Canada (Attorney General) v. Benjamin Moore & Co.

Canada (Attorney General) v. Benjamin Moore & Co.
Court (s) Database
Federal Court of Appeal Decisions
Date
2023-07-26
Neutral citation
2023 FCA 168
File numbers
A-188-22
Decision Content
Date: 20230726
Docket: A-188-22
Citation: 2023 FCA 168
CORAM:
GAUTHIER J.A.
LOCKE J.A.
ROUSSEL J.A.
BETWEEN:
ATTORNEY GENERAL OF CANADA
Appellant
and
BENJAMIN MOORE & CO.
Respondent
and
INTELLECTUAL PROPERTY INSTITUTE OF CANADA, CANADIAN LIFE AND HEALTH INSURANCE ASSOCIATION INC. AND THE INSURANCE BUREAU OF CANADA
Interveners
Heard at Montréal, Quebec, on February 16, 2023.
Judgment delivered at Ottawa, Ontario, on July 26, 2023.
REASONS FOR JUDGMENT BY:
GAUTHIER J.A.
CONCURRED IN BY:
LOCKE J.A. ROUSSEL J.A.
Date: 20230726
Docket: A-188-22
Citation: 2023 FCA 168
CORAM:
GAUTHIER J.A.
LOCKE J.A.
ROUSSEL J.A.
BETWEEN:
ATTORNEY GENERAL OF CANADA
Appellant
And
BENJAMIN MOORE & CO.
Respondent
And
INTELLECTUAL PROPERTY INSTITUTE OF CANADA, CANADIAN LIFE AND HEALTH INSURANCE ASSOCIATION INC. AND THE INSURANCE BUREAU OF CANADA
Interveners
REASONS FOR JUDGMENT
GAUTHIER J.A.
[1] The Commissioner of Patents refused to grant patents to Benjamin Moore & Co. (BM) in respect of its applications for Canadian Patent No. 2,695,130 (the 130 application) and Canadian Patent No. 2,695,146 (the 146 application) both relating to colour selection systems. The Commissioner refused the applications on the ground that the claims therein were directed to non patentable subject matter and therefore non-compliant with section 2 of the Patent Act, R.S.C. 1985, c. P-4 (the Act). This conclusion was based solely on her determination that the essential elements of those claims, as she construed them, were those providing the solution to the practical problem disclosed in the two applications, which did not include the computer (controller) and other conventional associated components.
[2] BM appealed the Commissioner’s decisions to the Federal Court pursuant to section 41 of the Act. These appeals were consolidated and heard together. The Intellectual Property Institute of Canada (IPIC) was granted leave to intervene in the consolidated appeals. This is somewhat surprising, given that the Attorney General (AG) had already conceded that the Commissioner had used the wrong test to construe the claims and agreed that the decisions should be set aside and be remitted back to the Commissioner for reconsideration. The parties even discussed the form of a consent judgment that could be proposed to the Court.
[3] Besides generally siding with BM’s position in respect of the Commissioner’s approach to claim construction in this case, IPIC took the debate one step further by asking the Court to adopt a revised framework to be followed by the Commissioner in assessing the patentability of computer-implemented inventions once the claims have been purposively construed, and to instruct the Commissioner to adhere to it in determining the patentability of such inventions.
[4] By the time the appeals were heard by the Federal Court, all parties agreed that it would be inappropriate for the Court to direct the Commissioner to issue the patents or to determine the subject matter patentability of the applications at issue as originally requested by BM in its notice of appeal. The only remaining remedy sought in the notice of appeal was that the matter be remitted to the Commissioner for reconsideration. The disagreement between the participants (parties and interveners) lay solely in whether the Court should issue binding instructions, and if so, what those instructions should be.
[5] The Federal Court granted the appeals (2022 FC 923, the FC Decision) on the basis that, as conceded by the AG, the Commissioner had failed to apply the proper test to determine the essential elements of the claims in these applications, following instead the approach described at section 13.05.01 (now section 12.02.01) (the version reproduced at para. 10 of the FC Decision is substantively the same as that used by the Commissioner in the decisions) of the Canadian Intellectual Property Office (CIPO)’s Manual of Patent Office Practice (MOPOP), before it was amended in response to the then-recent decision of the Federal Court in Choueifaty v. Canada (Attorney General), 2020 FC 837 (Choueifaty). In Choueifaty, the Federal Court set aside a decision of the Commissioner on the basis that the approach used to construe the claims in the patent application in issue, which was set out at section 13.05.02(c) of the MOPOP (now section 12.02.02(e)) was not in line with the principles of purposive construction set out in Free World Trust v. Électro Santé Inc., 2000 SCC 66 (Free World Trust) and Whirlpool Corp. v. Camco Inc., 2000 SCC 67 (Whirlpool).
[6] The Federal Court also noted that the Commissioner had erred by conducting a novelty analysis while determining the essential elements of the claims, contrary to Free World Trust and Whirlpool, which require claim construction to precede the novelty analysis (FC Decision at para. 36).
[7] It would seem that the reason BM was no longer asking the Federal Court to issue the patents is that this would have raised an issue as to the Court’s jurisdiction in the appeal before it. The Court agreed that it was not appropriate for it to make its own determination of the patentability of the subject matter of the applications, and that the matter should be remitted to the Commissioner for reconsideration. However, the Federal Court accepted IPIC’s invitation to include specific instructions in its judgment in the form of a binding test to be followed by the Commissioner in assessing the patentability of these computer-implemented inventions, and presumably others. BM and IPIC refer to this test as the “BM test”, which in their view, clarifies the law in respect of computer-implemented inventions in general (FC Decision at paras. 33, 53).
[8] The test set out by the Federal Court in its judgment is as follows:
3. In her assessment of the 130 and 146 Applications, the Commissioner of Patents is instructed to:
a. Purposively construe the claim;
b. Ask whether the construed claim as a whole consists of only a mere scientific principle or abstract theorem, or whether it comprises a practical application that employs a scientific principle or abstract theorem; and
c. If the construed claim comprises a practical application, assess the construed claim for the remaining patentability criteria: statutory categories and judicial exclusions, as well as novelty, obviousness, and utility.
[9] Before us, the AG appeals the FC Decision. This appeal focuses solely on the test stated at paragraph 3 of the Federal Court’s judgment. BM filed a motion to dismiss the appeal on the basis that the AG was attacking the reasons for judgment rather than the judgment. Justice Rennie dismissed the motion (2022 FCA 194), stating that paragraph 3 of the Federal Court’s judgment is a specific direction akin to a declaratory judgment. This is quite different from general references to reasons in a formal judgment, which do not form part of the judgment itself. Thereafter, IPIC, the Canadian Life and Health Insurance Association Inc. (CLHIA) and the Insurance Bureau of Canada (IBC) were granted leave to intervene.
[10] As agreed by all, this case is most unusual for a variety of reasons that will be discussed later on. On the one hand, the Commissioner does not want to be held in contempt of court under a test that, in her view, is ambiguous and does not deal with all the relevant case law of our Court not referred to by the Federal Court. On the other hand, BM, who was the appellant before the Federal Court, and who appeared to be more than willing to cooperate with IPIC to seek some instructions, now seeks an expedited determination of the present appeal based on the prejudice it claims to be suffering while a test it did not even request itself is being debated.
[11] Moreover, what is being asked of this Court is essentially to give an opinion on how to construe existing Canadian case law dealing generally with subject matter patentability in order to reduce it to what purports to be a “simpler” approach for the determination of the patentability of computer-implemented inventions. Not only is this a very complex issue that has been the subject of much debate before the highest courts in the United States and Australia, but our Canadian case law dealing with computer-implemented inventions, particularly in the context of applying judicial or statutory exclusions is not fully developed, consisting of only a few decisions. This is especially relevant, considering that most participants before us either criticized or attempted to distinguish particular decisions that are indeed relevant to the exercise, while simultaneously arguing that this Court was not being asked to depart from previous case law on the basis of the test of Miller v. Canada (Attorney General), 2002 FCA 370 at paragraph 10, or to make new law.
[12] Finally, I ought to note that despite the fact that IPIC argued that it is rare that our Court or the Federal Court will deal with issues relating to patentable subject matter in cases involving computer-implemented inventions, there is currently such a case before the Federal Court (T-657-22), in which the Commissioner considered the revised MOPOP following Choueifaty in a Practice Notice entitled “Patentable Subject-Matter under the Patent Act” (PN 2020-04), which represents CIPO’s current practice. This case is being held in abeyance until the present appeal is finalized, as the Federal Court, in that case, would otherwise have to consider the test set out by the Federal Court in the present case.
[13] As will be explained, the Federal Court erred for various reasons in including the test set out at paragraph 3 of its judgment. I further find that it would not only be premature, but quite unwise to attempt to settle issues that have yet to be properly considered by any court in Canada, and that the participants did not adequately address before this Court. I would thus allow the appeal, but only to delete paragraph 3 of the Federal Court’s judgment reproduced above. I would also add a direction that the Commissioner re-examine these applications on an expedited basis, in light of the most current version of the MOPOP with the benefit of these reasons.
I. Background [14] The Commissioner rendered her decisions on May 8, 2020, refusing both applications on the ground that, as mentioned, the claims were directed to non-patentable subject matter and therefore non-compliant with section 2 of the Act, as recommended by the patent appeal board. She described the two applications as follows in her two decisions: the 146 application is a “computer-implemented colour selection method” that focuses on the “the provision of appropriate colour combinations given a user’s selection of a threshold colour harmony or colour emotion value”. The 130 application is also a “computer-implemented colour selection method”, but it focuses on “providing the user with a combined colour score (e.g., colour harmony or colour emotion score) upon a user selection of at least three colours from a colour library” (Benjamin Moore & Co. (Re), 2020 CACP 16 at para 4) These two decisions were rendered before the decision of the Federal Court in Choueifaty was issued on August 21, 2020.
[15] On November 3, 2020, CIPO issued the updated Practice Notice PN2020-04 in response to Choueifaty. This Practice Notice not only provided an update on CIPO’s understanding as to how Free World Trust and Whirlpool were to be applied when determining the essential elements of claims, but it was also intended to ensure that the references to the “technological solution to a technical problem” found at various chapters of the MOPOP would no longer be applied.
[16] A few days later, on November 9, 2020, BM appealed the Commissioner’s decisions to the Federal Court. Considering the only remaining issue before us, the most relevant portions of the FC Decision are those relating to the test to be followed by CIPO. Although the Federal Court referred to IPIC’s submissions that such strict guidance was necessary because of CIPO’s alleged continued misapplication of the law in violation of Free World Trust, Whirlpool, Shell Oil Co. v. Commissioner of Patents, [1982] 2 S.C.R. 536, 142 D.L.R. (3d) 117 (Shell Oil), and now Choueifaty (FC Decision at para. 44), it never made any conclusion or finding in that respect. In addition, the Federal Court did not discuss the post-Choueifaty administrative case law cited by IPIC in support of that contention.
[17] The Federal Court made it clear that BM and IPIC were not asking it to choose its preferred interpretation of statutory provisions, but rather to direct CIPO “not to depart from the applicable jurisprudence” (FC Decision at para. 49). It also acknowledged that the appellant had only sought an order directing CIPO to re-examine the applications in accordance with the principles of Free World Trust, Whirlpool, and Shell Oil, and not to use the “Problem-Solution Approach” or the “Substance of the Invention Approach” (FC Decision at paras. 3 and 38). According to the Federal Court, the AG took no position on whether IPIC’s proposed framework accurately reflected the state of the law, simply submitting that the Court should decline to direct the Commissioner to adopt the appellant’s interpretation of the jurisprudence, just as it should decline to direct CIPO to apply the proposed framework (FC Decision at para. 45). At most, the AG agreed to refer the Commissioner to Choueifaty as guidance for reconsideration (FC Decision at para. 39).
[18] The most substantive explanation for adopting the test proposed by IPIC is found at paragraph 52 of the FC Decision. After noting that legal frameworks are questions of law within the purview of a reviewing court, the Federal Court simply stated that the test proposed by IPIC and endorsed by the appellant is in keeping with the Supreme Court’s teachings in Free World Trust and Shell Oil, and with our Court’s invitation in Amazon.com, Inc. v. Canada (Attorney General), 2011 FCA 328 (Amazon) at paragraph 68, to adapt “our understanding of the nature of the ‘physicality requirement’” for patentable subject matter as technology advances. The Federal Court also noted that IPIC’s test would ensure consistency in the law applied to patent applications by CIPO and the Courts, and in the treatment of computer-implemented inventions and all other types of inventions (FC Decision at para. 53).
[19] I ought to mention that, as noted by the AG, the FC Decision does not deal with or even refer to several cases the Commissioner expressly mentioned in her decisions, although she did not deem it necessary to apply them, considering her construction of the claims.
[20] I could discern no explanation either as to the usefulness of the test in light of the current practice of CIPO (the revised version of the MOPOP and PN2020-04) which, in its analysis, the Federal Court dismissed as not being relevant to the matter before it (FC Decision at para. 50). This presumably would include the relief to be granted. Moreover, in its conclusion, the Federal Court indicated that it was “directing CIPO on the proper procedure for claims construction and identifying patentable subject matter” (FC Decision at para. 54). However, the test set out at paragraph 3 of the judgment goes beyond these two elements of the patentability assessment by dictating the order in which the Commissioner should assess the other aspects of the patentability analysis (novelty, obviousness, and utility), mandated by the Act. In that respect, the Federal Court did not indicate why it could deviate from this Court’s statement in Amazon at paragraph 38 that once purposive construction is completed, the assessment of patentable subject matter, obviousness, novelty, utility, and excluded subject matter need not follow any particular order.
II. The issues and standards of review [21] The AG asks this Court to quash the test set out at paragraph 3 of the Federal Court’s judgment, or alternatively, to amend it to reflect that the claims must be directed to “more than a bare practical application”. The AG’s arguments before this Court are focused on the alleged incorrectness at law of the test in question as well as its ambiguity. In the AG’s submissions, if the test is upheld, Canada will become an international outlier in its approach to subject matter patentability, particularly in respect of computer-implemented inventions. In addition to generally supporting the AG’s position, the new interveners (the CLHIA and IBC) raise other arguments, including that Choueifaty was wrongly decided, and that the test adopted by the Federal Court would change the law on the patentability (or non patentability) of business methods in Canada, a matter they say is of vital importance to their membership.
[22] What started as a relatively simple matter, where the two parties to the statutory appeal had essentially agreed that the decisions should be remitted to CIPO for redetermination in accordance with Choueifaty because the Commissioner had not properly applied the principles of purposive construction to determine the essential elements of the claims, degenerated into what could be considered a reference on subject matter patentability under section 2 of the Act, and the statutory interpretation of subsection 27(8) of the Act in general.
[23] This is why, at the beginning of the hearing, the panel asked the parties to address how this Court could generally deal with the correctness of all the issues raised by this test in the context of this particular case, as it appeared essentially to be a reference seeking a pure declaratory judgment on issues that did not form the basis of the Commissioner’s decisions, given her determination of the essential elements of the claims (146 Decision at paras. 59-60; 130 Decision at paras. 60-61).
[24] In my view, the present appeal boils down to a single question:
Did the Federal Court err in setting out the test at paragraph 3 of its judgment?
[25] Though not determinative of the outcome of the present appeal, there are some issues, as will be explained under “General Observations”, with respect to the propriety of the remedy granted, and the basis on which the Federal Court issued what is akin to a declaratory judgment that ought to be discussed.
[26] Obviously, the appellate standards of review set out in Housen v. Nikolaisen, 2002 SCC 33 (Housen) apply in this appeal. The Federal Court viewed the question of the proper legal test or framework to be applied as a simple question of law reviewable on the standard of correctness. Considering that there is no discussion of the basis for including this test in its judgment, it is not clear on what basis the Federal Court felt it could exercise its discretion to set a test in its judgment on the specific statutory appeal before it.
[27] Assuming that the Federal Court had, in the context of this case, the discretionary power to issue specific instructions to the Commissioner as to how she should generally carry out her examination of patent applications involving computer-implemented inventions, this Court must nevertheless determine whether the Federal Court erred in exercising this discretion in the circumstances. This particular question is reviewable on the standard of a palpable and overriding error, unless the Federal Court made an extricable error of principle, which would be reviewable on the standard of correctness (Hospira Healthcare Corporation v. Kennedy Institute of Rheumatology, 2016 FCA 215 at para. 79 (Hospira)).
III. Analysis [28] Our Court in Amazon already provided guidance in respect of many of the issues raised in this appeal. In that case, it had the benefit of a 197-paragraph decision by the Commissioner, and a very detailed analysis by the Federal Court (2010 FC 1011). Having determined that the Federal Court had erred in adopting its own purposive construction of the claims, our Court clearly did not deem it appropriate to develop the type of test that the Federal Court adopted in this case. The guidance provided by this Court in its reasons in respect of the determination of the subject matter was based on the actual elements referred to and applied by the Commissioner in refusing Amazon’s application, which were also expressly addressed by the Federal Court. In my view, our Court simply could not go any further, considering the uniqueness and complexity of the facts in play and that the Commissioner had yet to construe the claims properly. Still, many statements in Amazon and other cases relied upon by BM and IPIC persuade me that the Federal Court erred in adopting the test it did.
[29] Given the lack of a detailed analysis in the FC Decision, the Federal Court’s statement that the current version of MOPOP (as amended by PN2020-04) was irrelevant, and the lack of consideration of all relevant case law, it appears to me that it did little more than “leapfrog” the test to our Court by including it in its judgment instead of its reasons. This exercise, which is quite different from dealing with an issue that was not necessary to determine the appeal before it in obiter, is inappropriate. It does not involve “judicial courage”, as IPIC argued before the Federal Court (Transcript of the Federal Court hearing of T-1340-20, Appeal Book, Vol. 1, Tab 4, p. 105).
[30] I will start my analysis with general observations regarding the propriety of the Federal Court’s adoption of a test, which was not included in BM’s notice of appeal, in a manner akin to a declaratory judgment, and without considering the applicable test for such declaratory relief. I will then consider the substance of the test included at paragraph 3 of the Federal Court’s judgment.
A. General Observations [31] As noted earlier, the remedies sought by BM before the Federal Court kept evolving. In its notice of appeal, BM sought an order setting aside the Commissioner’s decisions, requiring the Commissioner to allow the applications and grant the respective patents, or declaring that the applications disclose inventions pursuant to section 2 of the Act, and alternatively, an order directing the Commissioner to reconsider the applications. However, as mentioned, at the hearing, BM abandoned the first two remedies, leaving only the order for reconsideration. In the paragraph of its memorandum dealing with the orders sought, BM requested an order directing CIPO to use the test for purposive construction set out in Free World Trust and Whirlpool, and an order directing CIPO not to use the “Problem-Solution Approach” or the “Actual Invention Approach” (Appeal Book, Vol. 1, Tab 7, p. 229 at para. 93). At paragraph 87 of its memorandum, BM indicated that the instructions should be to follow not only Free World Trust and Whirlpool but also Shell Oil. Thus, even though Shell Oil was referred to, this again was not part of the relief sought. At the hearing, BM did “endorse” the test proposed by IPIC, in its submissions in chief, acknowledging that, in its view, it was correct at law.
[32] In concluding its oral submissions, the AG made it clear that the notice of appeal did not include a request for the adoption of an alternative framework, such as the one proposed by IPIC (Transcript of the Federal Court hearing of T-1340-20, Appeal Book, Vol. 1, Tab 4, p. 131). This clearly put BM on notice that this issue had to be addressed. In its reply, BM began by reminding the Court that IPIC was not the appellant and that its role as an intervener was only to assist the Court in understanding the issues and that despite this, most of the hearing was spent discussing IPIC’s arguments (Transcript of the Federal Court hearing of T-1340-20, Appeal Book, Vol. 1, Tab 4, p. 133). One can only agree with this statement, given the definition of “party” in the Federal Courts Rules, SOR/98-106. It is established that interveners cannot seek a remedy that was not sought by the parties themselves (Tsleil-Waututh Nation v. Canada (Attorney General), 2017 FCA 174 at paras. 54-55; Zak v. Canada (Attorney General), 2021 FCA 80 at para. 4).
[33] BM’s submissions in reply are difficult to follow insofar as the issue of the nature of the relief sought is concerned. What is clear is that at most, BM agreed that it was within the Court’s discretion to clarify the law and endorse IPIC’s framework, should it find that it accurately reflects the law. What BM did not do is seek an amendment of the remedies sought in its notice of appeal.
[34] Rule 337(c) dealing with the contents of a notice of appeal provides that a precise statement of the relief sought must be included. The same wording is used with respect to the contents of a notice of application under Rule 301(d). As mentioned in Mahjoub v. Canada (Citizenship and Immigration), 2017 FCA 157 at paragraph 51, originating documents are to be construed in order to gain ““a realistic appreciation” of their “essential character” by “reading [them] holistically and practically without fastening onto matters of form” (see also JP Morgan Asset Management (Canada) Inc. v. Canada (National Revenue), 2013 FCA 250 at para. 50). Nevertheless, it remains that, subject to limited exceptions, unless a request to include a particular framework for all computer-implemented inventions in the judgment is a remedy specifically sought in the notice of appeal, it should normally not be considered (Pfizer Canada Inc. v. Teva Canada Limited, 2016 FCA 218 at paras. 21-22; Boubala v. Khwaja, 2023 FC 658 at para. 27; Hendrikx v. Canada (Public Safety), 2022 FC 1068 at para. 27) (I do not mean here that the Court should then grant this request). I have much doubt that the limited exceptions to this general principle could apply here, given the ambit of the test adopted. One should be careful not to empty Rule 337(c) of its meaning by expanding these limited exceptions, especially when, as mentioned by the Federal Court in this case, the AG had not addressed the correctness of the test.
[35] I also ought to mention that there is no specific provision in the Federal Courts Act, R.S.C., 1985, c. F-7, like sections 18.1 (powers of the Federal Courts on judicial review applications) and 52 (Federal Court of Appeal’s power on appeal) that grants the Federal Court the power to issue this kind of general instructions in the context of a statutory appeal, such as this one. However, Rule 64 applies to all proceedings and refers to the Federal Courts’ discretion to grant declaratory relief which, like any other remedies, must have been properly sought. But, the discretion to grant such relief can only be exercised after considering the four-part test set out by the Supreme Court in Ewert v. Canada, 2018 SCC 30 at paragraph 81. In this case, there is no indication that the Federal Court turned its mind to this test. IPIC, relying on Steel v. Canada (Attorney General), 2011 FCA 153, and Defence Construction Canada v. Ucanu Manufacturing Corp., 2017 FCA 133 (Defence Construction), simply argued that the Federal Court’s discretion was not constrained and that judicial minimalism should not be applied here.
[36] However, these two decisions refer to the practice of courts dealing only with issues raised before them, which are necessary to determine the outcome of an appeal, and normally not to address other issues in obiter, even when properly before them. This is what judicial minimalism means. It is worth noting that in Defence Construction, our Court refused to determine in obiter the so-called jurisprudential issue that was otherwise properly raised by the appellant because it had been the subject of only a few decisions which did not cover all the relevant aspects of the issue, and because the matter was a complex one with compelling considerations on either side.
[37] Normally, the failure to consider the four-part test applicable to the general declaratory relief granted here would be an error in principle that would justify our intervention.
[38] That said, as the panel did not raise these issues at the hearing, and thus did not provide the parties with an opportunity to respond to the Court’s concerns in that respect, and because they are not necessary to dispose of the appeal, my discussion of these issues is limited to these general comments, which have no impact on the outcome of the appeal.
B. Did the Federal Court err in setting out the test at paragraph 3 of its judgment? (a) Purposive construction [39] The debate before the Federal Court was based on the Commissioner’s failure to apply Choueifaty to determine the essential elements of the claims, and later as to what the Federal Court actually decided in Choueifaty. Still, the Court failed to provide any guidance in that regard either in its reasons or in the first step of the test it sets out. There was probably little to add to what had already been said in Choueifaty and Amazon in respect of how to purposively construe the claims. Considering that our Court is not sitting in appeal of Choueifaty and that the Federal Court in that case was bound by Amazon and the general principles of purposive construction set out by the Supreme Court, I will not say much either. However, I must note that the error of the Commissioner in Choueifaty and in this case was not that she considered the problem and solution as part of her general assessment of the scope of the claims based on her reading of the applications as a whole, but rather that she identified the essential elements of the claims solely on that basis.
[40] As the Federal Court so clearly put it in Wenzel Downhole Tools Ltd. v. National-Oilwell Canada Ltd., 2011 FC 1323 (aff’d 2012 FCA 333) at paragraph 61:
To give a purposive construction to the claims of a patent, it seems to me that one should understand the purpose of the invention and the problem that the invention sought to address. For the most part, inventors come to their patentable inventions in order to solve a problem. What was the problem that the '630 Patent was intended to address?
[41] There is nothing ground breaking about such a statement, considering that letters patent are assimilated to a regulation as defined in subsection 2(1) of the Interpretation Act, R.S.C., 1985, c. I-21. It is trite law that the purpose and mischief intended to be addressed by the legislator is a relevant consideration in construing a regulation. Nevertheless, although identifying the problem and solution is relevant when construing the claims, it cannot be the sole or overarching element of the determination of their essential elements.
[42] Understanding the purpose, problem and solution may also be useful to identify to whom the patent is addressed. For example, even if the patentee in this case is a paint company, it appears (to me at least) that the monopoly claim could encompass the use of the computer-implemented system to select colours when used by artists, garden designers, furniture companies, or for organizing one’s wardrobe.
[43] Purposive construction is a difficult exercise even for judges, who nowadays appear to be relying more and more often on the presumption that all elements of a claim are essential unless established otherwise by the patentee or the applicant. This has resulted in a situation which, according to some, like the authors of a recent article entitled “Protection Against Infringement of Patents in Canada” (Ronald E. Dimock et al, “Protection Against Infringement of Patents in Canada” (2021) 36 CIPR 58), may not have been the result foreseen by the Supreme Court when it rendered its decisions in Free World Trust and Whirlpool. It appears to me that if there is indeed an overuse of this presumption, it places much emphasis on the art of claim drafting with little regard, at least before the issuance of a patent, to the fact mentioned in Amazon that claims can be expressed in a manner that is deliberately or inadvertently deceptive (Amazon at para. 44).
[44] In any event, I find it a step too far to conclude that a specialized administrative decision maker like the Commissioner is refusing to follow the case law of the Federal Courts and the Supreme Court of Canada in respect of purposive construction. In my view, CIPO and the Commissioner simply did not properly understand all of the subtleties of this difficult exercise. Our Court’s statement in Amazon at paragraph 42 that identification of “the actual invention” is relevant in the context of different aspects of the patentability assessment of the Commissioner, including patentable subject matter, could be at the core of the problem. As noted by the AG, the Commissioner did not appeal the decision in Choueifaty, as it did help clear up some misunderstanding in respect of purposive construction, such as the meaning of paragraph 55 of Free World Trust (Choueifaty at para. 38), and the current relevance of the Federal Court’s decision in Genencor International Inc. v. Canada (Commissioner of Patents), 2008 FC 608 (Choueifaty at paras. 34-35).
[45] I conclude on this first step of the test by noting that, as mentioned, the test does not provide any clarification as to the purposive construction exercise, instead simply stating that it must be completed. This was never in dispute.
(b) Subject matter patentability [46] The relevant sections of the Act referred to in this section as well as those of the 1970 version of the Act are reproduced in the Appendix to these reasons.
[47] Patentable subject matter is addressed in two distinct steps at paragraphs 3b and c of the Federal Court’s judgment. At paragraph 3b, with the use of the words “or whether”, the Federal Court creates a dichotomy between the statutory exclusions set out in subsection 27(8) of the Act on the one hand, and on the other, part of the reasoning adopted in Shell Oil and Amazon (new knowledge that needs to be practically applied) in determining whether the subject matter is patentable as an “art” within the meaning of an invention at section 2 of the Act. Then, it is only at paragraph 3c that the Commissioner should consider the statutory categories set out in section 2 of the Act (presumably those other than “art” to avoid redundancy), as well as judicial exclusions to patentable subject matter. The test appears to instruct that, only thereafter can the Commissioner consider novelty, obviousness, and utility.
[48] The first error that is apparent on the face of the test is that, as already mentioned, by dictating the order to be followed by the Commissioner, the Federal Court contradicted the statement made by our Court in Amazon that these patentability elements need not be considered by the Commissioner in any particular order (Amazon at para. 38). This is a fundamental error, considering that the Federal Court clearly stated that in adopting the test, it was simply following the binding authorities and was not making new law.
[49] The second error is in the actual order included in the test adopted by the Federal Court. Although the Federal Court meant to promote consistency, it did not explain nor seem to have considered the order in which the Commissioner currently considers inventions that do not involve computer implementation. Normally, the assessment of subject matter patentability begins with identifying the category of “invention” as defined in section 2 of the Act in which the subject matter of the patent falls.
[50] Computer-implemented invention is not a distinct category under section 2 of the Act, and depending on the nature of the invention claimed, it may fall under different categories. A manufacturing process may involve a computer. An improved refrigerator or oven may include a computerized component and a computer program to improve its working as a machine. One cannot simply assume that all computer implemented inventions fall under the category of “art”, nor can one construe this word as encompassing all other categories enumerated in section 2.
[51] It is difficult to understand why the Commissioner should look at the exclusions set out in subsection 27(8) before even examining whether the subject matter falls under the definition of an “invention” at section 2 of the Act. It may be practical for the Commissioner to do so in certain circumstances, but there is no basis for including this order in a legal test. The juxtaposition in paragraph 3b also implies a statutory interpretation of subsection 27(8) that gives no meaning to the “exclusions” therein, by simply requiring that the subject matter fall within the definition of an “art” under section 2 of the Act.
[52] Before us, IPIC and BM included in their submissions their statutory interpretation of subsection 27(8) based on the meaning of the word “mere”, but they did not point to any authority actually interpreting this particular provision of the Act. In response to a question from the panel, BM agreed that the word “mere” would be the equivalent of “as such” (this expression is used in statutory exclusions in England and has been the subject of several key decisions). In BM’s view, these statutory exclusions were meant to reflect the judicial exclusions developed in old common law cases, though BM did not discuss how said exclusions have been construed in other common law jurisdictions, such as Australia, which have relied on those very same old common law cases.
[53] In Amazon, this Court referred to subsection 27(8) without attempting to construe it. It simply noted at paragraph 60 that no Canadian jurisprudence had conclusively determined that business methods could not be patentable subject matter. The fact that it also had to address what kind of practical application was required to meet the definition under section 2 of the Act cannot be understood as being based on a construction of subsection 27(8).
[54] It is thus difficult to understand how the Federal Court could say that paragraph 3b of the test it adopted did not involve any statutory interpretation (FC Decision at para. 49), simply because it was using words of subsection 27(8). The addition of the word “only” and of “or whether” imply an interpretation of subsection 27(8) and creates a dichotomy that is not based on any authority having definitely interpreted subsection 27(8).
[55] Turning now to paragraph 3c of the Federal Court’s judgment, I fail to see any real reason why the Commissioner should deal with recognized judicial exclusions only at that stage. Many of these exclusions may be relevant to computer-implemented inventions, including for example those directed to professional skills (Tennessee Eastman Co. et al. v. Commissioner of Patents, [1970] Ex. C.J. No. 14, 62 C.P.R. 117, aff’d [1974] S.C.R. 111 and Lawson v. Commissioner of Patents, [1970] Ex. C.J. No. 13, 62 C.P.R. 101). BM acknowledges that professional skills are judicially excluded subject matter, but maintains its argument that a computer-implemented invention involving professional skills could nonetheless be patentable if it “comprises a practical application that employs [them]” (Paragraph 3b of the Federal Court’s judgment). This is just one of the judicial exclusions recognized in Canada (see enumerated exclusions in Monsanto Canada Inc. v. Schmeiser, 2004 SCC 34 at para. 133), and other similar exclusions, which have been found to be relevant to computer-implemented inventions in other common law jurisdictions, such as the US and Australia, and have yet to be considered by our courts.
[56] Turning now to the last portion of paragraph 3c, the Federal Court found it appropriate to include other patentability criteria, such as novelty, obviousness and utility, in the test it imposed, even though the Commissioner never carried out the exercise mandated under sections 28.2 and 28.3 of the Act (respecti