Relevance in Context: SCC decides R v Schneider

Trial judges play many roles, but one of their main tasks is determining what evidence is admissible. As a rule, all relevant evidence is admissible. But how do trial judges determine relevance, and what can they not take into account? This was the main question in R v Schneider, 2022 SCC 34 [Schneider], which the Supreme Court of Canada (SCC) decided this year.

Facts and History

In September of 2016, a young woman named Natsumi Kogawa was reported missing. Two weeks later, a news report showed a picture of Ms Kogawa with an unidentified male, later identified as William Schneider (the accused) by his brother, Warren.

Warren reached out to William. They later met up, and William told his brother that he had gone on three dates with Ms Kogawa. On their third date, William said they both took “medication” and told his brother “It’s true” (Schneider, para 11). Warren was not allowed to testify what he thought the accused meant by this statement.

The next day, the accused wanted to take heroin and die by suicide. He told his brother, and they went together to a park. Before taking heroin, the accused told his brother where Ms Kogawa’s body was, and asked him to give the police that information after his death. When the accused did not die, he asked for his brother’s cell phone and called his wife. Warren overheard some of the conversation, including reference to Ms Kogawa near the beginning of the call and, sometime later, “I did it” or “I killed her”—or something “along those lines” (Schneider, paras 16-17).

After overhearing the call, the accused’s brother gave an anonymous tip to the police and told them where they could find Ms Kogawa’s body. The police followed the tip and uncovered her body in a suitcase in west Vancouver.

After some investigation, Mr William Schneider was charged with second-degree murder. His brother testified at trial. The foregoing information was admitted by the trial judge and Mr Schneider was convicted.

Mr Schneider appealed to the British Columbia Court of Appeal. He had two grounds of appeal: that the trial judge erred in admitting his brother’s evidence about the phone call he overheard, and that the trial judge erred in answering a question from the jury.

Only the first ground of appeal was successful. A majority of the Court of Appeal allowed the appeal and overturned the conviction. For the majority, only the context of the phone call itself could be considered in determining whether or not the phone call evidence was relevant. Because the brother could not hear the accused’s wife’s side of the phone call, and some time separated the “I did it” statement from the reference to Ms Kogawa, the majority considered the evidence too speculative to admit. Justice DeWitt-Van Oosten, in dissent, considered the context of all the evidence, including the previous conversations between Mr Schneider and his brother, and held that the phone call evidence was relevant.

The Crown appealed to the SCC. The SCC allowed the appeal and restored the conviction.

Evidence Law, Hearsay, and Relevance

Evidence law is a law of rules and exceptions. All relevant evidence is to be admitted, except when an exclusionary rule applies. Hearsay evidence is presumptively inadmissible, except when it falls under an established exception. Even then, trial judges retain discretion to exclude such evidence (using a complex balancing beyond the scope of this article).

Hearsay

Hearsay is any statement made out-of-court by someone other than the person testifying about it. Due to concerns related to this kind of evidence’s reliability, hearsay is inadmissible if it is being led to prove the truth of its own contents. One exception to this rule is party admissions, which are statements made by a party to a case, if led against them. An accused is a party to his own case. When the Crown leads hearsay evidence about something the accused has themselves said, it is not excluded by the hearsay rule. Courts accept that parties should be held to their own words where they likely would not have said something untrue.

At issue here is a statement made by the accused (“I killed her”), by someone other than the accused (his brother), to prove its own contents (that the accused did, in fact, kill her). The accused is a party to the case. The evidence is being led by the Crown. The statement, therefore, is admissible as a party admission. Right?

The parties agreed that, if the statement was to be admissible, it was admissible as a party admission. But the issue raised by the accused is even more basic than that—the accused suggested that the evidence was not relevant. Irrelevant evidence is inadmissible.

Relevance

When determining admissibility, relevance is a binary question: Evidence either is, or is not, relevant. How relevant that evidence is, in terms of how strongly it should be weighed, does not matter when determining its admissibility. For admissibility, evidence is relevant if it is capable of proving a fact in issue (see Morris v The Queen, [1983] 2 SCR 190).

SCC Holds that all Context can be used to Determine Relevance

All nine justices of the SCC agreed in principle that all available context can be considered in determining whether a specific piece of evidence is relevant. Justice Malcolm Rowe penned a comprehensive majority opinion outlining the SCC’s jurisprudence on admissibility, relevance, hearsay, and party admissions. All nine justices agreed that, under the governing case of R v Ferris, 1994 ABCA 20 (CanLII), relevance could only be determined if there was enough context. The majority determined that, considering all the evidence including the conversations between Mr Schneider and his brother, the evidence about the phone call was “capable of non-speculative meaning” (Schneider, para 2).

However, not all the justices agreed in the result. Justices Andromache Karakatsanis and Russell Brown co-authored a concise dissent. In their view, it was not sufficiently clear that what the brother overheard—”I did it” or “I killed her”—was relevant evidence.

You might be wondering how a statement like “I killed her”—especially when made by a person accused of murder—could possibly be not relevant. The answer is surprisingly simple: the statement in question (“I killed her” or “I did it”) lacks context. 

It is not clear whether Mr Schneider was referring to Ms Kogawa at all. If he was, it is not clear whether he was admitting guilt or responding to a question (For example: “What did he say?” “That I killed her”, or “Did you take the garbage out?” “Yes, I did it”). This lack of context made it impossible to tell whether this evidence was relevant. The dissent noted that several minutes had passed between the mention of Ms Kogawa and the alleged statement. The conversation could have been about something else. And the fact that the accused’s brother could not hear the other side of the conversation meant that he could not say whether the comment was related to Ms Kogawa or not. 

In the dissent’s view, the majority relied on contextual information unrelated to the phone call itself. They criticized the majority decision for distinguishing between micro and macro context—what others might have called immediate or prior context—in a way that the Court of Appeal had not done. While the previous interactions might have informed the accused’s brother’s perception of that phone call, the dissent held that there was not enough context within his experience of the phone call itself. The accused’s brother was 10 feet away, actively trying not to listen, and could not remember the exact words used, or the words before or after the statement. It was impossible to determine whether the statement was properly in the context of Ms Kogawa’s death and disappearance or not.

Micro or Macro Context Narrows Relevance Too Far

In determining relevance, itself a binary question, the inquiry should not be complicated too much. Evidence law is an interconnected web of rules and exceptions, and further complicating the relevance analysis with different degrees of context, in my view, risks lowering the threshold bar too low.

Relevance is a threshold question, and the bar is already quite low. If evidence relates to a fact in issue, it is relevant. Not all relevant evidence will end up carrying weight — but weight assessed by the trier of fact, and that weighing is not for the trial judge. But where it is uncertain that evidence is relevant, and it is evidence that runs a real risk of being mis-weighed if its relevance is left to the jury, shouldn’t a system that assumes innocence until guilt is proven beyond a reasonable doubt err on the side of exclusion?

A criminal accused should not be immune from their own words. If they admit to committing a murder, and someone compellable overhears it, they should be held to their own (likely true) admission. However, it is not for the court to artificially construct context where the circumstances simply do not support it. It is a stretch, as the dissent said, to say that a person who is not involved in a conversation, especially not privy to both sides of it, suggests that a specific string of words, divorced from what came before and after,  imparts a particular meaning.

In this case, it was for the Crown to prove relevance. The bar might be low, but it cannot be so low that the mere possibility that the statement means what they assert it means can prove its relevance. That a conversation with a specific topic came up does not lead inexorably to the conclusion that that same topic was being discussed minutes later. In the absence of any context within the conversation itself, a court goes too far to tell a jury they can rely on a dubious statement as proof of guilt.