Purpose of s 53.1 Explained

Bauer Hockey Ltd v Sport Maska Inc (CCM Hockey) 2021 FCA 166 Locke JA: de Montigny, Rivoalen JJA affg 2020 FC 624 Grammond J

2,214,748 / FC Prosecution History

In this decision, the FCA affirmed Grammond J’s holding that the claims at issue were obvious, essentially on the basis of the deferential standard of review applicable to conclusions of mixed fact and law [16]. The most important aspect of the decision are Locke JA’s obiter comments on the interpretation of s 53.1 regarding the use of prosecution history. He provide a succinct analysis of purpose of the provision which will no doubt inform its interpretation in future cases.

In the decision under appeal, Grammond J held that no valid claims of Bauer’s 748 patent were infringed. The invention at issue relates to hockey skates. The uppers were traditionally made with two “quarters” that were sewn together. The 748 patent essentially claimed a one-piece quarter. It may seem obvious that the two pieces should be joined together, and this is exactly what Grammond J held [FC 154-58], [FC 184]. Grammond J addressed the “Beloit question,” [FC 146], referring to Beloit (1986), 8 CPR (3d) 289 (FCA) 295, to the effect of “If it was so easy, why was it not done before?” His answer was that one-piece quarters were expensive when implemented with directional material, such as leather, and became commercially feasible with the development of non-directional skate materials [FC 162–64]. The FCA affirmed, essentially on the basis of the deferential standard of review applicable to conclusions of mixed fact and law [16].

Locke JA noted that Grammond J began his discussion of obviousness by referring to the Sanofi 2008 SCC 61 [67] framework. Locke JA remarked (citations omitted):

[6] The Federal Court further noted, and I agree, that this framework should not be applied in a rigid manner. The only mandatory considerations are those laid out in section 28.3 of the Patent Act, which is concerned with obviousness to a person skilled in the art or science to which the invention pertains, having regard to “information disclosed before the [relevant date] in such a manner that the information became available to the public in Canada or elsewhere.”

The Sanofi framework has proven to be contentious and even unhelpful in some cases, particularly at the step of either identifying the inventive concept or construing the claims—see here, here and here—and this remark by Locke JA might be seen as the FCA starting to step back from reliance on the Sanofi framework. But I wouldn’t read too much into this statement, as the Sanofi framework has never been considered mandatory, and this was a case in which identification of the inventive concept was easy, so that whatever guidance might be provided by that framework is less significant.

On another point of passing interest, Bauer made the novel argument that “where inferences from findings of fact involve an evaluation of numerous factors (as with obviousness), intervention by an appellate court may be more appropriate in cases where the fact-finding judge is inexperienced” [17], relying on Actavis v ICOS [2019] UKSC 15 [78]–[79], where the UKSC stated (my emphasis) “Where inferences from findings of primary fact involve an evaluation of numerous factors, the appropriateness of an intervention by an appellate court will depend on variables including the nature of the evaluation, the standing and experience of the fact-finding judge or tribunal, and the extent to which the judge or tribunal had to assess oral evidence.” Bauer argued that less deference should be given to Grammond J’s inferences of fact because he had only previously been involved in one other patent infringement case [17].

Locke JA rejected this argument, noting that “To set the degree of appellate intervention based on the experience of the judge at first instance would require an appeal court to consider the judge’s background on every appeal. This would result in standards of review in a spectrum, rather than the two standards defined in Housen” [18]. Locke JA also noted that Actavis could be distinguished, as “Bauer proposes reduced deference in the case of a judge with limited experience, whereas Actavis was suggesting increased deference in the case of a judge with considerable experience” [19, original emphasis]. That is true, but it seems to me that giving relatively more deference to an experienced judge necessarily requires giving relatively less deference to an inexperienced judge; while Bauer’s argument is remarkable, but it does have a basis in Actavis. Accordingly, Locke JA did not rely on that distinction, but instead indicated that he would not be inclined to follow Actavis on this point: “In any case, I see no reason to modify the tools the Court already has to address decisions under appeal.” He noted that functionally, “[t]he advantages the judge at first instance has regarding assessment of factually-suffused issues, which prompted the Supreme Court in Housen to adopt a deferential standard, apply even to inexperienced judges” [19]. This strikes me as an entirely compelling point. I’d also note that following the Actavis line of reasoning leads to a morass. Not all appellate judges are experienced in patent litigation, so we might have four standards based whether the appellate judges are or are not experienced and whether the trial judge is or is not experienced.

Finally, the most important aspects of the decision relate to the interpretation of s 53.1 regarding the use of prosecution history. Grammond J had held that

[65] When an issue of claims construction arises, the patentee is always making representations to the Court as to the proper construction of the claims and the defendant is always attempting to rebut those representations. Therefore, in my view, as long as the issue is one of claims construction, section 53.1 applies and the prosecution history is admissible. In other words, there is no need to identify a particular representation and rebuttal every time a reference is made to the prosecution history. It is simply integrated in the interpretive process.

As Locke JA noted, “[t]his interpretation seems to open the door to unrestricted reference to the prosecution history to assist with claim construction,” and as such it is at odds with the text of the provision:

Subsection 53.1(1) is a detailed provision that contemplates admitting into evidence certain portions of a patent’s prosecution history for a certain purpose: “to rebut any representation made by the patentee in the action or proceeding as to the construction of a claim in the patent.” If this provision had been intended simply to brush aside the general prohibition against reliance on a patent’s prosecution history for the purposes of claim construction, it could have been much shorter.

This is not simply a textual point. The purpose of of s 53.1 “is to provide a tool to use against patentees who take one position concerning the meaning of a claim during prosecution of a patent application and another during litigation on the resulting patent. Accordingly, the concern relates to inconsistent statements” [37] (my emphasis). This limited purpose makes sense. On the one hand, it is particularly objectionable when the patentee wants to have it both ways, advancing a narrow interpretation during prosecution to obtain the patent, and then advancing a broad interpretation during litigation to expand the scope of its monopoly. On the other hand, permitting the general use of prosecution history, raises problems relating to “the public notice function of patent claims and the potential for inappropriately complicating litigation” [39], that were identified by the SCC in Free World 2000 SCC 66 [66] as reasons for refusing to allow the use of prosecution history for claim construction. Even though the legislature has now permitted some use of prosecution history, those concerns remain valid. The need to balance these considerations provides a purposive rationale for the limited nature of the exception, and that purpose is consistent with a straightforward reading of the text. While the legislature might reasonably have balanced these considerations differently, “it is not the role of the courts to participate in such a debate. Rather, we interpret and apply the legislation as written” [39].

Locke JA also noted that Grammond J’s interpretation seems to be inconsistent with the prior FC caselaw on s 53.1, specifically referring to Canmar 2019 FC 1233 affd 2021 FCA 7 (see here) and Lilly v Apotex 2020 FC 814. The interesting point here is that Locke JA did not refer to the decision of Crampton CJ in Allergan v Sandoz 2020 FC 1189, which, as discussed here, identified what appears to be a truck size loophole that is based squarely in the text of the provision. Crampton CJ’s decision certainly supports the point being made by Locke JA, namely that the exception is a limited one, and the text of the Act must prevail when it is clear. So the fact that Locke JA chose not to cite Allergan v Sandoz, while citing other FC caselaw, namely Lilly v Apotex, suggests to me that he did not want to be seen as approving Crampton CJ’s interpretation. I don’t take this as suggesting that Locke JA necessarily disagrees with Crampton CJ’s interpretation, but only that, given the magnitude of the loophole that Crampton CJ identified, the issue is best addressed by the FCA after the point is directly raised and argued, rather than by approving the decision in obiter comments.