A 95 year-old widow was robbed and left hog tied in her room with a ligature around her neck. Over a period of 48 hours she suffocated to death. During an undercover investigation, a suspect, Nette, admitted to an undercover officer that he had robbed and killed the widow. Nette was arrested and charged with first-degree murder under s.231(5) of the Criminal Code. The Crown achieved a conviction at the lower court (upheld by the British Columbia Court of Appeal) using the Smithers test, which the defence said was too low of a threshold.
- What is the proper threshold of causation required for second-degree murder?
Arbour, writing for the majority, states that the Smithers test applies for all forms of homicide and the additional test used in R v Harbottle applies in the case of first degree murder, rather than to all forms of homicide. On the topic of jury instruction, she held that it is better to state the test positively as a "significant" cause rather than negatively and that this is really the same concept as de minimus, which they take to mean a cause that is "not insignificant". They state that this test is still met by the appellant's actions and the jury would have decided the same thing with the new charge.
L'Heureux-Dubé, concurring, has a serious problem with the changing of the phrasing in the jury instruction from "not insignificant" to "significant". They say that this new test creates a much higher threshold of causation than the Smithers test, and that there is more than a semantic difference between "significant", "not insignificant", or "more than trivial".
When addressing a jury, the standard of causation for second-degree murder should be positively stated in that the actions of the accused must have been a "significant contributing cause" of death. However, the Smithers causation standard still applies to all forms of homicide.