Financial Services Commission

Commission des services financiers de l’Ontario

Neutral Citation: 2002 ONFSCDRS 123

Appeal P01-00026

OFFICE OF THE DIRECTOR OF ARBITRATIONS

BALJIT S. KUMAR

Appellant

and

COACHMAN INSURANCE COMPANY

Respondent

Before:

Nancy Makepeace, Director's Delegate

Counsel:

Michael J. Gillen (for Mr. Kumar)

Eric K. Grossman (for Coachman)

APPEAL ORDER*

Under section 283 of the Insurance Act, R.S.O. 1990, c.I.8, as amended, it is ordered that:

The appeal is dismissed, and the Arbitrator's order, dated April 27, 2001, is confirmed.
Coachman shall pay Mr. Kumar's appeal expenses.

August 9, 2002

Nancy Makepeace

Director's Delegate

Date

REASONS FOR DECISION

I. NATURE OF THE APPEAL

Mr. Kumar appeals from the Arbitrator's decision, dated April 27, 2001, that the incident in which he was injured was not an "accident," as defined in s. 2(1) of the SABS-1996.¹ This is the latest in a series of decisions about the boundaries of the definition of "accident " in the SABS-1996, and the first appeal decision to consider the application of this provision to an assault on a cab driver.

For the following reasons, I agree with the Arbitrator that Mr. Kumar's impairments were not directly caused by use or operation of an automobile. They were caused by his assailant, who struck him on the right side of his head with a hard object. Therefore, the incident was not an "accident" as defined.

II. THE ARBITRATION DECISION

The parties agreed about most of the facts, which the Arbitrator summarized, in part, as follows:

The Applicant, Mr. Kumar, was injured on November 29, 1998, at approximately 10:30 p.m. The incident occurred while he was operating a taxicab.
Mr. Kumar picked up a male passenger at a taxi stand at Square One Plaza in Mississauga. The passenger asked him to drive to Mineola Avenue. After turning into Mineola Avenue from Highway 10, the passenger directed Mr. Kumar to turn right at the first street and to stop the taxicab.
As Mr. Kumar was stopping the vehicle, the passenger moved from the rear passenger side of the car to the rear driver's side of the car and hit Mr. Kumar on the right side of the head with a hard object. At the same time, Mr. Kumar opened his door with his left hand to try to get out of the car. He screamed for help. He was wearing his seatbelt.
After hitting Mr. Kumar on the head, the passenger got out of the rear driver's side of the car and ran away.
The car was still in gear and began to roll into the ditch on the right side of the road. The car came to a stop upon impact with the ditch. Mr. Kumar sat dazed and bleeding. After a few minutes, he backed the car out of the ditch and drove to his apartment, which he shared with his brother.
His brother called the police and drove Mr. Kumar in the taxicab to the emergency department of the Mississauga Hospital.
The police caught the assailant after he robbed another taxi some time later that evening. He confessed that he needed money and had devised a plan to rob a taxi. The assailant told the police that during the first attempt (the one on Mr. Kumar), he got scared and ran away. He told police that he used a rock to hit Mr. Kumar. The assailant was charged and convicted of assault causing bodily harm, robbery and breach of probation.

There were two factual disputes at the arbitration hearing. First, Mr. Kumar testified that he lost consciousness after hitting his head against the steering wheel when the car went into the ditch. The Arbitrator rejected this evidence, relying on Mr. Kumar's contemporaneous statement to the police, which omitted any reference to hitting his head on the steering wheel, and his statement to Coachman's adjuster, a month later, in which he stated, "no part of my body hit the interior of the vehicle as I still had my seatbelt on. I did not sustain any personal injuries as a result of going into the ditch." The Arbitrator also relied on the hospital emergency record, which did not include any reference to losing consciousness. She found that Mr. Kumar "was dazed and disoriented after receiving the blows to the head and the taxi rolling into the ditch" and that his injuries "were the result of the blows to the head with a rock by the passenger in his taxi."

Second, Mr. Kumar submits that the Arbitrator implicitly accepted that he was restrained by his seatbelt, and he asks me to conclude that in this way the automobile directly caused his impairment by impeding his escape. Coachman submits I should not consider "the seatbelt argument" because the Arbitrator heard no evidence on point and did not make a finding. The Arbitrator came to the following conclusion:

The Applicant made the argument that the seatbelt restrained him from fleeing the attack, and, therefore, was an aggravating factor in the injury. I accept that the use or operation of the vehicle includes the use of the seatbelt. I do not find, however, that the use of the seatbelt directly caused the impairment. The restraint of the seatbelt may have contributed to the degree of the injury, but, in my view, it was not the efficient, predominate or direct cause of the impairment.² [italics added]

As I read this passage, the Arbitrator did not find it necessary to decide whether Mr. Kumar's seatbelt contributed to his impairment because she concluded this would not bring the incident within the scope of the definition of "accident." It is not my role, on appeal, to make factual findings. In any event, I agree with the Arbitrator that it makes no difference to the outcome.

The Arbitrator applied the two-part test set out by the Supreme Court of Canada in Amos v. ICBC³. She concluded that Mr. Kumar's commercial use of his taxicab on the night of the incident satisfied the "purpose" test because it "can be characterized as an ordinary use or operation of the automobile." Considering the "chain of causation" test, she compared Mr. Kumar's case to the "accident" cases under the present and previous versions of the SABS. She concluded that the test was not satisfied in this case. As I read the decision, the most important considerations were (i) that Mr. Kumar did not suffer any injuries apart from the injuries he suffered in the assault, and (ii) that the assault was part of an attempted robbery, not a dispute about the fare. She concluded:

Although there is a connection between the fact that Mr. Kumar was driving a taxicab, which has a commercial component, and an attempt at a robbery, in my view, the vehicle provided an opportunity for a robbery, but the use or operation of the vehicle was peripheral or incidental to the assault. In other words, the taxicab was the location of the attempted robbery which left Mr. Kumar injured, not the cause of his injuries.

I found that there was no injury caused by an impact with the automobile or as a result of the taxicab rolling into the ditch. As noted in Karshe,⁴ while it may not be necessary to find that the Applicant came into direct contact with a vehicle to bring the case within the definition of "accident" under the 1996 Schedule, there must be a direct link between the circumstance that caused the injury and the use of the automobile; it must be the "efficient, predominate, or direct cause of his impairment." In this case, I conclude that the efficient, predominate or direct cause of the impairment was the impact of the rock used by the assailant to hit Mr. Kumar's head and hand.

As a result, I find that Mr. Kumar was not injured as a result of an "accident" as defined in subsection 2(1) of the 1996 Schedule.⁵

III. THE PARTIES' POSITIONS

Mr. Kumar does not allege any error in the Arbitrator’s findings of fact.⁶ The only issue on appeal is whether the Arbitrator erred in her analysis and application of the definition of "accident " in s. 2(1) of the SABS-1996.

Mr. Kumar submits that the use or operation of an automobile "directly caused or materially contributed" to his impairments in five ways: (i) he picked up a person purporting to be a fare-paying passenger while operating the taxicab as a taxicab; (ii) the person's intention was to rob him because, as a taxicab driver, Mr. Kumar was an easy target; (iii) the assault commenced while the taxicab was in motion; (iv) while being assaulted, Mr. Kumar opened the driver's door, but was unable to escape because the car was in motion, he could not use his hand to put the vehicle in park, and he was restrained by his seatbelt; and (v) the taxicab was not merely the site of the assault, but was being used as an automobile and taxicab at the time.

Mr. Kumar submits that the Arbitrator erred in requiring use or operation of the automobile to be the "efficient, predominate [sic] or direct cause of . . . impairment." He argues that the SABS-1996 test – "directly causes an impairment" - is not that narrow. He submits that the correct test, consistent with the remedial character of the regulation, is the material contribution test that was reaffirmed by the Supreme Court of Canada in Athey v. Leonati⁷ and has been applied in numerous Commission decisions. He relies on several Commission decisions that he views as having taken a broader approach to the definition of "accident" in the SABS-1996.⁸He also relies on the minority of decisions under the SABS-1990 and SABS-1994 that favoured assaulted drivers.⁹

IV. ANALYSIS AND CONCLUSION

A. The Definition of "Accident"

Subsection 2(1) of the SABS-1996 defines "accident" as follows:

2(1) In this Regulation,

"accident" means an incident in which the use or operation of an automobile directly causes an impairment or directly causes damage to any prescription eyewear, denture, hearing aid, prosthesis or other medical or dental device. [italics added]

The predecessor regulation defined "accident" differently:

In this Regulation,

"accident" means an incident in which, directly or indirectly the use or operation of an automobile causes an impairment or causes damage to any prescription eyewear, denture, hearing aid, prosthesis or other medical or dental device. [italics added]¹⁰

The definition in s. 2 of the SABS-1990 was substantially similar to the SABS-1994 definition:

"accident" means an incident in which the use or operation of an automobile causes, directly or indirectly, physical, psychological or mental injury or causes damage to any prosthesis, denture, prescription eyewear, hearing aid or other medical or dental device; [italics added]¹¹

Mr. Kumar concedes that the deletion of the words "or indirectly" from the definition in the SABS-1996 reflects the drafters' intention to narrow the scope of coverage. The question is whether he is excluded as a result of the change.

B. Amos v. ICBC

A focal point of the cases is the Supreme Court of Canada decision in Amos v. ICBC. While stopped at a light, Mr. Amos' van was surrounded and attacked by a group of men. Mr. Amos was shot in the driver's seat, but managed to lock the doors, and eventually to drive away and escape. He claimed benefits under the British Columbia statutory accident benefits regulation, which provided coverage for "injury caused by an accident that arises out of the ownership, use or operation of a vehicle."

The British Columbia Supreme Court (Hardinge J.) reviewed the two leading cases interpreting "arising out of" coverage language: Stevenson v. Reliance Petroleum Ltd., 1956 CanLII 27 (SCC), [1956] S.C.R. 936, and Law, Union & Rock Insurance Co. v. Moore's Taxi Ltd., 1959 CanLII 81 (SCC), [1960] S.C.R. 80. In Stevenson, an employee of a fuel delivery company negligently spilled gasoline while pumping to a customer's storage tank, leading to a serious fire. The company's policy provided coverage "for loss or damage arising from the ownership, use or operation" of a motor vehicle. The Supreme Court of Canada accepted that the fire damage arose from use or operation of the truck. Justice Rand set out what came to be called "the purpose test " in the following passage:

The expression, "use or operation" would or should, in my opinion, convey to one reading it all accidents resulting from the ordinary and well-known activities to which automobiles are put, all accidents which the common judgment in ordinary language would attribute to the utilization of an automobile as a means of different forms of accommodation or service.¹²

Law, Union & Rock Insurance was the genesis of "the chain of causation test." In that case, the insured taxi company had a contract to transport mentally handicapped children to and from school. One of the children was hit by a truck when crossing the street. The general liability insurer argued that the company's automobile insurer was liable. However, the Supreme Court of Canada concluded that the injury arose from the driver's breach of his contractual duty to drop children off on the same side of the street as their homes. Justice Ritchie, speaking for the Court, discussed the phrase, "arising out of:"

The meaning to be attached to the words "arising out of" as they occur in the exclusion here in question has, of course, been the subject of much discussion in this case. Adamson C.J.M. has said that "The words are clear and must bear their own meaning. They refer to the immediate or proximate cause." On the other hand, the appellant contends that the words have a wider connotation and should be construed as meaning "originating from, incident to or having connection with the use of the vehicle", but that even if they bear the more restricted meaning the circumstances of the present case are such that the composite negligence of the taxi driver is not severable and that the proximate cause of the accident can, therefore, be said to have been the use and operation of the vehicle in stopping on the wrong side of the street. It is sufficient to say that the words "claims arising out of . . . the ownership, use or operation . . . of any motor vehicle" as used in this exclusion can only be construed as referring to claims based upon circumstances in which it is possible to trace a continuous chain of causation unbroken by the interposition of a new act of negligence and stretching between the negligent use and operation of a motor vehicle on the one hand and the injuries sustained by the claimant on the other.¹³

After a review of Canadian and U.S. authorities, Hardinge J. concluded that the purpose test and the chain of causation test must both be satisfied, and that "mere presence in a vehicle when injuries are sustained is not sufficient." He did not accept that Mr. Amos' assailants were attempting to hijack the van, or that the van itself contributed to his injuries. He concluded that "even if anything more than a minimal connection between the ownership, use or operation of a vehicle and accidental injuries is sufficient to establish the required nexus, mere presence in a vehicle when injuries are sustained is not sufficient." As the van was merely the site of the attack, the required causal connection was not satisfied.¹⁴

The British Columbia Court of Appeal found no error in the judgement at first instance, concluding that there must be "more than a 'but for', 'fortuitous', or 'incidental' relationship . . . between the ownership, use or operation of a vehicle and the accident resulting in injury," and therefore "it is not sufficient that the accident happens while the injured party is using or operating the vehicle."¹⁵

This decision was reversed on appeal to the Supreme Court of Canada, with Justice Major writing for a unanimous Court. Although the case arose in the context of British Columbia's statutory accident benefits scheme, Major J. held that private insurance principles favouring broad interpretation of coverage provisions applied. As is well known, Major J. adopted a two-part coverage test that combined the two leading analyses:

The purpose test: Did the accident result from the ordinary and well-known activities to which automobiles are put?
The chain of causation test: Is there some nexus or causal relationship (not necessarily a direct or proximate causal relationship) between the appellant 's injuries and the ownership, use or operation of his vehicle, or is the connection between the injuries and the ownership, use or operation of the vehicle merely incidental or fortuitous?

Major J. easily concluded that the appellant satisfied the first part of the test because he was driving his van down a street, which is an "ordinary and well-known" activity to which automobiles are put. The focus of the judgement was on the causation test. Major J. set out a number of principles:

(i) ". . . it is clear that a direct or proximate causal connection is not required between the injuries suffered and the ownership, use or operation of a vehicle. The phrase "arising out of" is broader than "caused by", and must be interpreted in a more liberal manner."[para. 21]

(ii) "Such proof is helpful in establishing the necessary nexus or causal link, but it should not be mandatory for an injured plaintiff to establish an assailant's intent. It is always open to the courts to draw reasonable inferences regarding causation from the facts." [para. 21-22]

(iii) "Negligence or fault in the use or operation of a motor vehicle does not need to be the cause of the injury. The liability for the injury may arise f rom a tortious act other than the negligent use of a motor vehicle." [para. 23]

(iv) ". . . a motor vehicle need not be the instrument of the injury to satisfy the causal connection requirement." [para 24]

Major J. posited the assailants' intent as the nexus between the vehicle and Mr. Amos' injuries:

Was the attack in this case merely a random shooting, or did it arise out of the ownership, use or operation of the appellant 's vehicle? While the appellant 's van may have been singled out by his assailants on a random basis, the shooting appears to have been the direct result of the assailants' failed attempt to gain entry to the appellant's van. It is not important whether the shooting was accidental or deliberate while entry to the vehicle was being attempted. It is important that the shooting was not random but a shooting that arose out of the appellant's ownership, use and operation of his vehicle. [para. 25]

Reasoning by analogy, the Court recognized the difficulty of the line-drawing exercise:

If the appellant had not been shot, but had lost control of his car while trying to get away from his assailants, the injuries suffered as a result of a subsequent car crash would surely be covered by the respondent. Similarly, if the appellant had suffered injuries as a result of being intentionally hit by those same assailants using a car instead of a gun, the respondent would not deny coverage. I do not think the instant case can be distinguished from the foregoing hypothetical examples. Generally speaking, where the use or operation of a motor vehicle in some manner contributes to or adds to the injury, the plaintiff is entitled to coverage, [para.26]

Amos v. ICBC remains the leading Canadian case defining the scope of motor vehicle insurance coverage. The Commission's decisions reflect different views about its application to "accident" cases under the SABS. The "arises out of" language considered in Amos is similar to the language used in the "right to sue" provisions of the Insurance Act,¹⁶ but differs from the "causes" and "caused by" language of the SABS. The Supreme Court of Canada reaffirmed in Amos that, "[t]he phrase, "arising out of" is broader than "caused by." However, an interpretation that the restriction on the right to sue is broader than accident benefit coverage runs counter to the notion of an "exchange of rights," whereby accident victims accept a restricted right to sue in exchange for expanded accident benefits.¹⁷ Reconciling the language of the Act and the SABS has, therefore, been an important issue in many of the Commission's "accident" cases. A number of Commission decisions held that the language of the "accident" definition of the SABS must be read together with the right to sue provisions of the Act, or that the broader language of the Act prevails where the SABS would otherwise restrict access to benefits.¹⁸ The Ontario Court of Appeal rejected this view in Alchimowicz v. Continental Insurance Co. of Canada.¹⁹ In the words of the court:

In Amos v. Insurance Corp. of British Columbia (1995), 1995 CanLII 66 (SCC), 31 C.C.L.I. (2d) 1 (S.C.C.), Major J., speaking for the court, found that the phrase "arising out of" is broader than "caused by", and must be interpreted in a more liberal manner.

However, the phrase "arising out of" does not appear in s.268 [the section that provides that motor vehicle liability policies shall provide accident benefits "subject to the terms, conditions, provisions, exclusions and limits set out in that Schedule."] or in the regulation and these are the sole sources of liability for no-fault benefits. We were shown opinions of arbitrators who have held that the language of s. 266 [the right to sue provision relating to SABS-1990 accidents] and s.268 can be merged and that the more restricted language of the schedule can be ignored. In our view, they are wrong. The legislation, for whatever reasons, set one test for exclusion of actions and explicitly left the determination of no-fault benefits to be determined by regulation in the schedule. If there is a difference, and there is in the words, it is authorized.

On that analysis, the appellant is left with one arm of the argument; that is, that "indirect" is broader than "direct." With this we agree.

This ruling was applied in subsequent Commission decisions. However, in Vijeyekumar v. State Farm,²⁰ the Court of Appeal stated that Amos applies to the definition of "accident " in the SABS-1994:

The wording of the British Columbia legislation under which Amos was decided differs slightly from the wording used in Ontario's Schedule. The British Columbia statute provides for the payment of benefits 'in respect of death or injury caused by an accident that arises out of the ownership, use or operation of a vehicle"; the definition of accident in the Schedule speaks of "an incident, in which directly or indirectly the use or operation of an automobile causes an impairment". Nonetheless, the two provisions are enough alike that the Amos test may be applied to the definition of accident under the Schedule.

Although the Commission decisions considering the narrower definition of "accident" in the SABS-1996 have continued to apply the two-part test set out in Amos, the weight of authority is that deletion of the words, "or indirectly," makes the broad "some connection" test adopted in that case inapplicable.

C. The SABS-1990 and SABS-1994 Cases

Although a substantial minority of decisions go the other way,²¹ Commission adjudicators have generally found that assaults do not fall within the definitions of "accident" in the SABS-1990 or SABS-1994. The Commission's appeal decisions have been consistent on point, apart from one early decision.²² The leading cases were decided before the decision of the Supreme Court of Canada in Amos, and before the Court of Appeal stated that Amos applies to the "directly or indirectly" definition of "accident."

The fact situations fall into three main categories: attacks on cab drivers over a fare dispute or in the course of a robbery, "road rage" incidents initiated by driving disputes, and apparently random attacks unrelated to the use or operation of the vehicle. Although some attention is paid to the "purpose" test, particularly in cases where the attack occurs outside the vehicle or after the vehicle is stopped, the primary focus is the "chain of causation" test. The Commission cases consistently state that while the notion of indirect causation broadens coverage beyond proximate cause, it is not sufficient that the attack occurred in or near the vehicle; the vehicle must be more than merely the situs (location) of the attack. The prevailing view is that the assailant's criminal act breaks the chain of causation between use or operation of the vehicle and the injuries suffered by the insured person. Commission adjudicators have rejected the submission that the requisite connection is established where the commercial use of the vehicle provided the motivation or opportunity for the attack. For example, the Arbitrator in the decision below relied on the following statement by Arbitrator Draper in Ekunah and Simcoe & Erie General, a decision that pre-dated Amos:

At most, it might be found that the assailant was attracted to Mr. Ekunah because taxidrivers are easy targets who are likely to carry cash. I do not believe, however, that automobile insurance is intended to insure against the risks of carrying cash.[p. 13]

The Commission’s approach was not significantly altered by the decision of the Supreme Court of Canada in Amos, in part because of Justice Major's statement that "arises out of" is broader than "caused by." Like the Court of Appeal in Alchimowicz, most Commission adjudicators took this to mean that the very broad "some connection" test is not applicable to decisions under the "accident" definition of the SABS. Exactly where the line is drawn has attracted much discussion.

The first decision to consider Amos was Hanlon and Guarantee, which was also the Commission’s first "road rage" case.²³ The dispute began when the two drivers jockeyed for position in the passing lane, and ended at the side of the road, when the other driver struck Mr. Hanlon with his cell phone. The other driver pleaded guilty to a charge of common assault.

Applying the two-part Amos test, Arbitrator Renahan found that Mr. Hanlon did not meet the purpose test because at the time of the injury, both drivers had stopped and exited their vehicles, time had passed and a number of deliberate acts had taken place. Nor did he meet the chain of causation test, because "[t]he chain of causation between the use of the vehicles and the injury was severed by new and independent acts which occurred after the vehicles had stopped." [p. 13]

Arbitrator Bayefsky came to a different conclusion in Overley and Co-operators,²⁴Mr. Overley's female companion struck another vehicle while attempting to back out of a parking spot at a tavern. When she left her vehicle to inspect the damage, she was assaulted by the driver of the vehicle she had hit. Mr. Overley remained in the car. The other driver then assaulted him where he sat, pulled him out of the vehicle and continued the attack. He was later convicted of assault.

Arbitrator Bayefsky held that neither the purpose test nor the chain of causation test required the vehicle to be in operation at the time of the incident. He distinguished Hanlon, where "[t]ime had passed and a number of deliberate acts had taken place between the last use and operation of the vehicles and the injury."²⁵He found that Mr. Overley satisfied the purpose test because sitting in the car after the collision is "an ordinary and well-known activity to which automobiles are put " and this activity continued until the assailant pulled him from the vehicle. He also concluded that the assault followed almost immediately after the collision, and was motivated by it, therefore "the connection between the injuries suffered . . . and the use of the automobile was not random, incidental or fortuitous; nor was the vehicle the mere situs of an attack unrelated to the use or operation of the car." [p.16] He was not persuaded that the assailant's actions were an intervening cause:

What makes an act 'new and independent in determining whether the chain of causation has been broken is not whether it is human or criminal in nature, but whether it is sufficiently connected to the use of operation of the vehicle in question. [p.18]

Arbitrator Bayefsky concluded that the incident satisfied the "some connection" test set out in Amos.

A series of Commission appeal decisions in the late '90s adopted a relatively narrow interpretation of the "accident " definition.

In April 1996, Ekunah and Simcoe & Erie was confirmed on appeal.²⁶ Director’s Delegate Naylor did not find it necessary to decide whether the "accident" definition in the SABS mandates a broader or narrower test than laid down in Amos because, in any event, Amos required that "there must be some causal connection between the injury and the use or operation of the vehicle."

The language of the Schedule requires a causal connection between the injury and the use or operation of the vehicle, although it need not be a direct cause. The cause may be indirect; the vehicle need not be the instrument of injury. However, it is not enough to show that an automobile was merely the location of the injury, that the victim or perpetrator were occupying it at the time of the injury, or that an automobile was involved in some peripheral or incidental way. The use or operation of the vehicle must have caused the injury, whether directly or indirectly.[p. 10]

The appellant challenged the arbitrator’s finding that robbery was not established as the motive for Mr. Ekunah's murder. However, Delegate Naylor stated that if anything, the evidence suggested robbery was not the motive;²⁷ the robbery theory was mere conjecture.

In March 1997, Director’s Delegate Draper confirmed the arbitration decision in Hanlon and Guarantee, and reversed Overley and Co-operators. In Hanlon he considered the application of Amos to the "accident " definition in the SABS:

The Supreme Court makes it clear that "arising from" is broader than "caused by." Proximate cause is not required to meet the "arising from" test, nor must the injuries arise from the negligent use or operation of the automobile. The liability for the injury can arise from a tortious act other than the negligent use of a motor vehicle. However, there still must be "some nexus or causal relationship" (the Court's restatement of the chain of causation test). Although the Schedule uses "causes, directly or indirectly;" rather than "arising from," the retention of the causation test in the second part of the Amos test makes the analysis relevant to this appeal.

My review of the Amos decision was helped by the analysis of Mr. Justice Macdonald in Royal Insurance Co. of Canada v. Guardian Insurance Co. of Canada (1995), 1995 CanLII 7228 (ON CTGD), 26 O.R. (3d) 290. I agree with his view that the Supreme Court's assessment of the factual situation in Amos is critical to the result. While the trial judge found no evidence that Mr. Amos assailants were attempting to hijack his van, the Supreme Court stated that "it is always open to the courts to draw reasonable inferences regarding causation from the facts."

Instead of viewing Mr. Amos as the unlucky victim of a random robbery, the Court focussed on the relationship between his continuing use of the van to resist his assailants and the firing of the gun. . .

As Mr. Justice Macdonald describes the facts cited by the Supreme Court, "the assailants efforts to gain entry to the vehicle and the appellant’s [Mr. Amos] efforts to keep the vehicle moving amounted to a contest for control."

It was out of this contest for control of the van that the gun was fired. Whether it was fired intentionally or accidentally, the Supreme Court held that it was an accident from Mr. Amos perspective. The Court concluded that this accident (the gun being fired) arose out of the ownership, use or operation of the van and caused Mr. Amos injuries. In addition, the Court found no "intervening act, independent of the ownership, use or operation of the vehicle which broke the chain of causation."

I agree with Mr. Justice Macdonald’s analysis and description of the crux of the Amos decision:

In my respectful opinion, the Supreme Court’s statement that "the shooting appears to have been the direct result of the assailant’s failed attempt to gain entry to the appellant’s van" was a determination that the gunman’s firing of the shot was the "accident" for coverage purposes, the "direct result" was compliance with the phrase "arising out of" and "the assailants failed attempt to gain entry to the appellant’s van" was compliance with the " ownership, use or operation of the vehicle" criterion of the coverage. [at pp.6-7, footnotes omitted]

Delegate Draper stated, "[t]he analysis in Amos is broad, but not without limits." [p.8] Reviewing the cases, including the cases referred to in Amos, he concluded,

the courts have consistently looked for some connection between the ongoing use of the vehicle and the event causing the injury, or for some involvement of the vehicle in the person's injuries. Where the use of the vehicle has ended and it does not contribute to the person’s injuries, the courts have not found the necessary connection. It is not clear to me, therefore, that Mr. Hanlon meets the Amos test. [footnotes omitted]

Turning to the "chain of causation" test, Delegate Draper dismissed Mr. Hanlon's submission that the addition of "indirectly" mandates broader coverage than in Amos:

I am unable to read the phrase, "causes, directly or indirectly," that broadly. In my view, the phrase demands a causal relationship between the incident involving the use or operation of an automobile and the person’s injuries, a narrower focus than an accident arising from the ownership, use or operation of a vehicle. Although the addition of "indirectly" may take the analysis beyond proximate cause, a causal relationship still must be found.

Delegate Draper adopted the analysis of Director's Delegate Naylor in Vineski and Federation Insurance Company of Canada.²⁸ He described the facts of that case:

In Vineski, Jonathan Vineski was riding his bike when he heard the sound of a car starting. He looked to his right to see where the car was and rode into a pothole. The front wheel of his bike twisted, snapping the bolts and causing him to fall to the road. The issue was whether he was entitled to accident benefits under the Schedule. Director’s Delegate Naylor upheld the arbitrator’s decision that this was an "accident" within the meaning of the Schedule; it was an incident in which the use or operation of an automobile caused his injuries, directly or indirectly.

Delegate Naylor concluded that:

"causes, directly or indirectly" takes us somewhat beyond a strict analysis of the doctrine of proximate cause . . . . It allows for consideration of a more remote causal link. However, the link cannot be so remote as to deprive the word "causes" of meaning. As I stated in Ekunah and Simcoe & Erie General Insurance Company, (April 22, 1996, OIC P-003550), it is not enough to show that an automobile was merely the location of the injury, that the injured person was occupying it at the time or that an automobile was involved in some peripheral or incidental way. The use or operation of an automobile must have caused the injuries, whether directly or indirectly. To determine this, . . . the role of the automobile in the whole scenario must be considered.

Delegate Draper agreed with the insurer that Hanlon and Vineski were distinguishable:

Jonathan Vineski reacted instantly to the use or operation of the automobile, triggering the incident in which he was injured. As pointed out by the arbitrator in this case, the use or operation of any automobile ceased well before the time of Mr. Hanlon's injury, and both Mr. Hanlon and Mr. Daly performed a number of deliberate acts resulting in the assault, [pp.14-15]

Delegate Draper concluded:

In Mr. Hanlon’s case, the only role played by the use or operation of an automobile was to create an atmosphere of hostility between him and Mr. Daly. This was not a situation where someone lashed out as an immediate reaction to the driving incident. Mr. Hanlon and Mr. Daly both took a number of intentional actions for which they bear responsibility. While it is possible to say that the driving incident provoked or motivated the assault, I am not persuaded that there is a causal relationship, even an indirect one, between the use or operation of an automobile and Mr. Hanlon’s injuries.

"Cause" has an instrumental aspect that is lacking here. The incident involving the use or operation of an automobile ended with no one being injured. No further consequences were inevitable or linked to any ongoing use or operation of an automobile. Sometime later, in a location removed from the " incident," there was a verbal confrontation between the two men outside their vehicles. At that point, Mr. Daly attacked Mr. Hanlon. No automobile was the target of the attack, was used in the attack or contributed to Mr. Hanlon’s injuries. I conclude that, like in Alchimowicz, the connection between the use or operation of the automobile and the injury is not sufficient to be covered by automobile insurance. [p.16]

Delegate Draper applied the same analysis to Overley and Co-operators. He conceded that the facts differed from the facts in Hanlon:

The time between the collision and the assault is considerably shorter and fewer deliberate acts took place in the interim. Unlike Mr. Hanlon, Mr. Overley played a passive role and was still sitting in Ms. Bartholomew's vehicle when he was first assaulted. The question is whether these differences are sufficient to lead to a different result. [p.6]

He concluded they were not:

This is not a case where the victim was assaulted because he resisted the assailant’s efforts to gain access to or take control of the vehicle. Nor is it a case where the vehicle was used in the assault or contributed to the injuries. After Ms. Bartholomew parked her car, turned it off and went to inspect the damage to the other vehicle, Mr. Grandbois assaulted her. He then assaulted Mr.Overley. Absent some continuing contribution of the vehicle, I conclude that the causal connection, even an indirect causal connection, was broken by Mr. Grandbois intervening actions. Mr. Overley’s recourse is to sue for damages, which he has done, or apply to the Criminal Injuries Compensation Board. [footnotes omitted]

Arbitrator Blackman adopted a different view in Assaf and Commercial Union, another road rage case. He relied on Delegate Naylor's statement, in Vineski, that '"causes, directly or indirectly' takes us somewhat beyond a strict analysis of the doctrine of proximate cause" and "allows for consideration of a more remote causal link. . . . the role of the automobile in the whole scenario must be considered." Arbitrator Blackman found, in Mr. Assaf's case, that "use (or misuse) of the vehicles " was integral to the sequence of events, "from the first encounter on the QEW to the impairment being sustained, 'an incident,' as required under the Schedule." In his view, the SABS-1994 did not require that use or operation of the vehicle continue at the time of the injury.

The facts in Lenti and Zurich were more unusual.²⁹ Mr. Lenti was injured by an explosion that occurred while he sat in the driver’s seat of his vehicle, just as he was about to turn the key in the ignition. Based on the police investigation, Arbitrator Leitch found that someone had planted a bomb underneath the car and detonated it from a remote location. Mr. Lenti was injured by shrapnel from the car and the bomb. The arbitrator also found that Mr. Lenti was the intended victim of the attack; it was not random. Applying Amos, he found that sitting in the car with the intention of moving it was a necessary part of the vehicle 's operation and an ordinary and well-known use to which vehicles are put. He also concluded that the causation test was satisfied because "Mr. Lenti’s vehicle became part of the instrument, not merely the situs of the injury."

The decision was overturned on appeal.³⁰ Delegate Draper stated:

The decisions in Amos and Alchimowicz also suggest that the type of causal connection required for coverage should be assessed in light of the purpose of accident benefits. While the inclusion of " indirectly" means that the connection need not be proximate, there still must be a causal connection between the injuries and the risks associated with an ordinary and well-known use or operation of an automobile. For example, an insurer could not successfully argue that someone injured as a result of driving into a pothole and losing control of the vehicle is not entitled to accident benefits because the injuries were caused by the negligence of the municipality, not the use or operation of an automobile. This is clearly a risk of the use or operation of automobiles, bringing it within the scope of accident benefits. [p.13]

The Delegate distinguished the facts from the situation where the victim is intentionally struck by an automobile:

Again, the key is that an automobile is being driven. Although it is an intentional act from the perspective of the assailants, the situation must be viewed from the perspective of the victim. Stated differently, being hit by an automobile is a risk that accident benefits are meant to cover. [p.18]

Again, the case was distinguishable from Amos in that "[t]here is no suggestion that it was an attack on the [vehicle] or an attempt to steal it." [p.17] He concluded:

I acknowledge that the line to be drawn in these cases is difficult and troublesome. If the assailant had wired the bomb to the ignition or disconnected the brakes, the outcome might be different. Although it is difficult to distinguish these situations as a matter of public policy, the starting point is that someone who is the victim of an assault is not entitled to accident benefits simply because the situs of the assault is an automobile. It would also be hard to explain why someone who is the victim of a bombing gets accident benefits while someone who is shot does not. [p.20]

In Kohli and State Farm,³¹ the arbitrator accepted that Mr. Kohli was attacked as he approached his vehicle in a parking garage, and that the attacker intended to, and did, steal the vehicle. Although the main basis for the decision was that the car was not instrumental in the injury, the arbitrator also rejected the argument that the assailant’s use of the car subsequent to the injury brought Mr. Kohli’s incident within the scope of the definition.

The appeal decision in Kohli was the first to consider the statement of the Court of Appeal, in Vijeyekumar, that Amos applies to the SABS. Delegate Draper held that Mr. Kohli's claim failed because he did not meet the first part of the Amos test:

In this case, I am not persuaded that the assailants intention to steal a car can be used to satisfy the first part of the Amos test. Assaulting someone to get their car keys is not "an ordinary and well-known activity to which automobiles are put." [ p. 6]

In Saharkhiz, the Court of Appeal reaffirmed its earlier ruling, in Vijeyekumar, that Amos applies to the definition of "accident" in the SABS-1994. When two passengers exited the cab after refusing to pay the fare, Mr. Saharkhiz pursued them. They argued, and after a few minutes, the passengers attacked the driver. The motions judge (Lederman J.) held that the incident was an "accident" as defined in the SABS-1994. It met the purpose test because the taxi was being driven as a taxi, the driver left his door open and engine running when he left the cab to discuss the matter with his assailants, and because the dispute arose out of the commercial relationship between the passenger and driver. The latter factor was critical to the judge 's causation analysis. He concluded that the cab was not merely the situs of the attack. Rather, "[t]he role of the taxi-cab throughout the sequence of events is crucial," and there was "an unbroken line of causation . . . beginning with the ride in the taxi-cab and ending with the assault." [para. 18.]

In brief reasons affirming the decision, the Court of Appeal agreed that the "the assault was the end product of the commercial relationship that had been created when the ride in the taxi-cab had started" and that the driver's injuries "were caused, at least indirectly, by the use and operation of the taxi-cab."

In my view, the recent decisions of the Court of Appeal in Vijeyekumar and Saharkhiz suggest a broader approach to the "accident" definition in the SABS-1990 and SABS-1994 than that taken in the Commission's leading cases. However, the significant amendment of that definition in the SABS-1996 requires a different analysis.

D. The SABS-1996 Decisions

Petrosoniak and Security National was the first decision to consider the amended definition of "accident" in the SABS-1996. The insured, a cyclist, lost control of his bicycle and fell onto the roadway after hitting a wet patch on the pavement. Arbitrator Novick concluded that the incident was an "accident" as defined:

My analysis must therefore begin with a consideration of the words used in the definition, which must be interpreted in accordance with their plain meaning, keeping the purpose of the legislation in mind. Consideration must also be given to the evolution of the definition of the word "accident" in the Schedule.

Counsel for the Insurer referred to various dictionary definitions of the word "direct." One of the definitions provided by the Shorter Oxford English Dictionary (volume 1) is "without intervening agency; immediate." Black's Law Dictionary defines "direct" as being "immediate; proximate; operating by an immediate connection or relation, instead of operating through a medium." The definition I find the most useful, however, is the definition of "direct cause" found in Black’s Law Dictionary. It is defined as "the active, efficient cause that sets in motion a train of events which brings about a result without the intervention of any force started and working actively from a new and independent source" (emphasis mine).

As I understand that definition, a series of events can be the direct cause of an incident, as long as there is no intervening agency or act. Consequently, if an unbroken chain of events involving the use or operation of an automobile leads to an injury, the injury can be said to have been directly "caused by" the incident. In the instant case, the Applicant was injured as a direct result of coming into contact with a wet patch of pavement. I have found that the fluid on the pavement was released by a motor vehicle. While the fact that the fluid fell onto the roadway, as opposed to falling directly onto Mr. Petrosoniak, may constitute an intermediate step in the process which ultimately led to his injuries, it does not, in my view, constitute an intervening act or a force "working actively from a new and independent source" as referred to in the definition cited above.

The Insurer conceded that the amended definition does not go as far as to require an applicant to come into direct contact with a vehicle. Counsel submitted, however, that the deletion of the word "indirectly" from the definition requires that there be a direct link between the circumstance that causes the injury and the incident. I agree with that statement. As the sole cause of the Applicant's injuries in this case was the existence of the oily substance on the pavement, in my view, the direct causal requirement or link has been met and the incident described therefore falls within the definition of "accident" set out in section 2.³²

The next decision was Karshe and Lloyd's. Mr. Karshe, a cab driver, was assaulted by two passengers after leaving the vehicle. They had proffered a $100 bill in payment for the fare. He refused it, fearing it might be counterfeit, and pursued them into a building, where they assaulted him. One of the passengers robbed the cab while Mr. Karshe remained outside. There was no evidence whether the cab was still running or the door remained open.

Relying on Saharkhiz, Arbitrator Blackman found that using an automobile as a taxicab is "an ordinary and well-known activity to which automobiles are put," and that the assault "was both the end product of a commercial relationship resulting from the use of the automobile as a taxi cab, and the means of obtaining access to an automobile being used as a taxi-cab." Accordingly, Mr. Karshe satisfied the "purpose test."

However, the arbitrator found that the Amos causation test, which requires only some relationship between the injury and the vehicle, did not apply to the amended "accident" definition. After reviewing the case-law, he applied the definition of "direct cause" adopted by Arbitrator Novick in Petrosoniak:

It can be said that the use or operation of Mr. Karshe's taxi-cab provided the opportunity, the motive, the atmosphere of hostility and/or the emotional impetus for a train of events culminating in an injury being sustained and was, therefore, a predisposing, secondary or indirect cause of Mr. Karshe 's impairment. However, with the legislative narrowing of the scope of coverage, that is not sufficient. I find that Mr. Karshe 's injuries were only sustained upon the intervention of a force starting and working actively from a new and independent source other than an automobile, namely brass knuckles being applied by passenger #1 against the left side of Mr. Karshe 's face. In the reality of Mr. Karshe's case, the use or operation of his taxi-cab was not the efficient, predominate or direct cause of his impairment.³³

To the same effect is Sarkisian and Co-operators, in which Arbitrator Sampliner found that the insured, who was shot while performing maintenance on his vehicle, was not killed in an "accident: "

. . . the car did not play a direct instrumental role in the chain of events leading to Mr. Sarkisian 's death. Mr. Sarkisian 's replacement of windshield fluid put him at the site of his car, but his vehicle played no role in his death or in any force giving rise to his death. Even if I were to accept that theft was the motive here, the direct and intervening instrument or cause of Mr. Sarkisian's death was the gunshot.³⁴

Arbitrator Miller adopted the same approach in Mahadan and Co-operators General Insurance Company. After unloading some bags of groceries from his trunk and closing the trunk, Mr. Mahadan fell when his foot became stuck in a groove in the pavement. He broke his left leg. The arbitrator found that the crack in the pavement that caused Mr. Mahadan 's fall was "a new and independent source" that had nothing to do with the use and operation of a motor vehicle, but was related to construction on the parking lot.³⁵

Elensky and Royal & SunAlliance Insurance Company of Canada concerned a long-haul trucker who was assaulted while returning to the truck after leaving it to ask directions.³⁶ Arbitrator Novick accepted Mr. Elensky's testimony that he had left the engine running, that he was trying to get into the cab to escape the assault when he was shot, and that he was robbed and goods were stolen from the truck. She did not accept that the assailants intended to steal the truck. In concluding that Mr. Elensky 's injuries were not sustained in an "accident," she considered that he had no commercial relationship with his assailants, that the attack was not directly linked to his use of the vehicle, and that the attack did not occur while he was in the truck. She concluded that Mr. Elensky's injuries came from a new and independent source that broke the chain of causation.

Arbitrator Novick's statement of the causation test in Petrosoniak was also adopted by Justice Chapnik in Chisolm v. Liberty Mutual Group, a case that involved a drive-by shooting.³⁷ She stated that "use or operation of the vehicle must directly cause the impairment."³⁸ Rejecting the plaintiff 's submission that it was only because he was trapped in his car that he was shot, she stated, "[n]evertheless, in this case, there was not an unbroken chain of events."³⁹

In Waters and Royal & SunAlliance Insurance Company of Canada, Arbitrator Novick summarized the cases by saying,

At this point, I think it is fair to say that criminal assaults such as the shooting in [Chisolm] and the vicious assault against the driver of a truck in Elensky, constitute intervening and independent acts that break the chain of events or causation.⁴⁰

Mr. Waters died of a heart attack while driving, and his car then collided with a pole or planter. Arbitrator Novick accepted the Coroner's opinion that the heart attack resulted from a pre-existing condition, and was not driving-related. She concluded,

As stated in the Petrosoniak decision, a series of events can be the direct cause of an incident, as long as there is no intervening force or act. In this case, however, I find that Mr. Waters' heart attack represented the intervention of a force that emanated from 'a new and independent source.'⁴¹

Arbitrator Novick also reaffirmed the principle, stated by Director’s Delegate Naylor in Ekunah and Simcoe & Erie General Insurance Company, that "it is not enough to show that an automobile was merely the location of the injury, that the victim or perpetrator were occupying it at the time of the injury, or that an automobile was involved in some peripheral way. The use or operation of the vehicle must have caused the injury . . . ,"⁴²

The only appellate decision to consider the definition of "accident " in the SABS-1996 is Correia and TTC Insurance Company Limited. Although there was no question Ms. Correia was injured in an "accident," she claimed benefits in relation to separate injuries she sustained in the course of treatment. Director's Delegate Naylor’s discussion of "proximate cause" is helpful:

In the context of private insurance policies, direct cause language signals proximate cause. Indeed, there is plenty of authority for the proposition that "direct cause" and "proximate cause" are interchangeable. If the insured peril is the originating factor, the test of proximity is whether the loss was a direct result of that peril. The test for directness is whether there has been a new intervening cause. Used in this sense, direct cause speaks to the quality of the relationship between the use or operation of the automobile and the injury, not simply to their closeness in time.

Arbitrators addressing whether an accident has occurred have adopted the definition of "direct cause" in Black's Law Dictionary:

The active efficient cause that sets in motion a train of events which brings about a result without the intervention of any force started and working actively from a new and independent source.

While the test applied involves a "moving" cause type of analysis, it has not been equated with "liability focused" proximate cause in which the interposition of a new act of negligence necessarily breaks the "chain of causation" between the automobile 's involvement and the injury. In other words, if someone is hurt in a car crash attributable to substandard highway maintenance, their injuries nonetheless are considered to be directly caused by the vehicle 's use or operation.

Under the principles of proximate cause, absent explicit language in a policy, if injury or death results from treatment necessitated by an accident, providing there is no intervening cause, the accident remains the direct or proximate cause. Direct cause does not mean the only cause or the most immediate cause.⁴³

Ultimately, Delegate Naylor accepted that Ms. Correia's treatment did not give rise to an intervening cause such as to break the chain of causation.

E. Conclusion

In summary, the SABS-1996 authorities are consistent in stating that the Amos test, which requires some causal nexus but not necessarily a direct or proximate cause, does not apply to the amended definition of "accident." Arbitrator Novick's view that a direct cause is "the active, efficient cause that sets in motion a train of events which brings about a result without the intervention of any force started and working actively from a new and independent source"⁴⁴ has received broad acceptance. Mr. Kumar's submissions do not persuade me to depart from this analysis.

The deletion of the words "or indirectly" from the "accident" definition is an important consideration in my analysis. Legislative evolution can provide strong evidence of legislative intent, although it must be used cautiously because some amendments are intended to confirm the law, not to change it.⁴⁵ In this case, it would be difficult to argue that the legislature intended only a formal change because of the many decisions that relied on the presence of the words "or indirectly" in the SABS-1990 and SABS-1994 definitions to extend coverage beyond proximate cause. I think the assault cases were very likely the primary focus of this amendment.

Mr. Kumar attempted to distinguish his case from Karshe, on which the Arbitrator relied, because Mr. Kumar was still driving his cab at the time of the assault, and the assault occurred inside the cab. Moreover, the assault was committed by a passenger, and Mr. Kumar's use of the vehicle as a taxicab provided the opportunity and motive for the attack. This distinguishes the case from Ekunah and Sarkisian, where the evidence did not establish the reason for the attacks. For these reasons, I agree with the Arbitrator that Mr. Kumar was operating the vehicle at the time of the assault, and therefore satisfies the purpose test.

However, that is not enough. He must show that "use or operation of an automobile directly causes an impairment." The Arbitrator did not accept that Mr. Kumar hit his head on the steering wheel or sustained any other injury when the car went into the ditch. She found that his only injuries "were the result of the blows to the head with a rock." There was no evidence that Mr. Kumar's attacker attempted to seize control of the vehicle, hijack it or steal it. There was no evidence that Mr. Kumar's attempt to escape the vehicle caused him additional injuries. Even if the seatbelt impeded his escape, there is no evidence the delay caused an impairment. The vehicle played no direct role in Mr. Kumar's impairments.

Mr. Kumar submits that under the "moving cause" test adopted in Petrosoniak and Correia, "[d]irect cause does not mean the only cause or the most immediate cause." However, in both those cases, it was use or operation of an automobile that set in motion the chain of events resulting in injury. As counsel for Coachman stated, these cases asked when the ball stops rolling; the issue in this case is whether it starts.

Relying on Athey v. Leonati, Mr. Kumar submits that use or operation of the vehicle directly caused his impairment because it made more than a minimal (de minimis) contribution. The issue in Athey v. Leonati was apportionment of liability and damages between multiple tortious and non-tortious causes. These questions do not arise in Mr. Kumar's case because the Arbitrator found that his injuries arose entirely from the assault. It seems to me that what Mr. Kumar is really arguing is a "but for" test: the assault would not have happened if he had not been driving his cab that day. But that approach has been rejected as over-broad in the pre-SABS-1996 cases. It falls far short of "directly causes."

Judges and Commission adjudicators have commented on the difficult line-drawing exercise involved in sketching the outlines of "accident." In this case, Mr. Kumar would clearly not be entitled to accident benefits if he were attacked while walking home one night. Just as clearly, he would be entitled if his car ran into the ditch and he banged his head on some part of the car without the involvement of any attacker. In Assaf and Commercial Union, Arbitrator Blackman began his analysis of the law by recognizing the plasticity of the term:

For Mr. Assaf to be entitled to benefits under the Schedule, he must have sustained an impairment as a result of an accident. Justice Holmes has described "accident" as:

a chameleonic term, taking on different hues and shades of meaning in different circumstances, context and classes of cases . . . It is indeed a term "susceptible of being given such scope that one would hardly venture to define its boundaries."⁴⁶

He returned to the point in Karshe, quoting from the leading "proximate cause" case, Leyland Shipping Company Limited v. Norwich Union Fire Insurance Society, Limited:

Causes are spoken of as if they were as distinct from one another as beads in a row or links in a chain, but - if this metaphysical topic has to be referred to - it is not wholly so. The chain of causation is a handy expression, but the figure is inadequate. Causation is not a chain, but a net. At each point influences, forces, events, precedent and simultaneous, meet; and the radiation from each point extends infinitely. At the point where these various influences meet it is for the judgment as upon a matter of fact to declare which of the causes thus joined at the point of effect was the proximate and which was the remote cause.

. . . To treat proximate cause as if it was the cause which is proximate in time is, as I have said, out of the question. The cause which is truly proximate is that which is proximate in efficiency. That efficiency may have been preserved although other causes may meantime have sprung up which have yet not destroyed it, or truly impaired it, and it may culminate in a result of which it still remains the real efficient cause to which the event can be ascribed.

. . . Where various factors or causes are concurrent, and one has to be selected, the matter is determined as one of fact, and the choice falls upon the one to which may be variously ascribed the qualities of reality, predominance, efficiency.⁴⁷

In my view, skepticism is advised when addressing "this metaphysical topic." In trying to untangle the causation net, it is tempting to search for the precise formula that describes the boundary of "accident." The varying judicial outcomes reflect the difficulty of this exercise.

Nor has Amos entirely resolved these questions. Several post-Amos judicial decisions have held that automobile insurance covers injuries caused by assault, but these are cases where the claimant or plaintiff also suffered injuries in a motor vehicle accident, apart from the injuries suffered in the assault, or the automobile played a central and ongoing role in the commission of the assault, beyond providing the opportunity and location.⁴⁸ As I read the cases, they are relatively fact-specific. In any event, Chisolm is the only assault decision I am aware of to consider the application of Amos to the SABS-1996, and Chapnik J. adopted the analysis accepted by the Commission 's adjudicators.

"Proximate cause" is a concept from negligence law, and I am not the first to observe that its application to insurance cases is problematic, especially in the context of "no fault" accident benefits.⁴⁹ It has been held that the incident must be viewed from the perspective of the victim, and the intentional act of a third party does not preclude coverage if related to use or operation of the vehicle.⁵⁰ Instead, the issue is the scope of the coverage, as a number of Commission adjudicators have recognized. For example, the arbitrator in McAllister and Dominion of Canada General Insurance Company commented on the availability of compensation under the Compensation for Victims of Crime Act.⁵¹ In Hanlon and Guarantee, Delegate Draper conceded that driving disputes are not unusual, but stated, "the cases fall short of establishing that automobile insurance covers situations where the only involvement of the automobile is to provoke an assault."⁵² Again, in Ekunah and Simcoe & Erie General Insurance Company, about the murder of a cab driver, the arbitrator stated:

If there is doubt about the propriety of this approach in the pre-1996 cases, I agree with Justice Chapnik that it is clearly warranted by the SABS-1996 amendment:

Even given the factual circumstances as taken that the plaintiff’s motor vehicle was stopped and the bullets were shot at the automobile itself, in my view, the shooting constituted an intervening act, independent of the vehicle 's use or operation which clearly broke the chain of causation. Moreover, it appears to me that a criminal assault which does not take place within the context of the vehicle 's ordinary and well-known use is not a risk that is sought to be protected by a motor vehicle liability policy. The overall objective of the legislation is to compensate insured people injured in motor vehicle accidents. This result is not an absurdity as alleged by counsel for the plaintiff. It reflects the intention of the legislature evidenced by the legislative amendments.⁵⁴

Considering the matter from a common sense point of view, and keeping in mind the legislature's decision to narrow the scope of coverage, I find that the event that set in motion the chain of causation leading to Mr. Kumar's impairment was an assault, not use or operation of an automobile. I find that while the taxicab provided the opportunity, location and motivation for the assault, it did not directly cause an impairment. Therefore I find that the Arbitrator did not err in concluding that Mr. Kumar was not injured in an "accident" as defined in the SABS-1996.

V. EXPENSES

I decline Coachman 's motion for an order that Mr. Kumar be ordered to pay its appeal expenses. Although the Arbitrator's decision is consistent with the weight of the Commission authority, there is contrary authority, and the recent Court of Appeal decisions in Vijeyekumar and Saharkhiz raise questions about some of the Commission's earlier decisions. This is the first appellate decision to consider the application of the SABS-1996 definition of "accident " to an assault on a cab driver, and Correia, the first appeal decision about the amended definition, provides some support for Mr. Kumar's position. Mr. Kumar raised legitimate issues and conducted his appeal expeditiously. I find this an appropriate case for the exercise of my discretion to award Mr. Kumar his appeal expenses.