One car hitting another car bumper to bumper.

The Laws of Causation


In every tort action, the plaintiff has the burden to prove his or her case. He or she has to show that the wrongdoer was negligent and that he or she sustained an injury. Although it may sound fairly simple and straightforward, it is often not so. He or she must also establish a nexus between the tortious act and the damage suffered. While a plaintiff may be able to prove that a tortfeasor was negligent, the claim will be dismissed if he or she cannot establish that the accident was caused by the negligent action and the injury results from the accident. Difficulties arise in this respect where due diligence would not have prevented an accident, or where several tortfeasors are involved and it is impossible to establish which one caused the injury. Also, when a plaintiff has a pre-existing condition, the injury could be the result of the negligent action, the pre-existing condition, or a combination of both. For example, although the evidence may show that a motorist was speeding at the time of an accident, the liability of the speedster is not crystalized until the evidence also shows that speeding was a cause of the accident.

The nexus between a wrongdoing and the accident and the injury is known as causation. In a case from New Brunswick, Snell v. Farrell, [1990] 2 SCR 311 p. 326, the Supreme Court of Canada described it as follows:

Causation is an expression of the relationship that must be found to exist between the tortious act of the wrongdoer and the injury to the victim in order to justify compensation of the latter out of the pocket of the former.

As already explained, causation works on two levels: (1) the link between the negligent act and the accident; and (2) the link between the accident and the injury. It is the chain reaction composed of tort → accident → injury.

The “But For” Test

The plaintiff will usually prove causation when he or she establishes that ‘but for’ the tortious act, he or she would not have sustained the injury. The legal test of ‘but for’ applies to establish causation at both stages of liability and damages. The test will be met when, on a balance of probabilities, the evidence shows that it is more likely than not that the negligent act caused the accident and the accident caused the injury. The standard is not to be applied too rigidly: “Causation need not be determined by scientific precision” (Snell, supra p. 328). It is neither necessary for the wrongful act to be the sole cause of the accident or of the injury: Athey v. Leonati, [1996] 3 SCR 458 paras 12, 17 & 19-20. The expression ‘balance of probabilities’ does not imply that liability rests in proportion to the probability of an injury being caused by the accident (Athey v. Leonati, supra para 28):

In a negligent action, the court must declare whether the defendant was negligent, and that conclusion cannot be couched in terms of probabilities. Likewise, the negligent conduct either was or was not a cause of the injury. The court must decide, on the available evidence, whether the thing alleged has been proven: if it has, it is accepted as a certainty […].

Balance of probabilities simply refers to the weight of the evidence and the likelihood of something being as it is claimed to be.

That said, as a minimum, in the case of the nexus between an accident and an injury, the plaintiff must show that the injury is more than just a ‘possible’ consequence from the tortious act. As explained by a majority of the Supreme Court in another New Brunswick case, Lévesque v. Comeau, [1970] SCR 1010 at p. 1012:

Only one medical expert testified for [the] appellant. Although pressed, he never stated that the accident in question was probably the cause of the deafness. What he said was that this was not impossible, while also stating that it was not impossible that the deafness was occasioned by another cause. If he had testified on the one hand that the accident was the probable cause and, on the other, that it was not impossible that it be something else, then I would agree that the trial judge erred in holding that the evidence adduced showed the accident and the other possibilities as equally probable causes, but, in my opinion, such is not a proper assessment of the evidence.

The Evidentiary Rules for Causation

Regarding the evidentiary process of showing a link between an accident and an injury, a useful illustration comes from Ouellette v. Dubé (1987), 79 NBR (2d) 323 paras 28 & 34 (QB):

[28]  The plaintiff has the onus of establishing a causal relation between the accident and the symptoms which he is now experiencing. It appears obvious that the plaintiff is experiencing some discomfort and that he has muscle spasms in the back which prevent him from going back to work. However, the difficulty lies in establishing a chain of causation between that situation and the accident in which he was involved in 1977.


[34]  The causal relation must be established on a balance of probabilities. The evidence must lead us to conclude that it is more probable than not that his present physical state is due to the accident of 1977 than to any other cause. That burden requires more than a simple “it is possible that the injuries are the result of the accident”, mentioned by Dr. De Yturralde, or a simple “it is conceivable that the injury was caused by the accident”, mentioned by Dr. Parton. None of the doctors were able to state that the accident was the proximate cause of the ailments from which the plaintiff is now suffering.

However, there is no strict evidentiary rule for causation. It can be established by positive evidence of causation as well as by evidence eliminating all other possible causes, or a combination of both. Another illustration comes from Mullin v. Tozer (1986), 70 NBR (2d) 403 paras 49-57 & 62 (QB). There is no doubt that the plaintiff had sustained a bad leg fracture. However, he reported lower back pain for the first time almost 2 years after the accident. His doctor testified that the pain may have been masked by medication and forced rest, only to surface later. He had no back pain prior to the accident, but had been involved in a second non-serious motor vehicle accident later in the year. Although somewhat reluctantly, the trial judge accepted that the “back condition … probably flows from the accident in question” and held the defendant liable (para 62).

Thus, when building a record for the trial judge, the plaintiff must focus on whether the negligent act caused the accident, and whether the accident caused the injury. Failure to prove either may be fatal to his or her case. As already discussed, courts will not apply a scientific standard to the test of causation. Proximity neither means closely related in time. For example, in Chartier v. Laramée, [1969] SCR 771, the plaintiff sustained a fractured leg in an accident. After his original discharge from the hospital, while being back for further treatments, he reinjured the healing leg after accidentally putting his weight on it when losing his balance. He had to undergo further surgery and a bone graft. The unanimous bench of the Supreme Court held the reinjury to be a result of the accident. In Athey v. Leonati, supra, the unanimous bench also accepted that a disc hernia triggered when performing stretching exercises at a gym was causally related to two prior car accidents; as in Chartier, the injury sustained in the collisions fragilized the plaintiff. For a case where a subsequent injury was found unrelated to the original accident, see Marney v. Bannister (1986), 68 NBR (2d) 253 para 13 (QB).

That being said, the ‘but for’ test is impracticable in certain situations. Due to the circumstances of the cases, the burden that it imposes on a plaintiff would be impossible to discharge. The Supreme Court of Canada has therefore provided for two types of exception to the usual rule.

Exceptions to the “But For” Test?

a) Reversal of Onus or Adverse Inference

The burden of proving causation typically rests on the plaintiff, along the lines already discussed, as part of the case that needs to be met. However, sometimes, courts will quickly transfer on the defendant the burden of explaining how the negligent action is not the cause of the injury. In Snell, supra p. 328, the Supreme Court of Canada explained that “the allocation of the burden of proof is not immutable. Both the burden and the standard of proof are flexible concepts”. The shifting of the burden will happen when a defendant has particular knowledge of a matter that may have caused the injury. In such instance, “very little affirmative evidence on the part of the plaintiff will justify the drawing of an inference of causation in the absence of evidence to the contrary” (Snell, supra pp. 328-329).

Adverse inferences will usually be invoked in cases of complex origin, where science may be lacking or the information is under the control of the defendant. One such example is where a railroad company has control of the information about the breaking capacity of a particular train: Stamper v. Finnigan et al. (1986), 74 NBR (2d) 271 paras 221-237 (QB), confirmed at (1988), 85 NBR (2d) 404 para 60 (CA), although the Court of Appeal saw enough evidence without the need to rely on adverse inferences. Medical malpractice is another field where adverse inferences are used, given the specific knowledge held by the defendants.

In Snell, supra, the ophthalmologist had pursued the cataract operation despite retroorbital bleeding. The removal of the cataract itself had caused some bleeding in the eye and it was impossible to determine from which action the optic nerve atrophy came from. Had the doctor postponed the operation when the bleeding behind the eye was noticed, its effect on the eyesight could have been better determined. The pursuit of the surgery when it should have been stopped made it impossible to identify the precise cause of the blindness. The matter was complicated further by pre-existing conditions of the plaintiff. Nevertheless, all three levels of courts applied the adverse inference rule and the defendant failed to provide a satisfactory explanation to the plaintiff’s condition and was therefore liable.

Technically, the ‘but for’ test remains the same, but the burden resting on the plaintiff will be lightened somewhat. Failure on the part of the defendant to provide an acceptable explanation will lead to an adverse inference of causation.

b) Material Contribution to the Risk of Injury

In specific types of cases, the courts will stray from the usual ‘but for’ test to establish causation. This scenario is distinct from the one of adverse inferences which is available anytime a defendant has control over the relevant information. Where several negligent tortfeasors contribute to an injury and their actions confound which one has caused the damage, the presiding judge will rely on the test of ‘material contribution to the risk of injury’: Cook v. Lewis, [1951] SCR 830; Snell, supra pp. 321-322 & 327. In Clements v. Clements, [2012] 2 SCR 181 paras 13-14, the majority of the Supreme Court explained:

[13]  To recap, the basic rule of recovery for negligence is that the plaintiff must establish on a balance of probabilities that the defendant caused the plaintiff’s injury on the “but for” test. This is a factual determination. Exceptionally, however, courts have accepted that a plaintiff may be able to recover on the basis of “material contribution to risk of injury”, without showing factual “but for” causation. As will be discussed in more detail below, this can occur in cases where it is impossible to determine which of a number of negligent acts by multiple actors in fact caused the injury, but it is established that one or more of them did in fact cause it. In these cases, the goals of tort law and the underlying theory of corrective justice require that the defendant not be permitted to escape liability by pointing the finger at another wrongdoer. Courts have therefore held the defendant liable on the basis that he materially contributed to the risk of the injury.

[14]  “But for” causation and liability on the basis of material contribution to risk are two different beasts. “But for” causation is a factual inquiry into what likely happened. The material contribution to risk test removes the requirement of “but for” causation and substitutes proof of material contribution to risk.

[Underlining added]

The typical scenario invoked to illustrate the test of material contribution to the risk of injury is the one found in Cook v. Lewis, supra. A plaintiff was injured by a gunshot during a hunting trip. Both defendants had fired at the same time and in the same direction, despite being aware of the potential presence of the plaintiff. One of the shots injured the victim but the jury could not determine which one. The plaintiff had established that both defendants were negligent and that he was injured by their tortious act, but could not establish which one had fired the shot. The Supreme Court confirmed the order for a new trial.

Another example of the doctrine can be found in the case of Wood v. Wawanesa, 2019 NBQB 125. Three motorists encountered at least 2 large bags of insulation on a 4-lane highway that had torn open and allowed their content to escape. It created blizzard like conditions which affected the drivers’ visibility. Nobody had seen where the bags came from, but the trial judge concluded that “it is more likely than not that the bags of insulation fell off an unidentified vehicle of some kind, most likely a truck or pick-up, where the operator had not ensured that the load was safely secured” (para 23). The condition created contributed to causing the collision and the unknown driver was found negligent. A chain reaction thus followed where the plaintiff’s SUV rear-ended a car before being also rear-ended. She sustained a fractured wrist in the collision but the evidence could not determine its cause: there had been two highspeed collisions, airbags going off, beating down the airbag to escape a death trap, struggling to open a jammed door, etc. The trial judge was not able to attribute the injury to a specific event (paras 123 & 125) and thus shared liability between all at fault parties (para 127). Although not stating so expressly, he thus applied the ‘material contribution to the risk of injury’ test.

As all other factual determinations, the application of the tests for causation requires a thorough understanding of all the facts of the case. It is wise to have proper evidence of the probable cause of the accident and of the injury, or at least proper evidence excluding all possible causes other than the one relied upon.

This paper is offered for the purpose of discussion only. It does not constitute legal advice and its distribution does not create a solicitor-client relationship. Please consult a lawyer if you require legal advice.