Some quick thoughts. I'll respond in more detail tonight.
A second question, about (A1) is about who's giving the warning. I assume you mean an intermediary. How do we handle the situation where the intermediary wouldn't have given a proper warning, REGARDLESS OF WHETHER Lilly's warning was adequate. (Hint - go look at Hollis. Does that complication create common issue problems?)
A fourth answer - there are problems applying a traditional but-for analysis to this sort of case so the conclusion that material contribution - whatever it means - isn't applicable doesn't tell us anything more than that but-for is applicable. But is that the traditional but-for. Read Resurfice carefully. It doesn't necessarily stand for the proposition that plain vanilla but-for is the only flavour of but-for.
----- Original Message ----
From: Jonathan D. Tweedale
Sent: Wednesday, August 8, 2007 4:31:25 PM
Subject: RE: Judges, Buses and A Butterfly in Tokyo
David,
Your thoughts regarding Heward v. Eli Lilly are very interesting. Respectfully, I have reservations about your insightful analysis. There may be an obvious error in my own analysis, of course.
I take your argument to be the following:
The certification judge’s decision to certify the class was premised upon a finding that the requisite causal connection was disclosed in the case of each member of the class. The certification judge’s basis for holding that a sufficient causal connection was disclosed was that without the Defendant’s negligent conduct, the Defendant would not have derived the profits it obtained (the “Certification Causal Finding”). The leave judge held that the Certification Causal Finding rested upon either the assumption that:
(A1) The class members would not have agreed to take Zyprexa if properly warned of the risks associated with the drug; or
(A2) Zyprexa would not have been approved for sale if Health Canada was properly warned of the risks associated with the drug.
The leave judge then noted (at para. 29) that absent assumptions (A1) or (A2), “the only way to determine the amount for which the defendants could be ordered to account … is to investigate whether each member of the class would not have taken Zyprexa if properly warned” – which is “the antithesis of a common issue”.
You pose the question whether the leave judge erred in failing to consider whether the appropriate test for causation in the case is the Resurfice contribution to risk test. To determine the answer to that question, you pose the further question whether it is impossible for the plaintiffs to establish that but for the defendant’s wrongful conduct, Zyprexa would not have been released to the public. You note that at minimum it would be difficult to prove such a proposition about what Health Canada would or wouldn’t have done (the difficult “decision causation” issue). Assuming, then, that material contribution is the test, you note that not only does it appear that the plaintiffs established causation (under that test) in Heward, but it appears that in ANY mass tort type of case causation would be established under the material contribution test.
As I understand it, your argument hinges upon the key premise that the appropriate test in Heward is the Resurfice material contribution test. (Otherwise – I take it you agree – the leave judge was correct in identifying the unwarranted assumptions made by the certification judge in support of the Certification Causal Finding.)
In my view, there is good reason to doubt that the material contribution test is the appropriate test in this case:
No doubt if the only way that the plaintiffs could have proved the requisite causal connection was by establishing (A2) then it is very likely that the material contribution test is applicable (for the reasons that you give near the end of your piece). However, let’s not forget about (A1). You took no issue with the leave judge’s statement that a plaintiff could prove the requisite causal connection by proving (A1) – each plaintiff on an individual basis, of course. It does not seem “impossible” to prove (A1) due to factors outside the plaintiff’s control (such as current limits of scientific knowledge). So, (A1) is not amenable to a material contribution analysis. It follows that it would be improper to attempt to resolve (A1) by merely proving that the failure to warn of the risks materially contributed to the class members agreeing to take Zyprexa.
It appears that while (A2) could be described as a suitable candidate for the material contribution test, (A1) cannot.
So, the question is: where a causal connection may be proven two ways, but one of the two ways is “impossible” due to factors outside of the plaintiff’s control – do we apply the material contribution test? I think not. As long as there is a possible route to proof of causation – in this case, (A1) – which is not “impossible” in a manner such as to merit application of the material contribution test, then the normal but-for test should apply. To see this, we need only refer back to the language of Resurfice at paras 24 and 25, where the SCC stated that it was a requirement for the applicability of the material contribution test that it is impossible for the plaintiff to prove that the negligence caused the injury using the but-for test. As long as we have a way to prove that the negligence that caused the injury that is not impossible to prove under the but-for test, the material contribution test would not apply.
It follows that the appropriate test for causation in Heward is the but-for test, not the material contribution test – and the leave judge was correct to grant leave on issues concerning causation.
A further question is whether it makes any difference to the causation analysis that the issue is posed at the certification stage.
I do not see how it could make a difference here. It would be contrary to reason that in a case where an individual plaintiff would be required to prove but-for causation, a group of plaintiffs banding together as a class would only be required to meet the softer “material contribution” threshold.
