Modernizing Circumstances: Revisiting Circumstantial Evidence in R v Villaroman

This blog also appears on the website:

My past two blog posts have a thematic connection and this post is no exception. I have modernity on the mind and so, apparently, do the courts. You may recall that theme in my discussion of the DLW decision (2016 SCC 22 (CanLII)) in which the Supreme Court of Canada, in the name of the “modern,” or the “modern approach” to be exact, entered into the time-honoured process of statutory interpretation only to come to the decision that the 2016 concept of bestiality under s 160 of the Criminal Code was no different than the common law concept of bestiality as subsumed into our codified criminal law in 1892. Justice Abella, hoping for a more modern approach, disagreed. Then, in my last blog post, I discussed the “smart” use of technological evidence to weave a persuasive narrative at trial. In the Didechko case (2016 ABQB 376 (CanLII)), the Crown relied, to great effect, on evidence emanating from the technological traces left by the accused to construct a case based on circumstantial evidence. Didechko serves as an exemplar of a thoroughly modern approach to another centuries-old process. Now, in this post, I will make another case for the modern as the Supreme Court of Canada in Villaroman (2016 SCC 33 (CanLII)) clarifies a very old rule on circumstantial evidence, one predating our Criminal Code, found in the English 1838 Hodge’s Case (168 ER 1136).

First, a few words on circumstantial evidence. We are all probably aware, contextually, of the difference between circumstantial and direct evidence. The most popular explanation in jury instructions and the best understood example involves rain and goes as follows: Imagine we wake up in the morning and when we peek out of the window to look at the weather for the day (this example is obviously pre smart phones) we notice the road, the sidewalk, and the ground is wet and rain is falling from the sky. We accept, therefore, that it is raining, and if we have been in Calgary all summer, we might even say “it is raining again.” A similar example was used in the Villaroman charge to the jury (at para 23). This is an example of direct evidence which, according to Watt’s Manual of Evidence 2016, page 49 at para 8.0 “is evidence which, if believed, resolves a matter in issue…the only inference involved in direct evidence is that the testimony is true.”

Circumstantial evidence is trickier and involves a more complex thought process. It differs from direct evidence as its probative value is found in the inferences to be drawn from the facts. Returning to our example, if we look out of our window and we see the road is wet but the sky is clear, we cannot directly aver to what the weather was like before we woke. We can, however, draw a “rational” or “reasonable” inference from the state of wetness and say “it was raining sometime before” but we did not observe that happen. We are not “direct” witnesses to this assumed event. In fact, we could be very wrong about our inference. For instance, if the road is wet but the sidewalk and ground is not, then we cannot safely assume it rained. A more “rational” or “reasonable” explanation may be that the City of Calgary street cleaners came by and washed the road. According to Watt’s Manual of Evidence 2016, page 50 at para 9.01, “it is critical to distinguish between inference and speculation.” An inference is “logical” (R v DD, [2000] 2 SCR 275, 2000 SCC 43 (CanLII) at para 18), “justifiable” (R v Charemski, [1998] 1 SCR 679, 1998 CanLII 819 (SCC) at para 33), “common sense” (Justice Moldaver in R v Walle, [2012] 2 SCR 438, 2012 SCC 41 (CanLII) at para 63), “rational” (R v Griffin, [2009] 2 SCR 42, 2009 SCC 28 (CanLII) at para 34) or, as preferred by Justice Cromwell writing for the Villaroman court, “reasonable” (at para 30). Conversely, speculation can lead to erroneous inferences. Speculation is tenuous as opposed to probative. Mere speculation strikes at the heart of the criminal justice system as it can ultimately lead to miscarriages of justice. It can cause the trier of fact to make an improper “leap” unsupported by the evidence.

To be cognizant of these improper “leaps” as a trier of fact is vitally important. As seen in Didechko, circumstantial evidence may be the only evidence of guilt or innocence. It is therefore essential, as a defence lawyer, to be able to argue persuasively that the circumstantial evidence does not amount to proof beyond a reasonable doubt as it is not reasonably sufficient to infer guilt. It is this argument, that the circumstantial evidence is “equally consistent with the innocence as with the guilt of the accused” (Fraser et al. v The King, [1936] SCR 296, 1936 CanLII 25 (SCC) at page 301), which is at issue in Villaroman but, as we will see, with a modern twist.

Mr. Villaroman was charged with various pornography related offences as a result of images found on his laptop computer, including a charge of possession of child pornography pursuant to s. 163.1(4) of the Criminal Code. As with most other possession offences, the possession element of the offence is where the circumstantial evidence was key to the prosecution’s case. The elements of possession are a curious mixture of statutory requirements and judicial interpretation, requiring proof of knowledge, consent, and control. Although section 4(3) of the Criminal Code clearly identifies knowledge and consent as elements of possession, the additional element of control is not found in the section. Rather, control is a judge-made requirement based on case authorities.

Thus in the Villaroman scenario, the prosecutor would have to prove Mr. Villaroman was aware of the child pornography on his computer, that he consented to the pornography being there, and that he had a measure of control over those images. The mere fact the images are found on his computer is not enough evidence of those essential elements. The Crown would need to figuratively, if not literally, place Mr. Villaroman’s fingers on the computer keys, at the time the prohibited images were knowingly captured by his computer, in order to prove possession. To do so, the Crown must rely on circumstantial evidence. In response, the defence must persuade the trier of fact that there are other reasonable or rational inferences which do not lead to guilt. As an aside, in Villaroman, Justice Cromwell equated “reasonable” with “rational” but, as mentioned earlier in this post, favoured the descriptor “reasonable” as the correct legal nomenclature (at paras 32 to 34).

The twist in Villaroman involves the source of those reasonable inferences or alternatives which lead to innocence. Traditionally, case authorities required that the inferences arise from the facts. In other words, there must be an evidential foundation for the defence’s position. However, by 2009 in the Khela decision ([2009] 1 SCR 104, 2009 SCC 4 (CanLII) at para 58), the Court found such a requirement effectively reverses the burden of proof by necessitating the defence “prove” facts in support of inferring innocence. Justice Cromwell in Villaroman makes it perfectly clear that this modern take does not invite speculation as long as it is within the range of reasonable inferences (at paras 35 to 38).  He gives two examples: one old and one new. In the 1936 case of Martin v. Osborne, [1936] HCA 23; 55 CLR 367, the High Court of Australia considered the admissibility of similar fact evidence as circumstantial evidence that the respondent, who was driving a commercial vehicle, was transporting people for pay contrary to legislation. In allowing the appeal against acquittal, Justice Dixon noted at page 375 (see para 40 of Villaroman) that “in the inculpation of the accused person the evidentiary circumstances must bear no other reasonable explanation” and further found at page 378 that the innocent inference was simply “too improbable.”

In the newer example from 2014, Justice Cromwell cited the Alberta Court of Appeal decision in Dipnarine (2014 ABCA 328 (CanLII), 584 AR 138) in which the court explained that circumstantial evidence need not “totally exclude other conceivable inferences” (at para 22) and that “alternative inferences must be reasonable and rational, not just possible” (at para 24). However, as the court further explains, “the circumstantial evidence analysis” (at para 25) is not a separate venture but is, in essence, the application of proof beyond a reasonable doubt. Ultimately, the trier of fact must “decide if any proposed alternative way of looking at the case is reasonable enough to raise a doubt” (at para 22). These reasonable alternate inferences can arise from either the presence of evidence or an absence of evidence. For instance, taking possession as an example, if there is no evidence of one of the necessary elements of knowledge, consent or control, the Crown has not proven the case and the accused person must be acquitted. This re-affirmation of the power of none is a step in the modern direction.

So what of Mr. Villaroman? At trial, the trial judge convicted Mr. Villaroman on the basis of the circumstantial evidence while the Alberta Court of Appeal set aside the conviction and entered an acquittal for the very same reason. The Supreme Court of Canada found the trial judge’s analysis was reasonable while the Alberta Court of Appeal’s position relied too heavily on “hypothetical alternative theories” (at para 67) which were “purely speculative” (at para 70). In other words, the appellate court “retried the case” (at para 69) by making that impermissible “leap” from the “reasonable” to the “improbable.”

The final nod to modernity in Villaroman is Justice Cromwell’s consideration of the form of the jury instruction on circumstantial evidence (at paras 17 to 24). In this discussion, Cromwell J sits firmly in today as he quotes approvingly from a passage written by Charron JA, as she then was, writing for the Ontario Court of Appeal in the Tombran decision (2000 CanLII 2688). There, in paragraph 29, she rejected the traditional “formulaic approach” to jury instructions in favour of “the modern approach to the problem of circumstantial evidence” which discusses all of the evidence, including circumstantial, within “the general principles of reasonable doubt.”

In modern terms this case suggests the jury need not be instructed in a finely constructed manner. Indeed, the Court, in a very modern turn, reiterates a theme they have been pursuing for years – that there are no “magic incantations” (WDS, [1994] 3 SCR 521, 1994 CanLII 76 (SCC) at page 533) or “foolish wand-waving or silly incantations” (a shout out to Professor Snape in Harry Potter) needed to “appeal-proof” jury instructions. The charge to the jury must remain nimble, tailored to each individual case and created by the judicial gatekeeper who is expected to weave a legal narrative for the trier of fact. Should there be no jury, then it is incumbent on the judge to be mindful in their approach to the evidence. To be modern, therefore, requires mental acuity and agility not pondering recitations of old rules but fresh iterations, perhaps on an old theme, but yet thoroughly modern.