What does that patent say, anyway? Appellate review of patent construction

In two recent judgments,  Cobalt v. Bayer and  ABB v Hyundai,  the Federal Court of Appeal has explicitly considered the appropriate standard of review for issues of patent construction, suggesting that this is an area ripe for review by the Supreme Court of Canada  The debate centres on whether patent construction should be reviewed on the correctness standard or the standard of palpable and overriding error, given the heavily factual nature of the trial Court’s inquiry into knowledge of the skilled reader of the patent.

The Supreme Court of Canada in Whirlpool has said patent construction is “a matter of law for the judge”. On appeal, issues of patent construction are reviewed on the “correctness” standard, a much less deferential approach than is afforded to factual findings from trial (which are reviewed for palpable and overriding error).

In Cobalt, the FCA held that the fact that a patent is deemed to be a “law” under the Interpretation Act does not necessarily mandate appellate review of construction on the correctness standard, noting that to accept that it does “is to permit form … to dictate substance.” Instead, the FCA proposes a distinction between “letters patent” and patent specifications. The term “letters patent” is to be limited to the certificate issued by the Canadian Intellectual Property Office on approval of a patent application. Only that certificate is to reviewed on the correctness standard, while the specification – the substance of what the patentee is telling the public – is to be reviewed for palpable and overriding error.

In both cases, the FCA actually applied the correctness standard “pending any future consideration by the Supreme Court of Canada” . The Court of Appeal also queried “whether the time has come to reconsider the view that appellate courts are to review patent construction on the basis of correctness.” Given this, it is clear that there will be no change until the Supreme Court has had the chance to weigh in.

In Cobalt, the FCA referred to the USSC’s decision in Teva v Sandoz as signaling a “growing acceptance” that deference should be accorded to the interpretation of patents reached by those who have seen the experts and evaluated them”, suggesting, at least, that the Court of Appeal shares this view and would prefer to defer to the triers of fact on matters of patent construction. Indeed, the FCA wrote:

How are appellate judges supposed to cleave off those aspects of claim construction that flow from the trial judge’s appreciation of expert evidence from the words of the claim per se? Can appellate judges really second-guess the trial judge, who, often over many days, has been educated in the relevant art and has seen and evaluated the experts? Who are the appellate judges to review on the basis of correctness, stepping into the shoes of the trial judge and imposing their own views of the matter?