In R v Lavallee and R v Mohan, the Supreme Court of Canada established a test for the admissibility of expert evidence which is somewhat different from that used in other common law jurisdictions. Over the course of several more recent decisions, the court has expressed an increasingly sceptical attitude towards expert evidence of human behaviour. Collectively, these cases have left the state of Canadian law unclear. Canadian commentators also disagree about how best to navigate a path between the Scylla of uncritical reliance on expert evidence and the Charybdis of leaving discriminatory legal reasoning undisturbed. This article describes two proposals for reforming the Canadian approach to expert evidence and suggests that only one has the potential to move expert evidence jurisprudence beyond its current impasse.