August 15, 2017
The recent acquittal by the Ontario Superior Court of three male police officers who had been charged with sexually assaulting a female colleague has, yet again, shone a bright light on the ongoing challenges posed by some of the bedrock principles of criminal law: the presumption of innocence and the heavy evidentiary burden of proving guilt beyond a reasonable doubt.
While many will find the outcome of this case to be wholly unsatisfactory, and will see it as yet another example of the criminal justice system’s failure to given an appropriate level of credence to the allegations of victims, to her credit, Judge Anne Molloy devoted a significant part of her written ruling to explaining these basic principles. As is not uncommon with cases of this nature, the outcome was, to a great extent, a matter of credibility.
Workplace investigators also have the sometimes difficult task of determining what evidence is reliable and credible. Workplace investigations are not criminal matters, so the standard of proof is the civil standard: proof on a balance of probabilities. In an investigation where people’s credibility is a factor, my clients often ask me how I decide who to believe — Doesn’t it all comes down to how the parties and witnesses behave during the interviews, they sometimes ask?
Although I usually consider the demeanor of the parties and the witnesses when they give their evidence as well as the firmness of their professed recollections, I have to be mindful that in many cases (even if a significant amount of time has not passed since the alleged incidents) memories can be faulty, narrators are sometimes unreliable, and specifics are evanescent. I also consider the fact that none of the witnesses in a workplace investigation testifies under oath or is cross-examined. As a result, I have usually given very little weight to these somewhat more traditional considerations.
Instead, I place much greater emphasis on the clarity, consistency, and overall plausibility of every witness’ evidence when compared to the evidence of other witnesses. This does not mean that the most organized or eloquent witness is always the most persuasive. Witnesses are usually nervous, and many ramble or go off on tangents. And as many seasoned investigators know, these digressions can sometimes yield important pieces of information. In my view, clarity, consistency, and overall plausibility speaks more to what has the “ring of truth” to it: what is more probable, or more likely, to have happened. Most of us know when something sounds farfetched or simply does not make sense.
In contrast, as far as Judge Molloy was concerned, the appropriate verdict was “not guilty” – and not because she was sure that the accused had behaved like perfect gentlemen. Even if she had believed that they were probably guilty or likely guilty, that would not have been enough. In a criminal case, the benefit of the doubt goes to the accused, and if she was not satisfied of their guilt beyond a reasonable doubt, she had no choice but to acquit them. In a workplace investigation, where the standard of proof is on a balance of probabilities, the evidence does not have to be as strong. An investigator does not need to be certain about exactly what happened, only what probably happened or is more likely to have happened.
When the outcome of an investigation comes down to a question of credibility, I also carefully consider the ability of all witnesses to resist the influence of self-interest and self-justification when providing evidence and framing their answers. I take into account the consistency of witness’ evidence with the documentary evidence, when available, as well as whether the assertions of witnesses are consistent with, or corroborated by, other objective evidence. Finally, I consider what seems to be most likely in all the circumstances established by the reliable evidence. This is sometimes described as determining what version of events is more inherently plausible. In a criminal case, even in the face of a clear verdict, sometimes the most that can be expected is that much has been shown to be unclear or uncertain.