Smith v. Safranyos

Citation: 2018 ONCA 760 | 143 O.R. (3d) 22

Court: Court of Appeal for Ontario

Judges: Strathy C.J.O., L.B. Roberts and Paciocco JJ.A.

Date: September 19, 2018

Case Summary

Highways — Non-repair — A driver failed to yield the right-of-way upon entering a highway from a local road and was T-boned by another vehicle. The trial judge found the driver negligent and the city liable for non-repair on grounds that there was no stop line on the local road at the intersection and that sightlines for vehicles approaching the intersection were not appropriate. The city's appeal from a finding that it was 25 per cent responsible for the accident was dismissed. The driver's negligence and breach of s. 136 of the Highway Traffic Act did not bar a finding of non-repair if the road was not in a sufficient state of repair to be safe for drivers exercising reasonable care. The trial judge did not use Ontario Traffic Manual guidelines as mandatory directions prescribing the standard of care.

Torts — Negligence — Causation — The driver who was struck was T-boned by another motorist who had consumed alcohol and was speeding. The trial judge found the motorist 25 per cent responsible for the accident. The motorist's appeal was allowed. The trial judge made a palpable and overriding error in finding that the motorist could have avoided the collision if he had been driving prudently.

Overview

[1] On June 16, 2007, at 1:03 a.m., a horrendous two-car collision occurred in the City of Hamilton at the intersection of Upper Centennial Parkway and Green Mountain Road. A vehicle operated by Ms. Dawn Safranyos, carrying four children, failed to yield the right-of-way upon entering Upper Centennial Parkway, a through highway, and was "T-boned" at highway speed by a vehicle operated by Mr. Daryl McHugh. Mr. McHugh had consumed alcohol and was exceeding the 70 kilometre per hour speed limit by at least 15 kilometres per hour. The occupants in Ms. Safranyos's vehicle were all seriously injured.

[2] Lawsuits were brought on behalf of the injured children and their family members against Ms. Safranyos, Mr. McHugh and the City of Hamilton ("Hamilton"). The basis for the claims against Ms. Safranyos and Mr. McHugh related to their alleged negligent manner of driving that played a role in causing the collision.

[3] The claims against Hamilton were based on the allegation that non-repair of the intersection, given its design and condition, was a cause of the accident.

[4] The claims against all three defendants were tried together, but the trial was bifurcated. Only the liability trial has so far been concluded. A damages trial has yet to be held.

[5] The trial judge found each of the defendants liable. She apportioned liability at 50 per cent to Ms. Safranyos, and 25 per cent to each of Mr. McHugh and Hamilton.

[6] Ms. Safranyos has not appealed the findings of liability or apportionment that affect her. Both Hamilton and Mr. McHugh appeal the findings of liability against them.

[7] I would dismiss Hamilton's appeal. The trial judge applied the correct legal test in finding Hamilton to be liable for non-repair under the Municipal Act, 2001, S.O. 2001, c. 25, s. 44 ("Municipal Act"). She was also entitled, on the evidence before her, to find that this particular intersection was in a state of non-repair.

[8] However, notwithstanding the deferential standard of review that applies to determinations of fact, I would allow Mr. McHugh's appeal. In my view, the trial judge committed palpable and overriding errors in finding him liable. I would dismiss the action against him, as I see no evidence upon which he could properly be found liable.

Background

[9] Upper Centennial Parkway is an arterial road running north and south with two lanes in each direction. Green Mountain Road is a local road with one lane in each direction. It runs east and west. The two roads intersect at a four-way intersection. In June 2007, stop signs only affected traffic entering the intersection from Green Mountain Road. There were no stop signs or traffic lights affecting drivers along Upper Centennial Parkway at the intersection. A drive-in theatre is located on the north side of Green Mountain Road, east of Upper Centennial Parkway.

[10] On June 15, 2007, Ms. Safranyos drove the four minor plaintiffs to the drive-in theatre to watch two movies. They arrived at approximately 8:00 p.m. and left at approximately 1:00 a.m. on June 16, 2007, after the second movie ended. Ms. Safranyos turned onto Green Mountain Road, heading west towards Upper Centennial Parkway.

[11] Wayne Jackson, a witness at trial who also attended the drive-in theatre that night, left the theatre at about the same time as Ms. Safranyos. He also intended to proceed through the Upper Centennial Parkway and Green Mountain Road intersection that night. His car was behind Ms. Safranyos's car, separated by one white BMW sedan, at the time of the accident. Mr. Jackson could look over the BMW and see Ms. Safranyos's car because he was driving a taller Chevrolet Suburban SUV.

[12] At the same time, Mr. McHugh was driving northbound along Upper Centennial Parkway towards the intersection with Green Mountain Road in a Chevrolet Cavalier. Earlier that night, Mr. McHugh had attended a house party and consumed four to six beers between 9:45 p.m. and 11:30 p.m. Mr. McHugh was driving alone. His friend, Dillon Baker, who also attended the same house party that night, was driving behind him in a separate car. They left the house party at approximately the same time. It was unclear how far behind Mr. Baker was from Mr. McHugh -- but he was not close enough to witness the accident. Mr. Baker, however, did arrive on the scene of the accident shortly after the accident occurred.

[13] The speed limit on Upper Centennial Parkway heading northbound towards Green Mountain Road varies. The speed limit is 80 kilometres per hour for some distance, but 80 metres from the intersection, it drops down to 70 kilometres per hour. Mr. McHugh was driving at least 85 kilometres per hour as he approached and entered the intersection that night.

[14] There was evidence before the trial judge that according to the Ontario Traffic Manual ("OTM"), a Transport Association of Canada guideline for designing roads and road markings used by the City of Hamilton, given the speed limit on Upper Centennial Highway an intersection warning sign should be located 335 metres from the intersection. The warning sign facing northbound traffic, including Mr. McHugh, was 125 metres from the intersection.

[15] At approximately 1:03 a.m., Ms. Safranyos signalled to turn left to head southbound onto Upper Centennial Parkway. She stopped at the stop sign controlling westbound traffic on Green Mountain Road. She looked left once, concluded that it was safe to proceed and then rolled towards the intersection. She did not stop at the intersection nor look left again to see that it was clear to continue. Ms. Safranyos accelerated into the intersection and was struck by Mr. McHugh.

[16] Mr. McHugh testified that he was about three car lengths, or 18 metres, away from the intersection when Ms. Safranyos accelerated into his lane. Mr. Jackson testified that Mr. McHugh's vehicle "came out of nowhere". The trial judge found that Mr. McHugh was at least 125 metres away when Ms. Safranyos stopped at the stop sign.

The City of Hamilton Appeal

A. Additional Relevant Facts and Reasons for Judgment

[17] The non-repair claims against Hamilton, based on the statutory tort set out under the Municipal Act, s. 44, relate to the design and condition of the east side of the intersection as it existed on June 16, 2007.

[18] The westbound stop sign Ms. Safranyos encountered on Green Mountain Road as she approached Upper Centennial Parkway was set ten metres back from the eastern edge of the intersection. The stop line that was once painted at the eastern edge of the intersection was removed in a "shave and pave" in 2004 and had not been repainted by June 2007.

[19] There was evidence before the trial judge that in 2000, the City of Hamilton prepared a pavement marking design for that intersection calling for a painted stop line on the eastern edge of the intersection, but not for the western edge of the intersection. There was also evidence presented that there are no records indicating that Hamilton reconsidered that design when they moved to an "as required" stop line policy in 2001, or after the "shave and pave" in 2004.

[20] There is also a steel guardrail on the east side of Upper Centennial Parkway, south of Green Mountain Road, running up to the intersection. In addition, there are elevation changes, as Green Mountain Road rises to connect with Upper Centennial Parkway. Mr. McHugh was driving beside that guardrail as he approached the intersection, going northbound in the curb lane of Upper Centennial Parkway. There was evidence before the trial judge that, given the location of the guardrail and the elevation changes, approximately 25 metres from the intersection the headlights of oncoming northbound compact vehicles driving in the curb lane, such as Mr. McHugh's Cavalier, can disappear from the view of westbound drivers such as Ms. Safranyos. The evidence was also to the effect that it is 25 metres from the intersection that drivers approaching an intersection will tend to look to their left for oncoming traffic.

[21] In finding that Hamilton failed to keep the highway in proper repair, contrary to s. 44, the trial judge featured two specific but related shortcomings. First, she concluded that to be in a reasonable state of repair, Green Mountain Road required a painted stop line where it met Upper Centennial Parkway. Second, the sightlines for vehicles approaching the intersection were not appropriate and could not meet the reasonable state of repair standard. She held that together "these breaches were causes" of the accident, and she rejected Hamilton's attempt to defend the condition of the intersection.

[22] As I detail below, in my view the trial judge was entitled to make a finding of non-repair relating to this specific intersection, in all of the circumstances. As indicated, the trial judge focused on two separate instances of non-repair. The evidence supporting the stop line non-repair is solid, but the foundation for finding non-repair based on the sightlines is not as concrete. It is clear from her decision, however, that the trial judge found that the sightline problems she identified enhanced the risks that the absence of a stop line presented to drivers exercising ordinary reasonable care. Put otherwise, although she featured these two issues in her decision, at base her finding was that the combined effect of these and other related factors rose to a level that goes beyond municipal discretion in intersection design and road markings, amounting to non-repair.

B. Hamilton Appeal -- Issues

[23] Hamilton alleges that the trial judge committed errors of law, and errors of mixed fact and law. I have reordered Hamilton's grounds of appeal for convenience in this analysis:

(1) Did the trial judge err in law by misapplying the reasonable driver standard when determining whether the intersection was in non-repair?

(2) Did the trial judge err in law in finding that the absence of a stop line constituted a non-repair by

(a) treating Ontario Traffic Manual guidelines as legally enforceable standards of civil liability; or

(b) improperly drawing adverse inferences and using Hamilton's policies in finding non-repair?

(3) Did the trial judge commit a palpable and overriding error in finding that the sightlines constituted a non-repair?

(4) Did the trial judge commit a palpable and overriding error in finding a causal link between the intersection's non-repair and the accident?

[24] I will describe the legal test for non-repair before analyzing these grounds of appeal, in turn.

C. Hamilton Appeal -- Analysis

[25] The Municipal Act, s. 44, sets out the statutory obligation Hamilton was found to have breached. It provides:

Maintenance

44(1) The municipality that has jurisdiction over a highway or bridge shall keep it in a state of repair that is reasonable in the circumstances, including the character and location of the highway or bridge.

Liability

(2) A municipality that defaults in complying with subsection (1) is, subject to the Negligence Act, liable for all damages any person sustains because of the default.

Defence

(3) Despite subsection (2), a municipality is not liable for failing to keep a highway or bridge in a reasonable state of repair if,

(a) it did not know and could not reasonably have been expected to have known about the state of repair of the highway or bridge;

(b) it took reasonable steps to prevent the default from arising; or

(c) at the time the cause of action arose, minimum standards established under subsection (4) applied to the highway or bridge and to the alleged default and those standards have been met.

Regulations

(4) The Minister of Transportation may make regulations establishing minimum standards of repair for highways and bridges or any class of them.

General or specific

(5) The minimum standards may be general or specific in their application.

Adoption by reference

(6) A regulation made under subsection (4) may adopt by reference, in whole or in part, with such changes as the Minister of Transportation considers desirable, any code, standard or guideline, as it reads at the time the regulation is made or as it is amended from time to time, whether before or after the regulation is made.

[26] In Fordham (Litigation guardian of) v. Dutton-Dunwich (Municipality), [2014] O.J. No. 5938, 2014 ONCA 891, at para. 26, Laskin J.A. set out the four-step test that is to be applied in analyzing whether a municipality is liable under s. 44 of the Municipal Act:

1. Non-repair: The plaintiff must prove on a balance of probabilities that the municipality failed to keep the road in question in a reasonable state of repair.

2. Causation: The plaintiff must prove the "non-repair" caused the accident.

3. Statutory Defences: Proof of "non-repair" and causation establish a prima facie case of liability against a municipality. The municipality then has the onus of establishing that at least one of the three defences in s. 44(3) applies.

4. Contributory Negligence: A municipality that cannot establish any of the three defences in s. 44(3) will be found liable. The municipality can, however, show the plaintiff's driving caused or contributed to the plaintiff's injuries.

[27] The first two steps of the four-step test -- (1) determining whether the plaintiff has first proved non-repair and (2) causation -- resolve whether there is a prima facie case of liability.

[28] If a prima facie case of liability is established, step three requires the municipality to establish, on a balance of probabilities, any statutory defences outlined in s. 44(3) that it seeks to rely on.

[29] If no statutory defences apply, then at step four, the municipality is entitled to attempt to show a plaintiff's contributory negligence. This fourth step is not material here, since no allegation of contributory negligence is made against the infant plaintiffs.

[30] Hamilton, however, claims that the trial judge committed errors of law with respect to her analysis of the first three steps.

(1) Did the trial judge misapply the test for non-repair?

[31] "Non-repair" will be established if the plaintiff proves "on a balance of probabilities that the municipality failed to keep the road in question in a reasonable state of repair": Fordham, at para. 26. The applicable legal test is, "was the road at the material time sufficiently in repair that those users of the road, exercising ordinary or reasonable care, could use it in safety?": Deering v. Scugog (Township), [2010] O.J. No. 4229, 2010 ONSC 5502 (S.C.J.), at para. 100, affd [2012] O.J. No. 2546, 2012 ONCA 386. In adopting the Deering standard of care test, Laskin J.A. elaborated in Fordham, at para. 28, that "[o]rdinary reasonable drivers are not perfect drivers; they make mistakes", but he cautioned, at para. 29, "a municipality's duty of reasonable repair does not extend to making its roads safer for negligent drivers".

[32] There is no question that Ms. Safranyos was a negligent driver, and that she contravened s. 136(1) of the Highway Traffic Act, R.S.O. 1990, c. H.8. Subsection 136(1) provides:

136(1) Every driver or street car operator approaching a stop sign at an intersection,

(a) shall stop his or her vehicle or street car at a marked stop line or, if none, then immediately before entering the nearest crosswalk or, if none, then immediately before entering the intersection; and

(b) shall yield the right of way to traffic in the intersection or approaching the intersection on another highway so closely that to proceed would constitute an immediate hazard and, having so yielded the right of way, may proceed.

[33] Building on this, Hamilton argues that the trial judge erred in law in misapplying the non-repair standard. It says that the trial judge ignored both of Laskin J.A.'s admonitions. It urges that the trial judge found Hamilton liable because the road was not safe for negligent drivers such as Ms. Safranyos, and in finding that Ms. Safranyos was merely mistaken when she failed to stop at the entrance to the intersection after the stop sign.

[34] I do not accept either argument.

[35] The fact that Ms. Safranyos was negligent is not a bar to a non-repair finding. A non-repair action can succeed even where a negligent driver was the immediate cause of the accident. The courts in both Deering and Housen v. Nikolaisen, [2002] 2 S.C.R. 235, [2002] S.C.J. No. 31, 2002 SCC 33 found municipalities to be liable to negligent drivers for non-repair. As long as road conditions that would imperil ordinary drivers constitute a "but for" cause of the accident, a liability finding against a municipality may be appropriate.

[36] The trial judge clearly understood this:

-- She described the law correctly, that the "state of the repair of the road hinges on an ordinary driver using reasonable care".

-- When making her non-repair findings she said on three separate occasions that the absence of a stop line and the impaired sightlines posed an unreasonable risk of harm to reasonable drivers.

-- She relied in her reasons on the impact that the alleged non-repairs would have for drivers generally, including

-- that the absence of a stop line created "ambiguity as to where a driver was required to stop" and "deceived drivers"; and

-- that the visibility of vehicles on Upper Centennial Parkway to drivers on Green Mountain Road was partially obstructed.

-- She noted the failure of Hamilton to follow its own stop line policies.

-- She noted that all but one of the accidents that have occurred at the intersection involved drivers turning from Green Mountain Road.

[37] Therefore, the trial judge did not misunderstand the law. The gauge she used was the proper one of whether the road was in a sufficient state of repair to be safe for drivers exercising ordinary or reasonable care.

[38] Hamilton argues that even if she stated the law correctly, she applied it incorrectly. Hamilton contends that where there is a stop sign, a driver of ordinary reasonable care would comply with s. 136(1) and stop immediately again before entering the intersection, whether or not there is a painted stop line. In other words, any driver failing to stop immediately before entering the intersection is not a driver of ordinary reasonable care.

[39] In support of this proposition, Hamilton cites Kennerley v. Norfolk (County), [2005] O.J. No. 4782, 16 M.P.L.R. (4th) 286 (S.C.J.), at para. 14, which described the failure to stop immediately before entering an intersection as reckless conduct, or an act of folly. Hamilton also quotes the dictum from Fordham, at para. 37, to the effect that "drivers who ignore important traffic signals, such as stop signs, are not driving with reasonable care"; and Chaschuk v. Lebel, [1981] O.J. No. 157, 12 M.V.R. 228 (C.A.), at p. 232 M.V.R., declaring that it may be negligent to miss a visible traffic signal that has only momentarily changed.

[40] Hamilton then argues that since it is always negligent to fail to stop immediately before entering an intersection marked with a stop sign, only negligent drivers are put at risk by the absence of a painted stop line. Therefore, a finding of non-repair based on the absence of a stop line is inappropriate given the Fordham decision.

[41] In my view, Hamilton's argument is untenable. It would immunize municipalities from s. 44 liability, regardless of the character or quality of their stop signs and markings.

[42] Its reasoning is also too simplistic. The question under s. 44 is the fact specific one of whether the road in question was sufficiently in repair at the material time that users exercising ordinary or reasonable care could use it in safety. Simply put, if an intersection is rendered unsafe for ordinary or reasonable drivers in whole or in part because of the placement or condition of stop signs and road markings, a municipality cannot avoid liability by relying on Highway Traffic Act, s. 136(1).

[43] The law does not say otherwise. In none of the cases relied upon by Hamilton was there any issue with the integrity or adequacy of stop signals. In this case there was, and the trial judge applied the correct test for non-repair in coming to the decision she did.

[44] The one place in her decision where the trial judge did exhibit confusion about the ordinary driver standard did not prejudice Hamilton. Specifically, after reviewing non-repair decisions, the trial judge commented that "[b]ased on the case law, Dawn Safranyos ought to be held to the standard of the ordinary driver using ordinary care". This does not reflect the law of non-repair. Given the trial judge's conclusion that the conditions said to constitute non-repair posed an unreasonable risk to ordinary drivers, whether Ms. Safranyos met the standards of the ordinary driver using reasonable care was immaterial on the question of non-repair. She did not have to satisfy the standard of an ordinary driver using reasonable care for Hamilton to be liable so long as its non-repair was a "but for" cause of the accident and posed an unreasonable safety risk to drivers who exercised reasonable care.

[45] For this reason, the trial judge's problematic characterization that Ms. Safranyos was "mistaken" in failing to stop is immaterial to the outcome of this appeal. Whether Ms. Safranyos was "mistaken" as the trial judge said here, or negligent as the trial judge said elsewhere, does not matter to Hamilton's non-repair liability. Ms. Safranyos's negligence only goes to the apportionment of liability.

[46] I would therefore not give effect to Hamilton's claim that the trial judge applied the improper legal test for non-repair.

(2) Did the trial judge err in law in finding that the absence of a stop line constituted a non-repair?

[47] Hamilton argues that the trial judge committed two legal errors in finding that the absence of a stop line constituted a non-repair. First, Hamilton argues that the trial judge erred by treating the OTM as imposing a mandatory rule when the OTM provides guidelines, not standards of civil liability: Fordham, at paras. 51-53; and Greenhalgh v. Douro-Dummer (Township), [2009] O.J. No. 5438, at paras. 66-68, affd [2012] O.J. No. 2012, 2012 ONCA 299](https://www.canlii.org/en/on/onca/doc/2012/2012onca299/2012onca299.html). Second, Hamilton argues that the trial judge erred by relying on adverse inferences in finding a non-repair. In my view, the trial judge committed neither error.

(a) Did the trial judge use OTM guidelines to define the standard of care?

[48] The OTM, Book 11 guidelines entitled "Markings and Delineation", at pp. 11 and 71, provide in relevant part:

The purpose of the Book is to promote uniformity of treatment in the design and installation of markings and delineation throughout Ontario. It is the practitioner's fundamental responsibility to exercise professional engineering judgement on technical matters in the best interests of the public, including safety and cost-effectiveness. The following standard terminology will provide the user of this Book with guidance on what practices are recommended and when judgement can be applied:

Must indicates a mandatory condition. Where "must" is used to described the design or application of the device, it is mandatory that these conditions be met in order to promote uniformity where delineation complements legally enforceable regulations . . .

Should indicates an advisory condition. Where the word "should" is used, the action is recommended but is not mandatory. "Should" is meant to suggest good practice in most situations and to recognize that there may be valid reasons not to take the recommended action.

3.8 Intersections

At both urban and rural intersections, a stop line (also called a stop bar) must be used to indicate the point at which a vehicle must stop in compliance with the STOP sign. A stop line must be a solid white retroreflective line between 30 cm and 60 cm wide. . .

Where there is no pedestrian crosswalk, the stop line must be located between 1.25 m and 3 m upstream of the projected nearside edge of the intersecting road. At STOP signs where visibility is restricted, the stop line should be located so that the driver of the vehicle properly positioned behind the stop line has an adequate view of approaching cross traffic in both directions. The stop line should also be positioned with reference to the clearance needs of cross traffic and pedestrians[.]

(Underlining added)

[49] The trial judge did not use OTM guidelines as if they established Hamilton's standard of care. She did not say that the guidelines were binding. Instead, she stated expressly that the "O.T.M. book 11 is instructive of the standard of care", and emphasized in her concluding remarks that "[t]he court is not simply relying on the O.T.M. in coming to its conclusion. Rather, the O.T.M. is one item combined with other cogent evidence that has guided the court when arriving at its conclusions."

[50] When she noted that, according to OTM standards "a stop line must be used", the trial judge was doing nothing more than what Howden J. did in Deering, at para. 243, by recognizing that where the OTM manual says "must" a court should ask whether there are compelling reasons not to follow that guideline. In Fordham, at paras. 52-53, Laskin J.A. quoted this proposition and explained that it did not apply in the case before him because the material guideline at issue in Fordham used the words "should", not "must".

[51] Nor does the trial judge's reasoning support the claim that she used the guidelines as mandatory directions prescribing the standard of care. The trial judge did not rely solely on the OTM guidelines in coming to her decision. She had other evidence supporting her non-repair finding, including opinions from three expert witnesses (Gerry Forbes, Robert Blankstein and Robert Gilchrist) that a stop line ought to have been used, that the stop line disappeared in a "shave and pave" and was not repainted as it should have been in the ordinary course, as well as the adverse inferences that I am about to address.

(b) Did the trial judge err by improperly using adverse inferences and Hamilton's policies in finding non-repair?

[52] The trial judge used adverse inferences against Hamilton to support her non-repair finding relating to the absence of a stop line. Hamilton failed to call three witnesses who the trial judge concluded would have been uniquely placed to explain whether a stop line was required. These potential witnesses were also uniquely placed to assist in interpreting municipal documentation, including a pavement marking design drawing that made a provision for a stop line at the westbound entrance to the Green Mountain Road and Upper Centennial Parkway intersection, as well as policy documents relating to road markings. One of the potential witnesses, Mr. Hart Solomon, was a traffic engineer who drafted the stop line policy. Another, Mr. Richard Galloway, authored the pavement marking design document. And the third, Mr. Rick Walshaw, was a Hamilton employee who was involved in the preparation of the policy and design documents.

[53] In substance, the adverse inferences the trial judge drew were that the evidence of these witnesses would not have been favourable to Hamilton, and that, in the absence of testimony from these witnesses to the contrary, the policy and road marking documents capable of being interpreted to support the need for a stop line at the intersection should be read as supporting the need for a stop line at the intersection.

[54] Hamilton does not appear to take issue with the trial judge's decision to draw adverse inferences against Hamilton for failing to call these witnesses. It argues, instead, that its policy documents and any adverse inferences were relevant only in evaluating Hamilton's statutory defences (step 3 of the Fordham test), and that the trial judge erred by using adverse inferences in proving a non-repair (step 1 of the Fordham test).

[55] I do not agree. The analytical rigidity Hamilton seeks to impose on the proper use of the adverse inferences available to the trial judge and of Hamilton's policy documentation is not supported by law. The adverse inferences drawn logically supported the non-repair finding and were therefore relevant to it. Similarly, since the trial judge interpreted the policy documents as supporting the need for a stop line at the intersection, those documents were tantamount to admissions by Hamilton to that effect, and therefore relevant in finding non-repair.

[56] I would reject this ground of appeal.

(3) Did the trial judge commit a palpable and overriding error in finding that the sightlines constituted a non-repair?

[57] The trial judge found that the sightlines at the intersection were not appropriate and constituted a state of non-repair. Hamilton argues that the trial judge erred as the conclusion is not supported by the evidence, or adequately explained. Hamilton also identifies what it says are specific palpable and overriding fact finding errors.

[58] The first two objections -- that the trial judge's finding is unsupported and not adequately explained -- can be addressed together. In my view, although the trial judge should have identified her factual sightline conclusions with much greater clarity, her conclusions are nevertheless discernible from the evidence she did accept.

[59] Specifically, the trial judge found that westbound drivers on Green Mountain Road had partially obstructed sightlines of northbound traffic on Upper Centennial Parkway as they approached and arrived at the stop sign. She also found that drivers proceeding northbound on Upper Centennial Parkway would lose sight of westbound vehicles on Green Mountain Road as they were approaching the intersection, until they were 35 metres from the intersection. These findings, and the foundation for them, can be gleaned when reading the reasons for judgment as a whole.

[60] To begin, it is clear that the trial judge found that westbound drivers on Green Mountain Road had unobstructed visibility of northbound vehicles travelling on Upper Centennial Parkway beginning when they were five metres away from the white fog line on the right-hand side of the northbound curb lane on Upper Centennial Parkway. In other words, drivers who would stop before entering the intersection would have no problem seeing northbound traffic. The trial judge said that all of the experts agreed to this and she made no comments in her decision to suggest that she did not accept this evidence.

[61] The sightline impairment evidence the trial judge accepted for drivers on Green Mountain Road relates to a "partially obstructed sightline" as they approach and arrive at the stop sign. She accepted the evidence of Mr. Forbes, who said that the guardrail on Upper Centennial Parkway is a visual obstruction for drivers on Green Mountain Road who are between ten to 25 metres away from the intersection. Mr. Forbes testified that the road design of the intersection required that drivers turning from Green Mountain Road onto Upper Centennial Parkway have a 175-metre unobstructed sight distance. However, a driver stopped at the westbound stop sign on Green Mountain Road would only have a 47-metre sightline before the guardrail became an obstruction.

[62] While the trial judge rejected the personal observations of Mr. Gilchrist, Mr. Gilchrist presented a photograph taken 25 metres from the intersection showing that headlights of northbound vehicles are not visible at that distance because of the guardrail. Thomas Klement, an expert witness called by Hamilton, agreed with this observation and it was evidently accepted by the trial judge. The trial judge also accepted the testimony of Mr. Forbes that the guardrail impaired sightlines for drivers travelling on Upper Centennial Parkway, of vehicles on Green Mountain Road. Instead of having the recommended 85 metres of sightline, Mr. McHugh did not have an unobstructed view of traffic on Green Mountain Road until he was 35 metres from the intersection.

[63] Since these findings and the evidence in their support can fairly be extracted from the reasons, I do not accept Hamilton's claim that the trial judge's finding that the sightline was obstructed is not transparent or supported by the evidence.

[64] Nor do I accept the discrete palpable and overriding errors that Hamilton claims the trial judge made in determining sightlines.

[65] She committed no error in rejecting the testimony of Mr. Klement. She found that Mr. Klement was unreliable because he did not respond to questions and was argumentative. She therefore gave little weight to his sightline evidence. That was her decision to make.

[66] I do not accept Hamilton's claim that the trial judge misapprehended Ms. Safranyos's evidence about what she could see from the stop sign, and that she used Mr. Blankstein's evidence to find impaired sightlines from the stop sign. Ms. Safranyos's evidence that, from the stop sign, she could see lights coming from her left is not inconsistent with the finding of partially obstructed sightlines from the stop sign. Nor is there any indication in the decision that the trial judge relied upon the testimony of Mr. Blankstein in finding partially obstructed sightlines from the stop sign. She relied, as she was entitled to, on Mr. Forbes's evidence.

[67] Controversy nonetheless remains. It is not obvious that the trial judge's partially obstructed sightline findings can properly, on their own, amount to a state of non-repair, as the trial judge found. I would not want to be taken as suggesting that they could on the evidence that was before the trial judge. Even if this is an error, however, it is not overriding because the trial judge also found that the impaired sightlines enhanced the risks posed by the absent stop line at this particular intersection, contributing to the risk this particular intersection presented to drivers exercising ordinary care.

[68] Specifically, the trial judge said:

In summary, with respect to the two "non-repair" items, the stop line and the sightlines only, the lack of the stop line is the most significant "non-repair" in the case at bar. When it is combined with the inappropriate, "non-repair" sightlines the cumulative degree of these "non-repairs" and their risks are elevated.

[69] Elsewhere, the trial judge expressed that changes in elevation also contributed to the risk posed by the absence of a stop line, given the sightline challenges, the placement of the stop sign, and the lighting conditions. She said:

The court finds the fact that the stop line was not present on Green Mountain Road created ambiguity as to where a driver was required to stop. The Court finds that at this particular intersection it is important to have a clear indication of where to bring your vehicle to a stop before entering the intersection. Of note, the elevation of Upper Centennial Parkway is higher than Green Mountain Road. There is a guardrail along the east side of Upper Centennial Parkway that partially obstructs sightlines of the intersection. In addition, there were no street lights along Upper Centennial except at the actual intersection.

[70] In addition to her controversial finding that the sightlines on their own amount to non-repair, the trial judge therefore clearly found that together the sightlines and the absence of a stop line contribute to the non-repair of this particular intersection. She did not purport to make a declaration that a painted stop line is always required, or invariably needed when sightlines are impeded. Nor could she have. Her task was to make a fact-specific assessment on the evidence before her. In my view, there is no basis for interfering with the non-repair conclusion of the trial judge relating to this intersection. It was hers to make.

[71] I would therefore dismiss Hamilton's ground of appeal relating to sightlines.

(4) Did the trial judge err in her causation analysis?

[72] Hamilton argues that the trial judge made palpable and overriding errors in applying the "but for" test of causation. It argues that the only proper inference from the evidence is that the accident would have happened in any event -- regardless of any non-repair problems -- because Ms. Safranyos saw the northbound vehicles yet chose to pull out into the intersection in any event.

[73] I disagree. Ms. Safranyos's evidence is that from the stop sign she could see northbound vehicles "above" the guardrail, and when she looked left from the stop sign she could see a light coming from a great distance away. She did not testify that she saw Mr. McHugh's vehicle approaching and decided to pull out in front of it.

[74] The trial judge had ample evidence that this intersection was confusing, even to drivers of ordinary care. She was entitled, in the circumstances, to accept Ms. Safranyos's evidence that had there been a stop line she would have stopped there, just as she had at the stop sign. There is nothing inconsistent in the trial judge's finding that, despite her negligence, Ms. Safranyos would have stopped had there been a stop line.

[75] I would not give effect to Hamilton's causation appeal.

D. Hamilton Appeal -- Conclusion

[76] For all of these reasons, I would dismiss Hamilton's appeal.

The McHugh Appeal

E. Additional Relevant Facts and Reasons for Judgment

[77] Three sets of factual findings are relevant to Mr. McHugh's appeal. They are the findings related to Ms. Safranyos's liability, Mr. Jackson's evidence and Mr. McHugh's liability. I will discuss each in turn.

(1) Findings Related to Ms. Safranyos's Liability

[78] The trial judge's material findings relating to Ms. Safranyos's role in the accident are brief.

[79] She found that Ms. Safranyos "made a stop at the stop sign located 10 metres back from the intersection, looked to her left [the direction Mr. McHugh was coming from] and did not look to her left again before proceeding into the intersection".

[80] Early in her reasons for judgment, the trial judge recorded her prior ruling under the Evidence Act, R.S.O. 1990, c. E.23, s. 22.1, in which she admitted into evidence "essential facts" derived from Ms. Safranyos's related provincial offence prosecution. Those facts included that Ms. Safranyos was a servient driver, that Mr. McHugh had the right-of-way, and that "[i]n entering the intersection, Safranyos created an immediate hazard".

[81] The trial judge did not refer overtly to this evidence elsewhere in her reasons for judgment, other than to note in her "Conclusion" that "the court considered its previous ruling under s. 22.1 of the Evidence Act".

[82] The trial judge made no other material factual findings relating either to Ms. Safranyos or to her operation of her vehicle before finding her to be 50 per cent responsible for the accident "for her inattention to traffic when she failed to yield the right-of-way and her lack of awareness when she turned onto Upper Centennial Parkway from Green Mountain Road".

(2) Findings Relating to Mr. Jackson's Evidence

[83] Mr. Jackson was the only independent witness to the collision to testify. He had been at the same drive-in theatre that the occupants of Ms. Safranyos's vehicle attended and followed the same route from the drive-in that Ms. Safranyos did. At the material time, Mr. Jackson's vehicle was two vehicles behind Ms. Safranyos's vehicle.

[84] Mr. Jackson described heavy traffic on Green Mountain Road leaving the drive-in theatre, and that cars were "bumper-to-bumper" right up to the Green Mountain Road and Upper Centennial Parkway intersection. He testified at trial that all the cars were, "I guess, stopping at the stop sign at one point because there was so much traffic." However, he also accepted during cross-examination that he made a previous statement in August 2008 that he was "certain [Ms. Safranyos's vehicle] did not come to a full stop at the stop sign before making [its] left-turn, [Ms. Safranyos's vehicle] followed the car ahead of [it] that also made a left-turn".

[85] I believe a fair reading of Mr. Jackson's evidence to be that he was uncertain whether Ms. Safranyos stopped at the stop sign, but the traffic was bumper to bumper at the intersection and he believed that everyone had to at least come to a rolling stop to allow the cars ahead of them to enter Upper Centennial Parkway.

[86] He recalled that Ms. Safranyos's vehicle was signaling to make a left turn as it approached the Upper Centennial Parkway, a turn that would require the vehicle to cross the two northbound lanes of Upper Centennial Parkway before entering the southbound lanes.

[87] Mr. Jackson testified that there was little traffic on Upper Centennial Parkway so everyone was "just going" at the intersection. "So, the car in front of [Ms. Safranyos's vehicle] came to the intersection and then she went, and then right after that car went, [Ms. Safranyos's vehicle] went right, right behind it." He said that Ms. Safranyos's vehicle did not stop at the intersection.

[88] He testified:

She stepped on the gas quite quick to get out into the intersection, and it was almost like she was avoiding something because she went so fast. She slightly turned to the left and then that's when the accident happened, I seen the other car at the last second . . . I only seen it like a split-second. As she was pulling out, I looked to the left and then that's when it happened. Came out of nowhere . . . I would say by the time she hit the intersection where the first lane hits, she was going at least 10, 15 somewhere around there, kilometres.

[89] Yet the trial judge summarized Mr. Jackson's evidence in less than nine lines in her reasons. The only information included in her summary of his evidence that is material to Ms. Safranyos's or Mr. McHugh's role in the collision was this:

He testified that when the Safranyos vehicle pulled out in the intersection he saw the other car at the last second and that the vehicle came out of nowhere from the left.

[90] The trial judge did not refer to Mr. Jackson's other evidence in explaining her conclusions, other than to note that Mr. Jackson saw a vehicle enter the intersection immediately ahead of Ms. Safranyos's vehicle, which Mr. McHugh did not notice.

(3) Findings Relating to Mr. McHugh's Liability

[91] The trial judge found Mr. McHugh to be 25 per cent responsible for the accident, concluding:

His alcohol consumption, which affected his reaction and perception time, his speed, his inattention, and the fact that he had a last clear chance to avoid a collision were causes of this accident.

[92] I pause here to note that Mr. McHugh interprets the trial judge's reference to "last clear chance" as an application of the now rejected, "last clear chance doctrine", "where liability [was] apportioned to a much greater degree to the party who had the last opportunity to avoid the loss": Treaty Group Inc. v. Drake International Inc. (2007), 86 O.R. (3d) 366, [2007] O.J. No. 2468, 2007 ONCA 450, at para. 28. Had the trial judge used "the last clear chance doctrine" instead of the "but for" test of causation re affirmed in Ediger (Guardian ad litem of) v. Johnston, [2013] 2 S.C.R. 98, [2013] S.C.J. No. 18, 2013 SCC 18, at para. 28, I would agree with Mr. McHugh that she had committed an error of law. It seems plain to me, however, that she did not apply the "last clear chance doctrine". Had she applied that doctrine she would have apportioned much greater liability to Mr. McHugh than Ms. Safranyos. In my view, the trial judge instead used the term "last clear chance" when examining whether Mr. McHugh could have avoided the collision if he had been driving in a prudent fashion.

[93] Returning to the trial judge's findings, I recognize that the findings in respect of Mr. McHugh work together. It is nonetheless helpful to identify the trial judge's findings relating to each component of her conclusion, beginning with her findings relating to speed, then alcohol consumption, then inattention, then last clear chance or opportunity to avoid the collision, and finally, reaction and perception time.

(a) Speed

[94] Eighty metres from the location of the accident, the speed limit on the Upper Centennial Parkway dropped from 80 kilometres per hour to 70 kilometres per hour. Mr. McHugh admitted that he did not alter his speed in response. Although his precise speed was never determined, Mr. McHugh admitted that he was travelling at a minimum of 80 kilometres an hour as the accident unfolded. His friend who was driving behind him, Mr. Baker, admitted that they were driving about 5 kilometres per hour over the speed limit along Upper Centennial Parkway. This led the trial judge to conclude that Mr. McHugh was speeding at the time of the accident, "going at least 85 kilometres an hour at the time of impact". She found that this speed was a cause of the accident.

[95] She explained:

If Mr. McHugh was driving the speed limit, in all probability the second half of the Safranyos vehicle would have passed through the curb lane before Mr. McHugh arrived at the intersection. Accordingly, no accident would have happened.

(b) Mr. McHugh's Alcohol Consumption

[96] It was not contested that Mr. McHugh had been at a party with friends, including Mr. Baker, where he consumed alcohol. Mr. McHugh testified that he had two beers at the party. Mr. Baker said it was two or three. The trial judge did not believe this evidence about the amount of Mr. McHugh's alcohol consumption.

[97] Instead, the trial judge accepted the opinion of expert toxicologist George Kupferschmidt that based on subsequent blood alcohol readings taken from Mr. McHugh, his blood alcohol level at the time of the accident, 1:03 a.m., was likely between 66 and 82 milligrams of alcohol in 100 millilitres of blood. This reflected the consumption of between four and six beers. Indeed, this level of alcohol consumption was supported by the four empty Moosehead beer bottles (two of which had Coors Light caps on them) that the police found inside Mr. McHugh's vehicle after the accident.

[98] In her "Findings Based on Analysis", the trial judge said:

It is significant to note that Mr. McHugh would have been impaired according to the Highway Traffic Act of 0.05 blood alcohol level. As well, the high level of the alcohol readings would have seen Mr. McHugh convicted of impaired driving under the Criminal Code had he been charged following this accident.

[99] In her reasons for judgment, the trial judge summarized Mr. Kupferschmidt's evidence to the effect that persons with blood alcohol levels of 0.05 or higher "may be a hazard on the road". She recapped his testimony that such blood alcohol levels may impair faculties required in the operation of a motor vehicle, and there was a "high possibility that a person with a blood alcohol level between 0.05 and 0.1 percent would display impairments".

[100] She made the following related findings in her "Findings Based on Analysis":

Mr. McHugh's alcohol consumption, inability to perceive or react in time to the Safranyos vehicle were clearly factors that partially caused this accident. That is, the court finds it is highly possible that alcohol impaired his abilities in terms of perception and reaction time. This fact may be logically inferred from the evidence. The court finds the plaintiffs have met their burden of proof about this discrete issue.

(c) Inattention

[101] The trial judge did not say a great deal about her "inattention" finding. She did not provide a discrete discussion about inattention. However, as I have already mentioned, she did say that it was "of note" that Mr. McHugh did not see the vehicle that Mr. Jackson described as having pulled onto the road immediately ahead of Ms. Safranyos's vehicle.

(d) Opportunity to Avoid the Collision

[102] The trial judge found that even though Ms. Safranyos's vehicle pulled into his path, Mr. McHugh could have avoided the collision if he had been driving prudently. She based that conclusion largely on the distance she found Mr. McHugh to have been from Ms. Safranyos's vehicle at the material times.

[103] Mr. McHugh estimated that he was eight to ten car lengths away from the intersection when he first saw Ms. Safranyos's vehicle, and three car lengths away when it entered the intersection. Based on the uncontested evidence that a car length is approximately six metres, this meant that Mr. McHugh was 48 to 60 metres away from the intersection when he first saw Ms. Safranyos's vehicle.

[104] The trial judge rejected this evidence. She found that Mr. McHugh was "approximately 125 metres away from the intersection when he first observed the Safranyos vehicle", which looked as if it was stopped at the time. This distance finding was based on Mr. McHugh's testimony that when he returned to the accident scene with his therapist, he estimated that he was near a warning sign when he first saw Ms. Safranyos's vehicle. It was uncontroverted that the warning sign was 125 metres from the centre line of Green Mountain Road at the intersection. Thus, on the trial judge's finding, Mr. McHugh's distance from Ms. Safranyos's vehicle when he first saw it was just under 21 car lengths.

[105] With respect to Mr. McHugh's distance when Ms. Safranyos's vehicle entered the intersection, the trial judge said:

The court does not accept the evidence of Mr. McHugh that he only noticed Ms. Safranyos move from the stop sign when he was just three car lengths away from the intersection. The court finds he underestimated that distance just as he underestimated his consumption of alcohol.

(e) Reaction and Perception Time

[106] The trial judge linked Mr. McHugh's reaction and perception time to his alcohol consumption and speed. Referring to Mr. McHugh's claim that he did not see Ms. Safranyos's vehicle move into the intersection until he was three car lengths away, she said:

If, and I underscore the word if, he did in fact only see her when he was three car lengths as was his evidence, it was because of his alcohol that he had consumed and his excessive speed. These two factors affected his perception of the Safranyos vehicle moving and affected his reaction time. That is, Mr. McHugh did not notice Ms. Safranyos move from a stopped position until it was too late. As Mr. Kupferschmidt indicated, 100 milliseconds of delayed reaction can be the difference of 2.6 seconds.

[107] Earlier in her reasons the trial judge had previously summarized Mr. Kupferschmidt's evidence this way: "Mr. Kupferschmidt indicated 100 milliseconds of delayed reaction can be the difference of 2.6 seconds."

F. McHugh Appeal -- The Issues and the Standard of Review

[108] Mr. McHugh has raised a number of grounds of appeal. Since I would allow the appeal on some of these grounds, I will not address them all. I will address only the following issues:

(1) Did the trial judge err in finding that Mr. McHugh could have avoided the collision?

(2) Did the trial judge err in finding that the collision would not have occurred if Mr. McHugh had been travelling at the speed limit?

(3) Did the trial judge misuse the evidence about Mr. McHugh's intoxication?

[109] In my view, each of these alleged errors involve questions of fact. The appropriate standard for appellate review of questions of fact is well-settled and not in dispute. An appellate court should not interfere with a finding of fact unless the trial judge made a palpable and overriding error: Housen, at para. 10. That is, an error that is obvious, plain to see, or clear, and one that is sufficiently significant to vitiate the challenged finding of fact: Wilk v. Arbour (2017), 135 O.R. (3d) 708, [2017] O.J. No. 164, 2017 ONCA 21, at para. 18.

G. McHugh Appeal -- Analysis

(1) Did the Trial Judge Err in Finding That Mr. McHugh Could Have Avoided the Collision?

[110] The trial judge committed a palpable and overriding error in finding that Mr. McHugh could have avoided the collision if he was not negligent. Taking the evidence of vehicle location, travel times and ordinary reaction times at its highest, and applying simple mathematics, Mr. McHugh could not have avoided the accident regardless of whether he breached the standard of care of an ordinary driver. Even leaving this aside, the trial judge failed to undertake an adequate examination of how the accident occurred in determining whether Mr. McHugh could have avoided the collision.

[111] I will begin with the evidence relating to the material location and movements of Mr. McHugh's vehicle. As indicated, the trial judge found that "on the totality of the evidence Mr. McHugh was approximately 125 metres away from the intersection when he first observed the Safranyos vehicle".

[112] Mr. Forbes, the expert whose evidence the trial judge accepted in terms of "stop lines, human factors, and sight [lines]", testified that a vehicle travelling at 80 kilometres an hour would travel 22.2 metres per second. The trial judge found that Mr. McHugh was travelling in excess of 85 kilometres per hour. Naturally, if that was so Mr. McHugh's vehicle would travel more than 22.2 metres per second.

[113] Applying simple mathematics, even employing the conservative speed of 80 kilometres per hour used in Mr. Forbes' testimony, it would have taken Mr. McHugh about 5.6 seconds to travel the 125 metres from the warning sign to the point of collision near the centre line of Green Mountain Road.

[114] The trial judge accepted that Ms. Safranyos's vehicle had stopped at the stop sign, ten metres from the entrance to the intersection. She also found that Ms. Safranyos was just leaving the stop sign when Mr. McHugh first observed her vehicle from 125 metres, or slightly more than five seconds of travel time, away. Mr. Forbes's evidence was that it would have taken Ms. Safranyos between 3.5 to 5.6 seconds to travel the 14 metres from the stopped position to the point of collision, four metres into the intersection.

[115] Ms. Safranyos therefore entered the intersection either immediately before, or at the same time that Mr. McHugh arrived at the intersection. Taking the evidence at its highest, he was less than three seconds away from the point of the collision when she did so. The fact that she made it only four metres into the intersection before the collision reinforces how close Mr. McHugh must have been when this occurred, and how brief his opportunity to react was.

[116] The trial judge received reaction time evidence from three expert witnesses. The range of reaction times for ordinary drivers was between 1.3 and 2.5 seconds. Even if Mr. McHugh reacted more quickly than most ordinary drivers, in my view it is unrealistic to expect that he could have reacted and then taken effective action to avoid the collision, since he had nowhere to go. The collision occurred as Ms. Safranyos's vehicle straddled both northbound lanes directly in front of Mr. McHugh.

[117] There was also other evidence supporting the conclusion that Mr. McHugh could not have avoided the collision that the trial judge failed to consider. For example, Mr. Forbes testified that if Mr. McHugh's vehicle was travelling at 80 kilometres an hour, Ms. Safranyos's vehicle should not have entered the intersection if Mr. McHugh was 135 metres away. Moreover, Mr. Jackson said that it was only a split second from the time Ms. Safranyos entered the intersection until the collision.

[118] In my view, on the evidence before the trial judge, it could not properly be found on the balance of probabilities that Mr. McHugh could have avoided the collision regardless of whether he was speeding or not.

[119] Indeed, the trial judge's entire approach to the causation issue was fatally flawed because she failed to make a finding as to how the accident actually occurred. Without paying close regard to the movement of Ms. Safranyos's vehicle, she was in no position to determine whether Mr. McHugh could have avoided the collision. Yet the trial judge's evaluation of the movements of Ms. Safranyos's vehicle was not only brief, but clipped or superficial. There was much more to be considered than the sole finding she made that Ms. Safranyos stopped at the stop sign, ten metres from the intersection, looked to her left, and did not look to her left again, failing to yield the right-of-way.

[120] For example, Ms. Safranyos testified that when she looked left from the stop sign, she saw lights coming. Yet she never looked left again. The trial judge made no attempt to consider the role that this played in the accident, including in assessing Mr. McHugh's responsibility.

[121] Ms. Safranyos testified that she initially intended to turn left but because the traffic was "unbelievable", decided to turn right. Yet the evidence of Mr. Jackson and the position of Ms. Safranyos's vehicle at the time of the collision suggest that she was turning left. The trial judge did not address this either.

[122] Mr. Jackson gave a detailed description of the movements of Ms. Safranyos's vehicle, including that everyone was "just going" to get onto the highway, that Ms. Safranyos's vehicle followed onto the highway "right behind" the car ahead of it, that she "stepped on the gas quite quick to get out into the intersection", and that "it was almost like she was avoiding something she was going so fast". Mr. Jackson estimated that by the time she got to the intersection, Ms. Safranyos was going ten to 15 kilometres an hour. The trial judge made no findings relating to this evidence, which, if credited, depicts a sudden, hurried and aggressive entry by Ms. Safranyos into Mr. McHugh's lane.

[123] I appreciate that it was for the trier of fact to make what she would of the evidence. The inferences were for her. The difficulty, however, is that none of the considerable evidence supporting Mr. McHugh's claim that he did not have time to react was considered by the trial judge in finding that Mr. McHugh had the last clear chance to avoid the accident, and she failed to engage in a sufficient evaluation of how the accident occurred. In my view, she erred in failing to do so.

(2) Did the Trial Judge Err in Finding That the Collision Would Not Have Occurred if Mr. McHugh Had Been Travelling at the Speed Limit?

[124] Mr. McHugh argues that the trial judge "misapplied the law" in finding that the collision would not have occurred if Mr. McHugh had been travelling at the speed limit. He takes issue, specifically, with her causation finding that, "[i]f Mr. McHugh was driving the speed limit, in all probability the second half of the Safranyos vehicle would have passed through the curb lane before Mr. McHugh arrived at the intersection. Accordingly, no accident would have happened."

[125] Mr. McHugh describes her thinking as the "butterfly effect", a reference to the theory that a minute factor can, cumulatively with other circumstances, produce a very large and unforeseeable effect. His point, as I take it, is that the trial judge relied upon the minute consideration of the role that Mr. McHugh's speed played in determining his location at the time Ms. Safranyos drove her vehicle into the intersection, and then assigned blame for the accident on Mr. McHugh because of the location where his speed happened to take him. I agree that this was the trial judge's reasoning. I also agree that this causation theory does not satisfy the "but for" test of causation.

[126] The "but for" test for causation was re affirmed in Ediger, at para. 28:

That is, the plaintiff must show on a balance of probabilities that "but for" the defendant's negligent act, the injury would not have occurred . . . "Inherent in the phrase 'but for' is the requirement that the defendant's negligence was necessary to bring about the injury -- in other words that the injury would not have occurred without the defendant's negligence"[.]

(Citations omitted)

[127] I would make two points. First, as this passage demonstrates, the required link is between the "negligent act" and the injury. In assigning causation, attention must therefore be paid to what makes conduct negligent. Speeding is negligent not because of the geographical location speed brings one to. It is negligent because of the inherent risks, such as reduced reaction time or loss of control of a vehicle that speed can produce. The injury must be linked to the negligent act of speeding and the inherent risks emanating from the negligence itself. The coincidence that, without the speed one vehicle was travelling, two vehicles would not have arrived at the same location is not enough.

[128] Second, causation requires "substantial connection between the injury and the defendant's conduct": Resurfice Corp. v. Hanke, [2007] 1 S.C.R. 333, [2007] S.C.J. No. 7, 2007 SCC 7, at para. 23; Clements (Litigation Guardian of) v. Clements, [2012] 2 S.C.R. 181, [2012] S.C.J. No. 32, 2012 SCC 32, at paras. 20-21 and 28; and Sacks v. Ross, [2017] O.J. No. 5171, 2017 ONCA 773, 417 D.L.R. (4th) 387, at para. 118, leave to appeal to S.C.C. refused [2017] S.C.C.A. No. 491, 2018 CarswellOnt 10678. In Clements, at para. 28, the manifestation of this limit was described by McLachlin C.J.C. as "a robust and common sense application of the 'but for' test of causation". In my view, finding causation based on the coincidental location a speeding vehicle happens to arrive at is not a robust and common sense application of the "but for" test.

[129] However, the trial judge did not rely solely on this causation theory. She also concluded that Mr. McHugh's speed contributed to his failure to react. This is a proper speed-based causation finding. The trial judge's "butterfly effect" causation error is not, therefore, an overriding error. Nevertheless, as I explained above, on the evidence the trial judge accepted regarding distance and speed, Mr. McHugh could not have avoided the collision regardless of whether he was speeding or not. The speed at which McHugh was driving cannot be a "but for" cause of the accident if the collision was unavoidable even for drivers exercising reasonable care.

(3) Did the Trial Judge Misuse the Evidence About Mr. McHugh's Intoxication?

[130] The trial judge misused Mr. Kupferschmidt's expert alcohol impairment opinion evidence in two respects. First, she reasoned that Mr. McHugh was impaired without a sufficient evidentiary basis. Second, she misinterpreted the effect that alcohol impairment would have on a driver's perception to reaction time.

(a) Mr. McHugh's Purported Impairment

[131] Mr. McHugh urges that the trial judge's findings of impairment are not supported by the evidence. I agree. Her findings are not supported by the evidence, and she made erroneous conclusions about the legal impact of the blood alcohol evidence. I will begin with this latter point.

[132] The trial judge relied upon Mr. Kupferschmidt's calculation of Mr. McHugh's blood alcohol range -- 66 to 82 milligrams of alcohol in 100 millilitres of blood -- to conclude that Mr. McHugh "would have been impaired according to the Highway Traffic Act of 0.05 blood alcohol level" and that, if charged criminally he would have been convicted based on his blood alcohol readings of impaired driving. Neither of these conclusions was available to the trial judge.

[133] First, the question before her was a factual one -- whether Mr. McHugh's ability to operate a motor vehicle was impaired by alcohol. That factual question is to be determined on the evidence, not by legislated levels of tolerance for blood alcohol.

[134] Second, the trial judge's use of the legislation betrays a material misunderstanding of these provisions and their application.

[135] The trial judge's conclusion that Mr. McHugh would have been convicted of impaired driving based on his blood alcohol readings is wrong in law. The evidence was that Mr. McHugh's blood alcohol level "straddled" the legal limit of 80 milligrams. In other words, it may have been over the 0.08 blood alcohol level or it may have been under 0.08, depending on where in the expert's range it actually fell. The possibility that blood alcohol is actually under the 0.08 level can raise a reasonable doubt in appropriate straddle cases: R. v. Ibanescu, [2013] 2 S.C.R. 400, [2013] S.C.J. No. 31, 2013 SCC 31; and R. v. Roberts, [2018] O.J. No. 2279, 2018 ONCA 411, at paras. 111-113. An important consideration is the extent to which the range exceeds the legal limit: R. v. Gibson, [2008] 1 S.C.R. 397, [2008] S.C.J. No. 16, 2008 SCC 16, at para. 73. Mr. McHugh's range falls predominantly below the legal limit. The trial judge's claim that he would have been convicted if he had been criminally charged is therefore highly questionable.

[136] More importantly, even if Mr. McHugh could be convicted because of his blood alcohol range, the conviction would not be for impaired driving. The Criminal Code, R.S.C. 1985, c. C-46 has two relevant offences, "impaired operation" in s. 253(1)(a) and "operation over 80" in s. 253(1)(b). It is incorrect to equate proof of the latter offence as proof of the former. If it could be concluded that all persons having over 80 milligrams of alcohol in 100 millilitres of blood are impaired, then there would be no need for the over 80 offence.

[137] Mr. Kupferschmidt gave evidence about the role of the over 80 offence. He explained that "impairment can begin at levels as [sic] 50 and that most persons in our society have some form of [impairment] at 100, the level of 80 was deemed to be a reasonable compromise". Being "over 80" does not, therefore, equate to impairment. The only factual conclusion that can be made if, while operating a motor vehicle a person had more than the criminal standard of 80 milligrams of alcohol in 100 millilitres of blood, or more than the regulatory, provincial blood alcohol standard of over 50 milligrams of alcohol in 100 millilitres of blood, is that their blood alcohol level exceeded those standards.

[138] The trial judge therefore misused the law to verify her impairment conclusion in a fashion that is manifestly prejudicial.

[139] In my view, her findings related to Mr. McHugh's impairment were also based on a misapprehension of the evidence before her. Mr. Kupferschmidt imposed significant qualifications on his evidence. He could not offer an opinion on Mr. McHugh's impairment because all he had were crudely quantified conclusions about the effects that alcohol can have on general populations. Those generalities about the effects alcohol tends to have at various blood alcohol levels -- "[impairment] can begin at levels of 50", "persons may start to exhibit impairment at 50 milligrams percent", "most persons have some impairment at 100", and "the majority if not all would show signs of impairment at 100 milligrams per cent" -- were unlinked by any evidence to Mr. McHugh. That is why Mr. Kupferschmidt spoke in terms of possibilities. He cautioned that he could not be more definitive because "[he] was not with Mr. McHugh. [He] did not see him that evening."

[140] In my view, the trial judge misused this evidence. In effect, she reasoned that since the unspecified odds are that Mr. McHugh would be one of those people who has impaired faculties at the blood alcohol range he exhibited, he should be treated as impaired. This was not appropriate. The trial judge had no evidence before her about whether Mr. McHugh exhibited any physical signs of impairment, or any other indicia of impairment to ground her finding. Indeed, the police and hospital records taken on the night of the accident did not indicate that Mr. McHugh was impaired. Care has to be taken in using general probabilities as proof capable of forensic application in particular cases. In my view, the trial judge did not properly understand the limits of the expert evidence.

[141] This, in my view, was a palpable and overriding error. It undercut the integrity of a finding that was central to the holding that Mr. McHugh was at fault. It would have been appropriate for the trial judge to factor in alcohol consumption as a general consideration in assessing negligence even without clear evidence of impairment, but her misuse of legal alcohol driving standards had the effect of unfairly exaggerating the significance of the evidence of impairment.

(b) The Effect of Alcohol Impairment on Perception to Reaction Time

[142] The trial judge clearly misapprehended the evidence relating to the possible effects of alcohol impairment on the perception to reaction time of a motorist.

[143] Mr. Kupferschmidt testified about the impact that alcohol consumption can have on reaction times. He read a select passage from a scientific study without particularizing the blood alcohol level he was referring to, and without identifying what degree of delayed reaction time alcohol can produce. That passage confirmed that "[w]hen demonstrated experimentally statistically significant increases in complex reaction time may be measured as milliseconds (MS) changes". Mr. Kupferschmidt continued reading:

These increases in response time might therefore be misinterpreted as inconsequential. However, a 100 millisecond increase in overall reaction time at a driving speed of 100 kilometres per hour would result in an additional 2.8 metres of distance travelled before a response is elicited.

(Emphasis added)

[144] The trial judge attempted to paraphrase this passage in summarizing Mr. Kupferschmidt's evidence. She said: "Mr. Kupferschmidt indicated 100 milliseconds of delayed reaction can be the difference of 2.6 seconds" (emphasis added). In my view, the 0.2 error (2.6 as opposed to 2.8) she made in repeating this evidence is not significant. The conversion of "metres" to "seconds" is. Recall that a vehicle travelling at 80 kilometres an hour travels 22.2 metres per second. Together, the trial judge's misperception of Mr. Kupferschmidt's testimony and the evidence of how far a vehicle would travel per second would suggest that for every 100 milliseconds of delayed reaction time, an impaired driver would travel close to 60 metres, or ten car lengths, before reacting. The passage Mr. Kupferschmidt read actually furnished a travelled distance per 100 milliseconds of delayed reaction time of 2.2 metres, or about a third of a car length, if the driver was driving 80 kilometres per hour.

[145] This proposition did not remain abstract. The trial judge linked her misperception of the evidence to Mr. McHugh in a passage which I have already quoted. It bears repeating here:

If, and I underscore the word if, he did in fact only see her when he was three car lengths as was his evidence, it was because of his alcohol that he had consumed and his excessive speed. These two factors affected his perception of the Safranyos vehicle moving and affected his reaction time. That is, Mr. McHugh did not notice Ms. Safranyos move from a stopped position until it was too late. As Mr. Kupferschmidt indicated, 100 milliseconds of delayed reaction can be the difference of 2.6 seconds.

[146] The trial judge took generic testimony that had been offered without any guidance on how many milliseconds of delayed reaction time alcohol tends to produce at any given blood alcohol level, converted that evidence from metres to seconds, and then relied on it as explaining why Mr. McHugh may not have observed Ms. Safranyos's vehicle until it was too late to react. As evidenced by the excerpt above, this misapprehension of the evidence was clearly a factor that led to the trial judge's erroneous conclusion that Mr. McHugh could have avoided the collision.

[147] As such, the trial judge's findings on the effect of Mr. McHugh's purported impairment is a palpable and overriding error. It created a grossly inaccurate image of the impact that alcohol has on reaction time and therefore distance travelled, which was crucial to the outcome of the trial.

H. McHugh Appeal -- Conclusion

[148] I would therefore allow Mr. McHugh's appeal. I would not order a new trial. On the uncontentious evidence presented at trial, the accident was caused by Ms. Safranyos's failure to stop her vehicle before entering the intersection, her failure to observe Mr. McHugh's vehicle travelling northbound in the curb lane of Upper Centennial Parkway approximately 125 metres away and her failure to yield the right-of-way to that vehicle, coupled with the non-repair of the intersection. Absent the trial judge's errors in the assessment of the evidence relating to Mr. McHugh, it cannot be said on the balance of probabilities that Mr. McHugh could have avoided the collision had he been driving prudently. It therefore cannot be said that Mr. McHugh's negligence caused or contributed to the collision. I would dismiss the claim and cross-claims against Mr. McHugh.

Costs

[149] The parties may make written submissions on costs. With respect to docket C63133, Mr. McHugh shall file his submissions within ten days of the release of these reasons and the respondents shall file their submissions within ten days thereafter. With respect to docket C63175, the respondents shall file their submissions within ten days of the release of these reasons Hamilton shall file its submissions within ten days thereafter. None of the costs' submissions shall exceed five pages in length, excluding costs outlines.

Disposition

Mr. McHugh's appeal allowed; City of Hamilton's appeal dismissed.