put it, CP virtually was the industry.

[28] There are two aspects to the finding of negligence in this case: the failure to inspect the crossing at night and the failure to take some steps to warn motorists travelling at night that a train [page330] is in the crossing. It seems to me that these two issues need to be addressed separately.

[29] I can find nothing in the trial judge's reasons to indicate that she found that CP fell below industry standards. I will therefore address the two issues on that basis.

[30] The decision of the Supreme Court in ter Neuzen v. Korn, 1995 72 (SCC), [1995] 3 S.C.R. 674, [1995] S.C.J. No. 79, 127 D.L.R. (4th) 577 deals with the question of standard of care in the face of an assertion that the defendant met the industry standard. In that case, at para. 39, Sopinka J. adopted an excerpt from Professor Fleming's treatise The Law of Torts, 7th ed. (Sydney: Law Book Co. Ltd., 1987), at p. 109:

Conformity with general practice, on the other hand, usually dispels a charge of negligence. It tends to show what others in the same "business" considered sufficient, that the defendant could not have learnt how to avoid the accident by the example of others, that most probably no other practical precautions could have been taken, and that the impact of an adverse judgment (especially in cases involving industry or a profession) will be industry-wide and thus assume the function of a "test case". Finally, it underlies the need for caution against passing too cavalierly upon the conduct and decision of experts.

All the same, even a common practice may itself be condemned as negligent if fraught with obvious risk. (Emphasis added)

[31] In my view, it was open to the trial judge to find that CP's failure to carry out nighttime inspections of this rural crossing was negligent. While the trial judge did not put it in these terms, her findings are reasonably open to the conclusion that failure to carry out nighttime inspections was fraught with obvious risk. This, it seems to me, is the only reasonable inference from para. 117 of her reasons:

In my view, in failing to carry out any inspections of this crossing at night, CP could not determine whether the use of a reflectorized cross buck alone was a sufficient protection for users of the roadway. This constitutes a failure to take reasonable care in the circumstances. I have not overlooked Mr. Stephens' [the Transport Canada inspector] evidence that he attempted to inspect a crossing from the perspective of a nighttime motorist. While I do not doubt his sincerity on this issue, it seems to me that it is almost impossible to accurately assess the safety of a crossing at night when such an inspection is carried out in the bright light of day.

[32] Further, I did not understand the appellants to be contesting this finding by the trial judge. They argue only that the failure to inspect at night could not have been the operative act of negligence unless it was "also proven that the inspection would have resulted in a change to the proximate loss-causing act or omission". As the appellants correctly point out, the trial judge had to ask the additional question of whether the measures already in place were reasonable in all the circumstances of the [page331] case. However, factors such as nighttime inspection and additional measures are not watertight compartments. The trial judge's finding was that had CP met the standard of care and carried out the necessary inspection of the Bothwell crossing, it would have realized that additional measures were required. To apply Professor Fleming's observation, had CP done what it should have, it would have realized, and any reasonable railway operator would have realized, that the crossing was fraught with obvious risks at night. In my view, that is the import of the trial judge's finding at para. 118 of the reasons:

If CP had carried out an inspection of this crossing at night, it would likely have concluded that additional warnings were necessary, given the steep angle at which the track bisected the roadway, the absence of any ambient lighting and the presence of crops on the side of the roadway, all of which conspired to make the presence of a train in the crossing very difficult to discern.

[33] Since the appellants have not identified any error of law by the trial judge in her conclusion that the failure to carry out the nighttime inspection was negligent, that finding is entitled to considerable deference: Housen v. Nikolaisen, 2002 SCC 33, [2002] 2 S.C.R. 235, [2002] S.C.J. No. 31, 211 D.L.R. (4th) 577, at paras. 26-33. Since the appellants have not pointed to any palpable and overriding error in this aspect of the trial judge's finding of negligence, that finding stands.

[34] The next issue is whether CP's failure to take additional measures constituted negligence. Justice Major summed up the test for negligence in these terms in Ryan, at para. 28:

Conduct is negligent if it creates an objectively unreasonable risk of harm. To avoid liability, a person must exercise the standard of care that would be expected of an ordinary, reasonable and prudent person in the same circumstances. The measure of what is reasonable depends on the facts of each case, including the likelihood of a known or foreseeable harm, the gravity of that harm, and the burden or cost which would be incurred to prevent the injury. In addition, one may look to external indicators of reasonable conduct, such as custom, industry practice, and statutory or regulatory standards.

[35] The appellants attack the trial judge's finding of negligence in two respects. They submit that she erred in law in failing to consider whether the measures already in place were reasonable in all the circumstances and in failing to identify what additional protection was reasonably feasible in the circumstances. They submit that the trial judge's conclusory statement to the effect that "some additional protection was necessary" is not sufficient. The appellants submit that each of the proposed measures had to be evaluated from the point of view of its technological feasibility, cost and efficacy. Otherwise, the trial judge was fixing CP with a standard of perfection, of doing whatever was necessary to avoid [page332] the accident, which would in effect make CP an insurer: see Kauffman v. Toronto Transit Commission, 1960 4 (SCC), [1960] S.C.R. 251, [1960] S.C.J. No. 5, at p. 255 S.C.R.; Empire Co. v. Sheppard, [2001] N.J. No. 35, 2001 NFCA 10, at para. 32; and Duddle v. Vernon (City), [2004] B.C.J. No. 1430, 2004 BCCA 390, at para. 22.

[36] It seems to me that the evidence concerning the cross- product is central to CP's submission. In effect, it asks the court to endorse its economic analysis of negligence. It has, through the vehicle of the cross-product, weighed the three factors identified by Major J. in Ryan of the likelihood of a known or foreseeable harm, the gravity of that harm and the burden or cost which would be incurred to prevent the injury. It has decided that absent exceptional circumstances, with certain low cross-products, as at the Bothwell rural crossing, the risk of harm was a reasonable one to take. However, the trial judge was not bound by CP's economic analysis. The trial judge was obliged to consider this particular crossing and decide whether because of the conditions at this particular crossing it was unreasonable for CP to only use a passive warning system at this unilluminated crossing. She was also entitled to take into account other evidence adduced by the respondents, especially the uncontradicted evidence from Professor Smiley, as to the efficacy of speed advisory signs.

[37] The trial judge's reasons could have been clearer on this aspect of the case. Although the trial judge conducted an admirable review of much of the evidence and the applicable law, her findings on this important aspect of the case are somewhat conclusory. Nonetheless, the trial judge did address the first two Ryan factors concerning the likelihood of harm and the gravity of that harm. She noted, at para. 88, the evidence of a significant number of fatalities and injuries at passive crossings.

[38] The trial judge also addressed the issue of the burden or cost that would be incurred to avoid the injury, albeit in a somewhat conclusory form, at paras. 105 and 120 of her reasons:

I do not propose to summarize all of the evidence on the issue because it is clear that a variety of technology existed to provide warning that a train was occupying the intersection, including flashing lights, bells and gates, street lighting and reflectorization of train cars and the back of cross bucks. It must be noted, however, that some of this technology was in its infancy in 1994. For example, CP has recently begun to reflectorize the backs of cross bucks. This creates a strobe effect when an approaching car's headlights catch the reflector as the train passes through the crossing. It does not appear to have been in use in 1994, however. . . . . .

In summary, therefore, if CP had undertaken an inspection of this crossing at night, it would have concluded that some additional protection was [page333] necessary, the technology for which clearly existed and the cost of which would not be an impediment, according to Mr. Churcher and Mr. Lukianow. With additional protection, Mr. Zsoldos would have received sufficient warning of the presence of the train in the crossing, thereby allowing him to avoid the collision.

[39] The appellants submit that this was not sufficient, that the trial judge had to deal with each proposed measure individually, and consider the feasibility, including cost, of each measure. I do not agree. In my view, it was sufficient that the trial judge found that measures were available in 1994 that a reasonable railway operator could have employed to meet the standard of care. It was not for the trial judge to dictate which of the proposed measures CP should adopt. It seems to me that for the appellants to succeed on this aspect of the case, they must show that the trial judge's finding was unreasonable; that no jury properly instructed could find that any of the measures were reasonable in the circumstances.

[40] In my view, there was evidence from which the trial judge could find that at least some of the measures identified in para. 105 of her reasons were known to the industry in 1994, and could have been adopted by a reasonable and prudent operator in the circumstances. The central problem identified by the trial judge, which is fully supported by the evidence, is that at night it was impossible for a motorist to know that a freight train was in the crossing until it was too late to stop. The evidence accepted by the trial judge was that the measures taken by CP, at this crossing, were not sufficient. There would be a major benefit if freight cars bore reflective tape, but many freight cars come from the United States, where there is no requirement for reflectorization. The cross bucks, with reflecting tape or paint only on the front, warned of the presence of a railway crossing, but did not warn that there was a train in the crossing. The advisory sign erected by the municipality was similarly insufficient. The expert evidence adduced by the respondents and accepted by the trial judge was that an experienced railway operator would know that motorists, especially those familiar with the crossing, do not reduce their speed in response to such advisory signage. The trial judge also found that the sign was placed too close to the tracks to allow the motorist sufficient time to react and slow in order to stop before the crossing.

[41] In Ryan, Major J. held that in addition to considering the likelihood of a known or foreseeable harm, the gravity of that harm and the burden or cost, the courts may look to external indicators of reasonable conduct, such as custom, industry practice and statutory or regulatory standards.

[42] The appellants submit that industry practice demonstrates conclusively that an active warning system, such as flashing lights [page334] and gates, was not reasonable. While the respondents point out that in other nearby locations, lights and gates are used on this very same line, in my view, the trial judge found that this evidence was not helpful in identifying reasonable measures because of the different circumstances at those crossings. However, at least two passive measures, illumination of the crossing by a street light and reflectorization of the back of the cross bucks, were reasonable and known to the industry in 1994.

[43] Overhead lighting, referred to at trial as luminaries, was not in use anywhere in Canada at rural crossings. The appellants, therefore, seemed to have met industry standards in that respect. But, failure to illuminate this crossing was fraught with danger. Conformity with standard practice does not insulate CP from negligence when that practice itself is negligent. The appellants' own witness, Clarke Stephens, an employee of Transport Canada, testified that even back in 1993 "illumination of night time trains should be considered for existing crossings where a train visibility is a problem". Mr. Stephens' evidence focused on problems associated with highway approach gradients or curvature that "do not allow a vehicle or headlights to illuminate the grade crossing before the stopping distance".

[44] However, on the trial judge's findings, in the absence of any other measure, failure to illuminate the Bothwell crossing was negligent because a motorist travelling at the expected speed would not have been able to stop in time. CP did not recognize the problem because they were negligent in failing to carry out any nighttime inspections of the Bothwell crossing. A trier of fact could draw the conclusion that failure to illuminate the crossing in some fashion fell below a reasonable standard of care even without expert evidence since "the practice does not conform with basic care which is easily understood by the ordinary person who has no particular expertise": ter Neuzen, at para. 43.

[45] The evidence concerning the reflectorization of the backs of the cross bucks is less satisfactory. The point of putting reflector tape on the back of the cross bucks is that it produces a strobe effect as a vehicle's headlights approach a crossing occupied by a moving train. The respondents adduced evidence that reflectorization was in use in some states in the United States. On the other hand, the evidence was also relatively clear that this technology was not in use in Canada and the respondents did not adduce any evidence to the contrary. In view of the feasibility of overhead lighting, I need not pursue this issue further.

[46] The appellants also submit that the trial judge misapprehended the evidence with respect to the cost of remedial measures. The trial judge said the following, at paras. 106 and 120: [page335]

Both Mr. Churcher and Mr. Lukianow testified that cost is no object in eliminating or reducing danger at a crossing. . . . . .

In summary, therefore, if CP had undertaken an inspection of this crossing at night, it would have concluded that some additional protection was necessary, the technology for which clearly existed and the cost of which would not be an impediment, according to Mr. Churcher and Mr. Lukianow.

[47] In my view, this was a reasonable interpretation of the evidence. For example, Mr. Lukianow, a manager with CP, testified as follows:

Q. Now would you agree with me that if a danger is recognized at an intersection that the costs to protect the public and railway employees from that danger really should not be a factor?

A. Correct.

[48] Similarly, Mr. Churcher, the employee of Transport Canada testified as follows:

Q. All right, would you -- would it be fair to say that the tension point in developing these regulations is whether or not the risk at a particular crossing warrants an expense at the crossing?

A. We, in Transport Canada at that time never considered the cost of, of improving a crossing. We only looked at what had to be done to make it safe.

(2) Causation

[49] The appellants submit that the trial judge failed to properly address the issue of causation. This submission is tied to the submission concerning the standard of care and can therefore be briefly dealt with. CP submits that because the trial judge never identified the standard of care, she could not find that if it met the standard of care the accident would have been avoided. The trial judge addressed causation, at para. 120 of her reasons, as follows:

With additional protection, Mr. Zsoldos would have received sufficient warning of the presence of the train in the crossing, thereby allowing him to avoid the collision.

[50] There was circumstantial evidence to support this finding. The respondent adduced evidence from Professor Smiley and Mr. Scherer concerning reaction times, braking distances and stopping distances, where the motorist was entirely reliant upon the vehicle's headlights. In particular, Professor Smiley's evidence, summarized by the trial judge at paras. 74 and 75, supported her conclusion. The trial judge said this:

She testified that in this case because of the rural setting, in the absence of any natural or artificial lighting, the steep angle of the crossing and the [page336] presence of foliage obscuring the train's silhouette, a motorist would be very reliant on his headlights to see the train in the crossing. Her opinion was largely, but not exclusively, predicated on an assumption that Mr. Zsoldos was using his low beams.

She noted that the train engine's headlights would have travelled through the crossing well before they might otherwise have assisted Mr. Zsoldos to discern the train's presence in the crossing. There was approximately 2200 feet in distance between the engine and the 35th car. (Emphasis added)

[51] Professor Smiley testified as follows:

Q. Is there any consideration or have -- has your group given any thought to placing light standards or just standard lights in the intersection?

A. Ah, that I think -- another issue there is cost and that's not something that we're looking at within our study. That's not something we chose to focus on.

Q. Okay, is that something that would have any benefit in a situation like Jason Zsoldos' situation?

A. Yeah, most certainly. If the train was illuminated in some way that would be very helpful.

Q. Okay and just so we can make sure we understand the point why would that be helpful?

A. Ah, simply because the -- there would be an additional source of illumination right at the crossing itself, illuminating the train and you wouldn't be relying on getting close enough with your headlights that the headlights would cast light on the train. You would be able to see that train before the -- you got close to the crossing. . . . . .

Q. All right, so if this court finds that Jason Zsoldos was in probability going faster than -- or going 80 kilometres or faster over that or through that intersection would he have had an opportunity to perceive and react in time to stop to avoid the collision.

A. No, no he would not.

Q. If -- at high beam or low beam?

A. High beam or low beam. (Emphasis added)

[52] It was open to the trial judge, based on this evidence to draw the common sense inference that had the crossing been illuminated, the accident could have been avoided.

(3) Contributory negligence

[53] The appellants submit that the trial judge erred in her assessment of contributory negligence and, in particular, in relying upon the analogy to seatbelt cases. The trial judge's findings [page337] concerning contributory negligence are found in paras. 138-39 of her reasons:

. . . Mr. Zsoldos' level of impairment was one factor that contributed to the accident that occurred, but it was not the only one. The accident would have occurred whether Mr. Zsoldos was impaired or not, given CP's failure to take reasonable care to protect motorists approaching this crossing. I have also concluded that Mr. Zsoldos was driving too quickly in the circumstances, given the very dark conditions of that evening.

In assessing the level of Mr. Zsoldos' contribution, I consider him to be 25% responsible. This is consistent with findings of contributory negligence in cases where an injured motorist has failed to wear a properly adjusted seatbelt.

[54] Contributory negligence can arise in three ways. First, the plaintiff's negligence may have been a cause of the accident in the sense that his acts or omissions contributed to the sequence of events leading to the accident. Second, although the plaintiff's negligence is not a cause of the accident, the plaintiff has put himself in a position of foreseeable harm. Third, as in seatbelt cases, the plaintiff may fail to take precautionary measures in the face of foreseeable danger: see P.H. Osborne, The Law of Torts, 3rd ed. (Toronto: Irwin Law, 2007), at pp. 104-105, and Lewis N. Klar, Tort Law, 3rd ed. (Toronto: Carswell, 2003), at p. 457.

[55] I agree with the appellants that the trial judge erred in fixing the amount of contributory negligence on the basis of an analogy to the seatbelt cases such as Snushall v. Fulsang (2005), 2005 34561 (ON CA), 78 O.R. (3d) 142, [2005] O.J. No. 4069 (C.A.). She did not identify any precautionary measures that the respondent should have taken that would have reduced the severity of his injuries.

[56] That, however, does not end the inquiry. The issue is whether the error had any effect on the verdict. A plaintiff will be found contributorily negligent only where his or her negligence caused the accident or contributed to the severity of his injury. It is not necessary that the plaintiff's negligence be the only cause, but it must have been a proximate or effective cause. Thus, if the plaintiff was negligent but that negligence did not in fact cause or contribute to the causation of the accident or damage to the plaintiff, it is not contributorily negligent. The burden was on the appellants to prove the necessary causation: see G.H.L. Fridman, The Law of Torts in Canada, 2nd ed. (Toronto: Carswell, 2002), at p. 473. The basic test for determining causation is the "but for" test: Resurfice Corp. v. Hanke, 2007 SCC 7, [2007] 1 S.C.R. 333, [2007] S.C.J. No. 7, at para. 21. There is no suggestion that the "but for" test could not be applied in this case.

[57] The appellants focused their attention primarily on the evidence of impairment. They submit that "driving while intoxicated is itself a proximate cause of the collision" and unlike failure to [page338] wear a seatbelt does not merely aggravate the injuries. They submit that the conventional cap of 25 per cent in seatbelt cases therefore has no application to this case. The appellants have assembled a number of cases suggesting that percentages in excess of 50 per cent have been applied where the plaintiff was intoxicated.

[58] The difficulty with applying those cases in this case is the trial judge's finding that the accident would have occurred whether or not the respondent was impaired. This suggests that the appellant's impairment, which was below .08, was not a cause of the accident. This is consistent with the expert evidence of Professor Smiley and Mr. Scherer that even using a PRT of 1.5 seconds, if the respondent was travelling in excess of 66 kilometres per hour, the collision could not have been avoided.

[59] Thus, the only conduct by the respondent that contributed to the accident was excessive speed or, as it was put by the trial judge [at para. 138], that the respondent "was driving too quickly in the circumstances, given the very dark conditions of that evening". In those circumstances, the 25 per cent assessment of responsibility to the respondent may have been somewhat excessive, but there was no cross-appeal by the respondent. It follows that I would not give effect to this ground of appeal.

Disposition

[60] Accordingly, I would dismiss the appeal with costs fixed at $50,000 inclusive of GST and disbursements.

Appeal dismissed.