The truth about factual causation issues in most personal injury actions is that, where there’s any issue at all, it’s not one that involves difficult questions of law or any need to venture into realms of philosophy or metaphysics, abstract, concrete or something else. Instead, the issue is whether the evidence is enough to establish that the alleged negligence was a sufficient legal cause of the injury. The evidence may be complicated. The decision may require the trier-of-fact, judge or jury, to make difficult choices, such as who to believe or what conflicting evidence to accept or reject. In the not-trite aphorism that lawyers involved in the dispute resolution portion of the legal system soon learn, the law in a particular case is clear. What isn’t clear is how to apply the law to the facts.

I’m going to mention a 4 articles published in the last 3 years which should interest members of the legal profession, and others, who, for whatever reason, are interested the subject of causation in negligence. These articles may help to clarify, for some readers, some of the murkiness in the current state of Canadian common law case-law governing proof of factual causation in negligence actions.

Don’t worry, I’m not the author of any of them. That means the articles are short enough for convenient reading. Three of the articles are by leading academic lawyers. The fourth is by a lawyer with experience in the Canadian civil litigation, medical malpractice, trenches.

The articles are:

Jane Stapleton, “Unnecessary Causes” (2013)  129 LQR 39.

René Brewer, “The End of Material Contribution to Injury (2013) 42 Adv Q 217.

Jane Stapleton, “An ‘Extended But-For’ Test for the Causal Relation in the Law of Obligations” (2015) Oxford Journal of Legal Studies.

Oxford J Legal Studies (2015)
doi: 10.1093/ojls/gqv005
First published online: March 7, 2015

Ernest Weinrib “Causal Uncertainty” (2015) Oxford Journal of Legal Studies.

Oxford J Legal Studies (2015)
doi: 10.1093/ojls/gqv020
First published online: July 2, 2015

Unless you know somebody, the last 2 are currently available online, only.

I discuss these articles, and their significance to Canadian jurisprudence, after the break. Most of the discussion deals with Professor Weinrib’s article.

Warning: the entire post is about 6,400 words according to Word, but the wording counting tool in the version of Word I’m using counts parts of citations as words, so there’s a bit less than that.

I’ll start with René Brewer’s “The End of Material Contribution to Injury (2013) 42 Adv Q 217, because, if you’re a practitioner, and don’t want to read anything else, read this one. Also, I’ll be brief.

Based on her footnotes, she’s read large sections, if not all, of my articles  “Factual Causation In Negligence After Clements” (2013) and “The Snell Inference and Material Contribution: Defining the Indefinable and Hunting the Causative Snark” (2005).

Despite that, Ms. Brewer makes some valid points about Clements, the but-for test, and Snell “inference approach” and how Clements’ explanation of Snell could be seen to be very plaintiff friendly.  While she’s correct about that possibility,  I haven’t seen, at least in reported cases – take from that as you will – any indication that anything that’s been written by the SCC, starting with Resurfice, about proof of factual or legal causation  in negligence cases has made any significant difference to actual results since 2007, compared to what the situation was between 1997 when Athey was released and 2007 when Resurfice was released.  Judges have, for the most part, changed their terminology to conform to new law, but that’s cosmetic.

Putting it another way, I have yet to read, or hear of, any reasons where a trial judge said something to the effect of: “I have found that the plaintiff has failed to establish, on the balance of probabilities, that the defendant’s negligence was a but-for cause of the [the injury, the accident]; however, if there were still a material contribution to injury alternative for establishing factual causation, as the Athey material contribution test was understood before Resurfice. I would have found that the plaintiff had satisfied that alternative.”

In addition, many lawyers who practised personal injury defence in the Athey “material contribution to harm” interregnum  – you know: the period, which we’re now told, occurred as a result of the Bar’s and Bench’s misunderstanding of  Athey – will claim that the judicial approach to the use of the Athey material contribution to harm test as alternative method for establishing factual causation was already very plaintiff friendly. Plus ça change?

Should you decide to read Ms. Brewer’s article, keep in mind that it remains to be seen what the SCC had in mind in Henry v. British Columbia (Attorney General), 2015 SCC 24 where (albeit not in a negligence-causation context) the majority cited Clements for the proposition that there is a “materially contributed to the harm suffered” version of method for establish factual causation and the concurring judges did not challenge this statement but described it as a version of a but-for test. I discussed Henry last May in this blog. The author of the concurring reasons was Chief Justice McLachlin.

Professor Weinrib explains why and how corrective justice justifies the conclusion that, in some instances,  a plaintiff should be permitted to satisfy the causation requirements in negligence even though the evidence does not allow a valid conclusion that the negligence of a defendant, or group of defendants, satisfies the but-for test on the balance of probabilities, in the sense that the evidence justifies the conclusion that it was the negligence of that defendant, or that group of defendants, which was necessary for the occurrence of the plaintiff’s injury. The evidence establishes what caused the injury. It just doesn’t sufficiently identify who caused the injury.

The abstract for Professor Weinrib’s article, “Causal Uncertainty”, which you can find here is:

“This article examines, from the standpoint of corrective justice, three basic situations of causal uncertainty in the law of negligence. In the first situation the uncertainty is about which of the possible defendants wrongfully caused the injury. In the second situation the uncertainty is whether the injury resulted from the innocent or the wrongful aspect of the defendant’s activity. In the third situation the uncertainty is whether the defendant’s negligence failed to prevent the materialisation of an independently existing risk. In viewing the parties as the doer and the sufferer of the same injustice, corrective justice confirms the centrality of causation to negligence liability. The article shows, nonetheless, how these three situations of uncertainty can be resolved within the corrective justice framework. Moreover, what emerges from the corrective justice analysis is that causal uncertainty is not a homogeneous problem; its solution varies with the kind of situation in which the uncertainty appears.”

The article is not an attempt to make existing case law more coherent. Professor Weinrib explains, at pp. 5-6:

  “The failure to recognise the difference between these problems has itself contributed to confusion about causal uncertainty. This is especially evident in connection with the first two. One reason that the English jurisprudence is tied up in knots is that is employs judgments about the second problem to address the issues that arise under the first one. Conversely, the Canadian jurisprudence treats the first problem as paradigmatic and then marginalizes the second. Clarity about the distinctiveness of the problems is a practical, as well as a theoretical, desideratum.

  In any case, my strategy here is to stand back from the extensive and ramified case-law that has developed over the last decades. Instead, I focus on the basic situations represented by the paradigmatic cases from which that case-law has been elaborated. If, as I have just suggested, the case-law in various jurisdictions has gone astray by ignoring the distinctions between these situations, little progress can be made by taking the present case-law as the starting point for analysis. More promising is a re-examination of the normative contours of the basic situations themselves. This is what I attempt here. The normative contours in question are those of corrective justice, which in my view provides the only plausible general understanding of private law relationships.” [footnotes omitted]

The basic situations were set out earlier, at pp. 4-5:

  “This article concerns three basic problems of causal uncertainty in the law of negligence. One of these arises when a wrongfully injured plaintiff cannot prove who the wrongdoer was. This is the problem of identifying the tortfeasor. The classic example is the case of the negligently injured victim of a hunting accident who cannot prove which of the other hunters fired the shot that wounded them. The other two problems arise when the plaintiff cannot prove that the defendant’s negligent conduct, rather than a concurrent innocent factor, was the cause of the injury. At issue here is whether the injury was the result of the tortious or the innocent factor. One problem involves an innocent factor that originates in the defendant’s activity … [as in the case] … where an employee alleged that his dermatitis was caused by his employer’s admittedly negligent failure to provide shower facilities, but could not show that he would not have contracted dermatitis even if he had showered. The other problem is the result of an innocent factor that arose independently of the defendant’s activity. The paradigmatic example is the defendant’s misdiagnosis of a medical condition from which the patient would probably not have recovered in any case. It is important to recognise that these are different problems.” [Footnotes omitted, words in brackets added.]

A consequence of this “re-examination of the contours of the basic situations”, is that Professor Weinrib provides a normative rationale and justification, at least in some instances, for the dictum, in Clements v. Clements, [2012] 2 SCR 181, 2012 SCC 32, that fault and creation of risk may, in some circumstances, satisfy the causation requirements in tort, in the limited circumstances set out in Clements through the application of the “material contribution to risk doctrine”. That is, Professor Weinrib provides a legitimate theoretical basis, consistent with existing doctrine – beyond the fact that the SCC declared a rule of law –  for allowing a plaintiff to satisfy the causation requirement of the tort, in certain instances, even though the plaintiff is unable to establish, on the balance of probability, that the negligence of any defendant was necessary for the occurrence of the injury the plaintiff suffered.

Here, I get to to admit that “Causal Uncertainty” leave me eating half a plate of crow. I have argued, since the release of Clements, that Professor Weinrib’s corrective justice explanation of negligence law did not justify the Clements material contribution to risk test. As I read “Causal Uncertainty”, he has shown that corrective justice, as he explains it, can justify the use of that doctrine in some instances.

Readers of this blog (and many others who, for whatever reason, don’t read this blog) will recall that the circumstances under which the Clements material contribution to risk doctrine is said to apply are:

“[33]    We have seen that the jurisprudence establishes that while tort liability must generally be founded on proof that “but for” the defendant’s negligence the injury would not have occurred, exceptionally proof of factual causation can be replaced by proof of a material contribution to the risk that gave rise to the injury.

[34]     In Resurfice, this Court summarized the cases as holding that a material contribution approach may be appropriate where it is “impossible” for the plaintiff to prove causation on the “but for” test and where it is clear that the defendant breached its duty of care (acted negligently) in a way that exposed the plaintiff to an unreasonable risk of injury.  As a summary of the jurisprudence, this is accurate.  However, as a test it is incomplete.  A clear picture of when “but for” causation can be replaced by material contribution to risk requires further exploration of what is meant by “impossible to prove” (Resurfice, at para. 28) and what substratum of negligence must be shown.  I will discuss each of these related concepts in turn.

…

[39]      What then are the cases referring to when they say that it must be “impossible” to prove “but for” causation as a precondition to a material contribution to risk approach?  The answer emerges from the facts of the cases that have adopted such an approach.  Typically, there are a number of tortfeasors.  All are at fault, and one or more has in fact caused the plaintiff’s injury.  The plaintiff would not have been injured “but for” their negligence, viewed globally.  However, because each can point the finger at the other, it is impossible for the plaintiff to show on a balance of probabilities that any one of them in fact caused her injury.  This is the impossibility of which Cook and the multiple-employer mesothelioma cases speak.

…

[46] (2)     Exceptionally, a plaintiff may succeed by showing that the defendant’s conduct materially contributed to risk of the plaintiff’s injury, where (a) the plaintiff has established that her loss would not have occurred “but for” the negligence of two or more tortfeasors, each possibly in fact responsible for the loss; and (b) the plaintiff, through no fault of her own, is unable to show that any one of the possible tortfeasors in fact was the necessary or “but for” cause of her injury, because each can point to one another as the possible “but for” cause of the injury, defeating a finding of causation on a balance of probabilities against anyone.”

To make sense of [46](2)(b) it’s essential to remember this about para. [46](2)(a)’s “where … the plaintiff has established that her loss would not have occurred “but for” the negligence of two or more tortfeasors, each possibly in fact responsible for the loss”.  What the “possibly in fact” means is tha the but-for test does not, in fact, identify which tortfeasors negligence was a cause on the balance of probability. All it tells us is that, with respect to each of the tortfeasors, their negligence could satisfy the but-for test in certain circumstances. The obvious circumstance is if the negligence of all of the other tortfeasors had not occurred. Another possibility, in some circumstances, could be if the negligence of some of the other tortfeasors had not occurred.

The problem described in clause (a) is that the evidence is not enough to single out one, or any combination, of the multiple tortfeasors, even all of them, on the balance or probability. The point of the “any combination, even all” is that the evidence is not enough for the court to say: “aha, the evidence is enough to identify you, or the group of you, and exclude the others. It may not be the case that all of the combinations of one or more of the tortfeasors are equally likely. What is the case is that the evidence doesn’t reach the balance of probabilities in respect to any one tortfeasor, or any combination of them – doesn’t allow the identification of one wrongdoer or a specific combination. That’s why there’s finger-pointing.

Now, Professor Weinrib doesn’t claim that his explanation for the manner by which corrective justice provides

1. a principled, normative, justification,
2. consistent with existing negligence doctrine,
3. for the position that the causation requirements of the negligence cause of action may be satisfied, in certain circumstances, against certain wrongdoers
4. even though the evidence, in the particular case, does not permit a valid (i.e. consistent with incontestable principles of logic) conclusion
5. that, as against a particular defendant, the plaintiff has satisfied the causation requirements on the balance of probabilities

justifies the Clements material contribution to risk doctrine.

I do.

Professor Weinrib doesn’t mention Clements except in two footnotes, the first of which is attached to this statement about the state of Supreme Court of Canada decisions attempting to explain proof of factual causation: “two incompletely successful recent efforts by the Supreme Court of Canada to sum up and clarify its own approach”. (See the text associated with note 5. The other reference is in note 14 which points out another problem in Clements. But, I’m not here, now, to again complain about Clements, so I’ll move on.)

He explains how corrective justice supports the imposition of liability (assuming all of the other requirements of the cause of action are satisfied) in the two wrongdoer example known in Canadian tort law as Cook v. Lewis, [1951] SCR 830, 1951 CanLII 26 (SCC), known in scholarly and philosophical circles as the “2 hunter” example. The problem in this example is that the evidence shows that it was one of the two negligent hunters who necessarily fired the shot that injured the plaintiff. However, the evidence doesn’t permit a valid conclusion, on the balance of probability, as to which of the two it was.

Some readers (arguably but not necessarily those of you who have, as needed, adventured with scholars, even philosophers) should be able to figure out where I’m going, if not necessarily where I’ll end and how I’ll get there.

Professor Weinrib points out in that in the two hunter example one of the two wrongdoers necessarily injured the plaintiff through negligence. It was the bullet that hunter shot which struck and injured the plaintiff. While the bullet from the other hunter’s shot did not strike the plaintiff, the fact and manner of that hunter’s shot is what, on the evidence, prevents the plaintiff from being able to establish which of the two hunters fired the bullet that actually struck the plaintiff.  All that the evidence allows is the conclusion that the negligence of one of the two resulted in a bullet striking and injuring the plaintiff. The negligence of the other impaired the plaintiff’s ability to establish which of the two wrongdoers committed the negligence that caused the injury to the plaintiff’s body. But, we don’t know which one is which. So we’re left with the situation where both could have been responsible for one or the other of the events; but obviously not both events.

Here’s where the scholarly stuff comes in. (Some might even call it …. ssshhhh …. abstract metaphysical theory).

For reasons that have to do with the meaning of corrective justice, when properly understood – don’t forget that Clements refers to corrective justice as the basis for the causation requirement in tort: see para. 7 and para. 37) – corrective justice focuses on breach of right, not (just) causing injury. The injury is the evidence of the breach. The injured person’s right to not be injured by wrongful conduct implies the existence of a viable remedy to compensate for wrongful breach of that right by restoring, in some fashion, the relationship between the injured person and the wrongdoer to what it was before the breach of right occurred. That remedy is part of the right; not separate from it.

What we have, then, in the Cook v Lewis situation is a situation where the negligence of each of the wrongdoers in some way breached the injured person’s right; however, we don’t know, because of the lack of evidence, know which negligence was the cause of of which breach and, thus, which negligence actually caused the injury for which the plaintiff seeks compensation.

Be that as it may, in the Cook v Lewis situation – the two wrongdoer situation: it’s a paradigm of what’s called alternative (or alternate causation) because, in the real world, it was actually one or the other of the two whose negligence caused the physical injury –  the fact that the multiple wrongdoers are each necessarily responsible for a wrongful breach of an aspect of the injured person’s right is normatively sufficient to justify the imposition of liability on both wrongdoers (assuming all other aspects of the cause of action are satisfied). Remember that, despite how it was often summarized, the rule in Cook v. Lewis wasn’t a liability rule, it was a causation rule. It was just that the only issue on the facts of Cook was the causation issue.

Professor Weinrib points out, correctly, that his corrective justice based explanation of the result in Cook v. Lewis applies even if there are more than two wrongdoers, provided (again) that all of the (relevant) alternative choices, as causes, are wrongful. That’s because it is not based on the number of wrongdoers involved, so long as there’s more than one but on the nature of the relationship between each of the wrongdoers and the injured person.

That move, although he doesn’t claim this in the article – I do, here – is what allows this corrective justice rationale for the result in Cook – for the conclusion that the causation requirement may be satisfied in some instances where the evidence doesn’t permit proof of factual causation on the balance of probabilities – to be extended to the Clements explanation of the requirements for material contribution to risk.

To be clear, the explanation provided by Professor Weinrib, on its face, does not apply to all instances of multiple tortfeasors which could satisfy the requirements of the Clements material contribution to risk doctrine, as explained in para. 46(2).

That’s because the problem of lack of evidence which creates the problem addressed in Clements –  the ability of each of the multiple wrongdoers to point the finger at another wrongdoer , where “each can point to one another as the possible “but for” cause of the injury” – isn’t limited to instances of alternate (alternative) causation as existed in Cook v Lewis. 

Recall that what the Clements example means  – using a 3 tortfeasor example – is that the evidence is not enough for the trier-of-fact to conclude, on the balance of probability, that it was any one of the 3, or any combination of the 3 (including all three) not all three, whose negligence was necessary. That is, the evidence isn’t enough for a conclusion on the balance of probability that the probable cause was necessarily a particular 1 of the 3, a particular 2 of the 3, or all.

An example of a situation which, on its face, literally falls with the Clements‘ explanation of when the material contribution to risk test applies is one where – again using 3 tortfeasors – on the evidence, the negligence of each, ignoring the negligence of the other two, was sufficient (together with all of the other required conditions) to cause the injury. This is called duplicative or oversubscribed causation. It is a a doctrinal problem of too much evidence identifying the negligence of too many negligent actors as but-for causes, not one of not enough evidence to identify, to the required standard of proof, even one wrongdoer, or a group of wrongdoers, as the person or persons responsible for a but-for cause.

If you’ve read this far, you’ll say “hold on there, Cheifetz, you know a wrongdoer can’t argue “but there’s somebody else at fault, whose negligence was also; that’s the joint liability rule”, or something like that. I’ll respond, yes, but you’re conflating (a) causation and fault with (b) liability AND (c) the problem is that, where there are multiple, independently (of each other) sufficient, negligent causes, then by definition the but-for test can’t apply, since none of them are necessary since each of the others is, etc.

So, you reply, “well that’s what Clements material contribution is premised on, right?”

Actually no. It’s premised on there not being enough evidence to allow the plaintiff to validly, in a forensic sense, point the (right) finger at any one of the group of tortfeasors, or any particular combination of the group, including all of them, even though the evidence establishes that the negligence at least one, but not all, of  the group, would be a but-for cause, if the negligence of the other(s) had not occurred. In our 3 tortfeasor example, as I indicated, the evidence isn’t enough for a conclusion on the balance of probability that the probable cause was necessarily a particular 1 of the 3, a particular 2 of the 3, or all. That’s merely a forensic problem in the application of the but-for test.

We have a different, doctrinal problem, not merely forensic, problem where we have enough evidence to establish, to the required standard of proof, two or more independently (of each other) sufficient causes. Each one makes the other one “unnecessary”. That’s why, doctrinally, the but-for test, defined in terms of necessity (the orthodox explanation) is, as Clements states it is (at para. 8) necessarily fails.

“The test for showing causation is the “but for” test. 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. This is a factual inquiry. If the plaintiff does not establish this on a balance of probabilities, having regard to all the evidence, her action against the defendant fails.”

(Emphasis in original.)

Put in other words, there’s no evidential causal uncertainty, in the Clements mc-r scenario, as to what the cause was, just uncertainty as to who to blame among the wrongdoers.

If you want to call Clements mc-r a responsibility (liability) rule, assuming all of the other requirements of the cause of action are satisfied, go ahead. You’ll be correct enough, in practice, because, in practice, unless the plaintiff is able to satisfy all of the other requirements of the cause of action, in the particular cases, it’s worth asking how many judges will (want to) undertake the remainder of the Clements material contribution to risk analysis.

What remainder you ask? Isn’t all that’s required, on the causation issue, is that the plaintiff show the facts of the case fit with para. 46(2) of Clements?

I say probably not.

Why? Recall that Clements purports to clarify the material contribution to risk test first declared in Resurfice Corp. v. Hanke, [2007] 1 SCR 333, 2007 SCC 7. As explained in Resurfice, the mc-r test had at least two, or maybe three, conditions which the plaintiff had to satisfy.

“24    However, in special circumstances, the law has recognized exceptions to the basic “but for” test, and applied a “material contribution” test.  Broadly speaking, the cases in which the “material contribution” test is properly applied involve two requirements.

25    First, it must be impossible for the plaintiff to prove that the defendant’s negligence caused the plaintiff’s injury using the “but for” test.  The impossibility must be due to factors that are outside of the plaintiff’s control; for example, current limits of scientific knowledge.  Second, it must be clear that the defendant breached a duty of care owed to the plaintiff, thereby exposing the plaintiff to an unreasonable risk of injury, and the plaintiff must have suffered  that form of injury.  In other words, the plaintiff’s injury must fall within the ambit of the risk created by the defendant’s breach.  In those exceptional cases where these two requirements are satisfied, liability may be imposed, even though the “but for” test is not satisfied, because it would offend basic notions of fairness and justice to deny liability by applying a “but for” approach.”

The Court said two, explicitly. I say maybe three. There was, according to Resurfice, a third if also had to be shown, in the particular case, that “it would offend basic notions of fairness and justice to deny liability by applying a ‘but for’ approach. At least to me, it wasn’t (and still isn’t) certain what Resurfice should be understood to means on that issue. I’ve written that the quoted phrase seems to be an additional requirement. That is, the “mere” fact that the fact that the facts, in a particular case, satisfy the first and second criteria does not, on its on, seem to necessarily provide a “yes” answer to the “offend basic notions …” issue. (The most up-to-date, published, version of that explanation starts a p. 262 of my “Factual Causation after Clements” (2013) 41 Adv. Q. 170. The discussion is short enough, covering only 25 pages.)

The point is that Clements restated, explained, chose your term, the meaning of the first Resurfice mc-r requirement: the impossibility requirement. I’ve quoted the string of relevant paragraphs again, beginning with para. 34. I’ll quote that paragraph again.

“[34]    In Resurfice, this Court summarized the cases as holding that a material contribution approach may be appropriate where it is “impossible” for the plaintiff to prove causation on the “but for” test and where it is clear that the defendant breached its duty of care (acted negligently) in a way that exposed the plaintiff to an unreasonable risk of injury. As a summary of the jurisprudence, this is accurate. However, as a test it is incomplete. A clear picture of when “but for” causation can be replaced by material contribution to risk requires further exploration of what is meant by “impossible to prove” (Resurfice, at para. 28) and what substratum of negligence must be shown. I will discuss each of these related concepts in turn.”

Clements, however, is silent on the balance of the Resurfice condition(s), unless we find hidden meaning in the concluding sentence of paragraph 45. There, after rejecting aspects of what the BCCA had said about the material contribution test as set out in Resurfice, the majority stated, in the last sentence: “However, in broad terms, the Court of Appeal correctly identified the circumstances where a material contribution to risk approach may exceptionally be imposed.”

If we’re to look at what the BCCA wrote that the Court might have been referring, we need go no farther than 2010 BCCA 581, paras. 63 and 64. Paragraph 63 begins: “In summary, having regard to the over-arching policy that the material-contribution test is available only when a denial of liability under the but-for test would offend basic notions of fairness and justice”. Paragraph 64 concludes: “I do not consider it either unfair or unjust, or … “just plain wrong” not to fix Mr. Clements with liability when Mrs. Clements has been unable to show factually that his negligence was a cause of her damages.”

Draw your own conclusions.

Clements also leaves us a doctrinal problem created by the Court’s failure, in Clements, to remember that there are doctrinal problems in the application of but-for, not just evidential.

That’s why there’s still plenty of room for more spilled ink, or better put in this millennium, excited pixels.

Will this make a difference, in practice, in the next lawsuit(s) arising out of a multiple vehicle collision where we know, from the evidence, that, in real terms, the consequences (on the people already in the pile up) of some of the conditions that resulted from some of the later trucks that drove into the pile, didn’t make a difference to what their injuries would have been if those later conditions had not occurred, because of what had already occurred?

Will the driver of truck 4 be able to say that her negligence didn’t make a difference given what were the necessary consequences of the negligence of the drivers of trucks 1, 2 and 3, let alone the train locomotive engineer, because, as it happened, the train also crashed into the pile up. It was on the tracks at a flat crossing. The train would have had time to stop had the engineer been paying attention. The pile-up was on the tracks because the barrier came down suddenly trapping a vehicle. It was still there when the chain of collisions started. The barrier descended because it had been damaged a few days earlier. The railway had been alerted but failed to respond timely.

It turns out, for what it’s worth, that even had the engineer applied his brakes on time, the train wouldn’t have stopped because the locomotive repair shop – in this case, not the railway company operator – hadn’t serviced the brakes properly.

But we don’t need so complicated a situation.

If your eyes (or mind) haven’t glazed yet, ask yourself what the judges of the courts in your jurisdiction would do with the facts of a simple example. A vehicle driven by D, (owned by a garage G, rented to D) strikes the vehicle of P. P is injured. D’s negligence was in not paying attention such that D never applied the brakes, at all, before impact. Had D been paying proper attention, D would have had time to stop. But, as it turns out, even if D had been paying attention, and applied the brakes properly, that wouldn’t have made a difference.The brakes had failed at some point before the accident, but D didn’t know that . The brakes had failed because of negligent work on them by (the mechanic) of the garage G.

Or make it even simpler. In the morning, you take your car in to the dealership to have the brakes serviced. You get the car that evening. The accident occurs as you are driving to wherever it is you planned to go.

Whose negligence was a but-for cause? Can D say: “but my negligence wasn’t necessary, because even if I had applied the brakes, that wouldn’t have made a difference”? Can G say: “but the mechanic’s negligence wasn’t necessary, didn’t make a difference because D never applied the brakes”?

These are the facts of a real U.S. case: Saunders System Birmingham Co. v Adams, 217 Ala. 621, 117 So. 72 (Ala. 1928). You can read the case, here. Adams (the plaintiff) sued only the garage. Adams lost.

That brings us to the second new article lawyers and judges interested in the workings of factual causation should read: Jane Stapleton, “An ‘Extended But-For’ Test for the Causal Relation in the Law of Obligations” (2015) Oxford Journal of Legal

Professor Stapleton outlines an extended version of the but-for test which, unlike the orthodox version, is capable of providing valid answers to instances of factual causation to which the orthodox but-for test,  because of the manner in which it is defined, cannot be validly  applied, in the sense that if it is applied the result is nonsensical.

abstract here

“This article explores the question of what character relations must have before the orthodox law of obligations will describe them as ‘causal’ relations. The article does not purport to identify the metaphysical nature of ‘causation’. Instead it provides a non-reductive account of what is essential before the law has described the relation between a specific factor and the existence of a particular indivisible phenomenon as ‘causal’. Section 1 presents a simple test for this relation—an ‘extended but-for test’—that can be deployed in a straightforward way without engaging with theoretically complex and often problematic accounts of causation based on the notion of sufficient sets, such as Wright’s NESS account. Section 2 demonstrates how important principles relating to the separateness of a legal entity and to legal responsibility can resolve theoretical puzzles and in turn illuminate why the orthodox law of obligations does not choose to describe as ‘causal’ a relation wider than the one identified in this article.”

Those of you interested in more will have to adventure with the scholar and philosophers, even read footnotes. For those who interested, Prof. Stapleton concedes that even her version of the but-for test doesn’t provide an answer to the Saunders System problem.

Jane Stapleton “Unnecessary Causes” (2013)  129 LQR 39. The more recent Stapleton “Extended But-For” builds on this paper and her earlier work. “Unnecessary Causes” discusses Clements mc-r in some depth at pp 62-63.

This post may be considered a continuation of a few of my earlier posts on the subject of “adventuring with philosophers”, including these two:

https://davidcheifetz.ca/2012/10/27/adventuring-with-philosophers-part-1/

https://davidcheifetz.ca/2012/11/04/causation-in-the-air-or-adventuring-with-philosophers-manque-part-2/

Plus ça change, plus c’est la même chose

 If you’ve read this far, then bear with me just a bit longer. Recall this statement in Clements, at para. 28, explaining the results in the SCC’s decisions in Snell, Athey, Walker Estate and Resurfice:  Snell, Athey, Walker Estate and Resurfice were all resolved on a robust and common sense application of the “but for” test of causation.” In it its first explanation of the “how” of the but-for test’s application, the Court wrote at para. 9 of Clements: “The “but for” causation test must be applied in a robust common sense fashion.” At para. 10, the Court referred to “A common sense inference of “but for” causation from proof of negligence [that] usually flows without difficulty”. [Word in brackets added.] The Court reiterated this explanation at para. 23: the “robust, common sense application of the “but for test, in accordance with Snell“.  Finally, however, at para. 46(1), the Court wrote, in part: “As a general rule, a plaintiff cannot succeed unless she shows as a matter of fact that she would not have suffered the loss “but for” the negligent act or acts of the defendant. A trial judge is to take a robust and pragmatic approach to determining if a plaintiff has established that the defendant’s negligence caused her loss.”

Anybody care to explain what happened to “common sense”? Should we understand that the “robust and pragmatic” approach entails? amounts to? applying common sense? If “robust common sense” and “robust and pragmatic” have the same meaning, does that entail that “pragmatic” is a synonym for “common sense”. If not, and the absence of “common sense” was an oversight, is the approach the robust, pragmatic, common sense approach”?

Anyway, the Court, in Clements, didn’t explain, as such, what we should understand to be the content of the mandated approach to the application of the but-for test; in other words, the “how” rules. But, I suppose, the Court did provide the “why”: to achieve “corrective justice” between injured person and wrongdoer (paras. 7, 9, 19, 21, 41,); to “me[e]t the goals of negligence law of compensation, fairness and deterrence, in a manner consistent with corrective justice” (para. 19); “To allow recovery where the injury was the result of neutral factors would neither further the goals of compensation, fairness and deterrence, nor comport with the theory of corrective justice that underlies the law of negligence” (para. 21). Those of you interested in a recent appellate attempt to provide content to the robust and pragmatic approach should read  St-Germain c. Benhaim, 2014 QCCA 2207 (CanLII) – see the  majority reasons of KASIRER and BÉLANGER, JJ.A (in English). The Ontario Court of Appeal decision in Aristorenas v. Comcare Health Services, 2006 CanLII 33850 (ON CA) leave to appeal to SCC denied 2007 CanLII 10550 (SCC), which also attempted an explanation, is explained in St-Germain.  Aristorenas was applied in Ediger v. Johnston, 2011 BCCA 253, reversed on other grounds 2013 SCC 18.