Contributory Infringement in Canadian Law
Author(s): Norman Siebrasse
Contributory infringement arises when a party knowingly supplies a direct infringer with a product especially adapted for use in a patented invention. Canadian courts have regularly stated that there is no liability for contributory infringement in Canadian law, in the absence of inducement. This article shows that there are nonetheless few cases actually refusing to impose liability on a contributory infringer, and none at the appellate level. The article argues that the reasoning and results in almost all the leading cases support a rule that the supply of a product especially adapted to infringe, and with no substantial non-infringing use, constitutes indirect infringement, even in the absence of active inducement. The current shape of the law is a result of a misreading of the early leading case of The Copeland-Chatterson Company Ltd v Hatton, in combination with the problematic decision in Slater Steel Industries Ltd v R Payer Co, which is the only prominent case refusing to impose liability in such circumstances. Slater Steel has been confined to its facts in both subsequent Court of Appeal decisions to address it, and this article argues that Slater Steel was wrongly decided on its facts. The article concludes that it is open to the courts to recognize that liability for contributory infringement may be imposed in Canadian law.
The problem of indirect infringement concerns the potential liability of a party—for convenience, the “indirect party”—which in some manner facilitates infringement by a direct party. The basis of an action for indirect infringement is generally the supply of an unpatented product by the indirect party, which is then used by the direct party to infringe the patent at issue, as when the indirect party supplies an unpatented compound that is then used by the direct party in a patented combination, or for a patented purpose. It is well established that the indirect party cannot be held liable solely for having supplied an unpatented product that is subsequently used by the purchaser to infringe; some extra element is required.
See AB Hassle v Canada (Minister of National Health and Welfare), 2002 FCA 421 at paras 57–58 [AB Hassle]; Pharmascience Inc v Sanofi-Aventis Canada Inc, 2006 FCA 229 at paras 36–60 [Pharmascience], rev’g on this point Genpharm Inc v Procter & Gamble Pharmaceuticals Canada, Inc, 2002 FCA 290 [Genpharm]. In Canada, liability for indirect infringement is normally established on the basis of inducement, which requires some kind of active encouragement or influence beyond mere supply of the product. In US law, even in the absence of inducement, liability may be imposed for what is referred to as contributory infringement, which arises when the indirect party knowingly supplies a product especially adapted for use in the patented invention.
35 US § 271(c) (providing that a party who supplies a component of a patented invention “constituting a material part of the invention, knowing the same to be especially made or especially adapted for use in an infringement of such patent, and not a staple article or commodity of commerce suitable for substantial non-infringing use, shall be liable as a contributory infringer”).
It is normally said that Canadian law does not impose liability for contributory infringement in the manner of American law.
Care is needed in respect of terminology. In US law, the term “contributory infringement” is commonly used to refer to the specific type of indirect infringement arising under 35 US § 271(c). The terminological distinction between contributory infringement and inducement appears to have been introduced on codification of US law in 1952; prior to that, “contributory infringement” was the general term corresponding to what we now call “indirect infringement”: see Global-Tech Appliances, Inc v SEB SA Eyeglasses, 563 US 754 at 761 (2011). However, even today “contributory infringement” is also sometimes used as a synonym for indirect infringement: see e.g., Donald S Chisum, Chisum on Patents (New York: Matthew Bender, 1997), ch 17 “Contributory Infringement” (dealing with both §§ 271(b) and (c)). In Australian law, the terms are also used interchangeably: see e.g. Apotex Pty Ltd v Sanofi-Aventis Australia Pty Ltd,  HCA 50 at para 295. In Canadian law, the terms are also sometimes used interchangeably: see e.g. MacLennan v Les Produits Gilbert Inc, 2008 FCA 35 at para 13 [MacLennan] (using the term “contributory infringement” for what is now more commonly referred to as inducement). Furthermore, in UK law, contributory infringement may also be used to refer to the specific type of indirect infringement that arises under s 60(2) of the Patents Act 1977 (UK), c 37 (see e.g. Grimme Maschinenfabrik GmbH & Co KG v Scott (t/a Scotts Potato Machinery),  EWCA Civ 1110 at para 4 [Grimme]), just as the same term in US law is often used to mean the specific type of indirect infringement arising under 35 US § 271(c). However, the elements are distinct under the different statutory provisions, so the term “contributory infringement” means different things in US and UK law. In this article, I use the term “contributory infringement” in the specific sense in which it is used in US law, to mean liability for the supply of a product with no non-infringing use (though without suggesting that Canadian law should adopt every detail of US law on this point), and not in the sense in which it is used in UK or Australian law. See John C Osborne, “Contributory Infringement” (1950) 14 CPR 75 at 89 [Osborne] (noting that “it is better to avoid the awkward phrase [contributory infringement] and to speak simply of direct and indirect infringement”). A clear statement of this position is found in MacLennan v Les Produits Gilbert Inc, where the Federal Court of Appeal stated, “In Canada, the courts have consistently held that selling a component intended to be incorporated in a patented combination (or process) without anything further, does not constitute an inducement to infringement, even where this component cannot be used for any other purpose.”
MacLennan, supra note 3 at para 33 [emphasis added]; and see to the same effect ibid at paras 38, 40; and see also Fox’s Annotation to Slater Steel Industries Ltd v R Payer Co (1968) in 38 Fox Pat C 139 at 140 (Ex Ct Can) [Fox’s Annotation to Slater Steel] (stating that “[t]here is, in Canadian law, no doctrine of contributing or constructive infringement such as exists in American law”). In this article, I argue that while the Canadian courts have consistently held that selling a component intended to be used in a patented invention does not in itself constitute indirect infringement, the rule against contributory infringement, reflected in the emphasized phrase, is not on such firm footing.
2.1 Effective Enforcement and the Risk of Overbreadth
Indirect infringement generally arises when the patentee does not wish to sue the direct infringer. This may be because it is impractical to do so, as when there are a large number of direct infringers;
See Dawson Chemical Co v Rohm and Haas Co, 448 US 176 at 188 (1980) (discussing the classic US contributory infringement case of Wallace v Holmes, 29 F Cas 74 (Conn Cir Ct 1871), and noting that the court permitted the action against the indirect party “rather than requiring the patentee to undertake the almost insuperable task of finding and suing all the innocent purchasers who technically were responsible for completing the infringement”). or because the patentee prefers not to, as when the direct infringer is also a customer of the patentee;
See MacLennan, supra note 3 at para 8 (noting that “[t]he appellants chose not to sue the forestry operators who, according to their own theory, are guilty of direct infringement of their patent when they use the Gilbert tooth as a replacement”). or, commonly, both. Thus, imposing liability for indirect infringement is intended to ensure effective enforcement of the patentee’s exclusive rights. This reflects the central question regarding infringement, set out by the Supreme Court in Monsanto Canada Inc v Schmeiser: “did the defendant, by his acts or conduct, deprive the inventor, in whole or in part, directly or indirectly, of the advantage of the patented invention?”
Monsanto Canada Inc v Schmeiser, 2004 SCC 34 at para 44 [Monsanto], quoting Saccharin Corp v Anglo-Continental Chemical Works (1900), 17 RPC 307 at 319 (Ch) [Saccharin].
The risk, on the other hand, is that of overbroad enforcement, because, by its nature, a finding of indirect infringement imposes liability for acts that do not in themselves infringe. In AB Hassle v Apotex, the Court of Appeal summarized the problem of overbroad enforcement as follows, in the context of a pharmaceutical patent for a new use of a known compound:
Thus [the defendant] cannot be prevented from obtaining [marketing authorization] solely on the basis that it will sell [the known compound]. If it were otherwise, then serious policy issues would arise. If there was any likelihood that a patient would consume a generic product for a patented use, then the generic product would not be approved. This would prevent new uses from being approved for existing drugs because there is always the possibility that someone somewhere will use the drug for the prohibited, patented purpose. This would result in a real injustice: since a generic company cannot possibly control how everyone in the world uses its product, the prevention of the generic from marketing the product would further fortify and artificially extend the monopoly held by the patent holders. The patent holder would, therefore, effectively control not just the new uses for the old compound, but the compound itself, even though the compound itself is not protected by the patent in the first place. The patent holders, as a result, would obtain a benefit they were not meant to have. In the end, society would be deprived of the benefit of new methods of using existing pharmaceutical medicines at a lower cost.AB Hassle, supra note 1 at para 57, quoted and adopted in Sanofi-Aventis Canada Inc v Apotex Inc, 2006 FCA 357 at para 19 [Sanofi-Aventis]; and see to the same effect Pharmascience, supra note 1 at para 58.
Although this statement is directed at the pharmaceutical context of the case, the problem of overbroad enforcement it describes is more general.
The central problem of the law of indirect infringement, therefore, is to balance effective enforcement of the patentee’s exclusive rights against the risk of overbroad enforcement. Since imposing liability for indirect infringement will almost always allow for more efficient enforcement of the patent rights, the policy issues related to indirect infringement generally turn on whether a particular rule will unduly restrict the unpatented use.
2.2 Contrast with Other Jurisdictions
In most other common-law jurisdictions, the law of indirect infringement looks to both the nature of the product and the indirect party’s knowledge or actions beyond mere supply; as the range of potential uses becomes narrower, the requirement for knowledge or inducement becomes more stringent.
Australian law, for example, considers three categories of products, relaxing the intent requirement as the product becomes more specific to the infringement. For products capable of only one use, the supply is infringing in itself.
Patents Act 1990 (Cth), ss 117(1), (2)(a). At the other extreme, for staple commercial products, the supply is infringing only if the supplier induced infringement.
Ibid, ss 117(1), (2)(c). Paragraph (c) applies “in any case,” which includes staples. The intermediate category, where the product is capable of more than one use but is not a staple, attracts an intermediate intent requirement, under which supply alone is not sufficient to establish indirect infringement, but neither is inducement required; instead, the supplier is liable if it had reason to believe that the product would be used for an infringing purpose.
Ibid, ss 117(1), (2)(b).
While the details vary in UK and US law, a similar trade-off is seen, with a product that is less specific to the infringement requiring a heightened degree of knowledge or inducement on the part of the indirect party.
In UK law, an indirect party may be liable for supplying an essential element of an invention to a direct infringer if the indirect party does or should know that the direct infringer intends to put the invention into effect, unless the item is a staple commercial product, in which case liability must be based on inducement: Patents Act 1977 (UK), c 37, ss 60(2), (3). In US law, an indirect party will also be liable for inducing infringement, and for supply of a component of a patented invention which it knows to be especially adapted for use in the patented invention, and which is not a staple of commerce (subject to the relevant knowledge requirements): 35 USC §§ 271(b), (c).
2.3 Commercial Staples
In MacLennan, the Court of Appeal stated that the mere sale of a component intended to be incorporated in a patented invention does not constitute indirect infringement, even where this component cannot be used for any other purpose. The court acknowledged that this rule “may seem questionable at first glance,”
MacLennan, supra note 3 at para 34. and offered the rather weak explanation that “[a]t the very least, it would be incongruous if the sale of an article, which in itself is not protected and which is therefore legal, becomes illegal without any other action being taken by the seller.”
Ibid. To say that this explanation is weak is not to criticize the court. On the contrary, the court was being careful not to overstate the point, and rightly so, as the strength of the argument turns on the nature of the product.
In the context of the supply of a staple commercial product, the point is very strong, as enforcement against indirect infringement would limit legitimate activities. For example, Innes v Short
(1898), 15 RPC 449 (QBD) [Innes v Short]. concerned a patent for the use of a common compound, metallic zinc, as an anticorrosive in steam boilers. While liability was imposed for inducing infringement, the court remarked:
It would be nonsense to say that a person is to be restrained from perfectly legitimate trade, namely, selling an article of commerce, because he happens to know … that the buyer intends to put it to some improper use.Ibid at 451, quoted with approval in MacLennan, supra note 3 at para 35, cited in Copeland-Chatterson v Hatton et al (1906), 10 Ex CR 224 at 243, 1906 CarswellNat 10 [Copeland-Chatterson], and cited by François Grenier, "Contributory and/or Induced Patent Infringement" (1987) 4 CIPR 26 at 27 [Grenier].
Clearly, a party cannot be made liable for supplying a staple item of commerce, without knowledge or intent that it will be used by the direct infringer for an infringing purpose, as those sales are almost always for legitimate purposes. Asking the vendor to proactively verify the purchaser’s intent would be an impossible burden on the sales clerk, and surely would never be attempted, even if liability were imposed. Such a rule would fail both aspects of the policy balance: it would have a chilling effect on legitimate trade and do nothing to reduce infringement.
Going one step further, if a vendor sells a staple item of commerce, that sale does not become an infringement simply because the purchaser tells the vendor that he intends to use it to infringe a patent. It might be suggested that imposing liability in such a case would not present a problem of overbroad enforcement, because a purchaser would never be prevented from acquiring the product for legitimate purposes, and there would be no verification burden at the point of sale. However, as a practical matter, a typical sales clerk could not be expected to know what acts might infringe a patent and, as staples of commerce are by their nature widely available, the direct infringer, forewarned after being once refused, would simply be more circumspect in his approach to a different supplier.
Accordingly, it is not surprising that on this point the law is the same in Canada, the United States, the United Kingdom, and Australia: the supply of staple commercial products attracts liability for indirect infringement only on the basis of inducement.
See 35 US § 271(c); Patents Act 1977 (UK), c 37, s 60(2); Patents Act 1990 (Cth) s 117(2)(c). Inducement requires some positive or “active” step by the indirect party that relates to the infringing act—most typically, inducement is established on the basis of supply of the product accompanied by instructions to use the product so as to infringe.
See e.g., Sanofi-Aventis Canada Inc v Novopharm Ltd, 2007 FCA 167 at para 11 (in which the court stated, “Infringement by inducement may be established, for example, by inferences reasonably drawn from the contents of the product monograph for the generic drug product,” though no inducement was found on the facts). Requiring active steps generally moves the locus of liability up the management chain of the indirect party, from a lowly sales clerk to a party with responsibility for marketing the product, and it is not an unreasonable burden to ask the person who is marketing the product for a specific purpose to verify whether that purpose is wrongful.
Full patent clearance is notoriously difficult in some fields, but an indirect party marketing a product for a specific purpose is normally in at least as good a position as the purchaser to verify whether that specific purpose is patented.
2.4 Products with a Substantial Non-Infringing Use
The problem is different when the product in question is not a staple, yet has a substantial non-infringing use. The point has recently arisen prominently in the general context of a patent claiming a new use for a known pharmaceutical, and the question is whether a generic producer will infringe by selling the unpatented pharmaceutical itself for non-infringing purposes, even though it knows that infringement will probably or inevitably result. Canadian law is perfectly clear on this point that there is no liability for indirect infringement in this context, absent some form of inducement, such as directions to infringe provided in the product monograph.
In the pharmaceutical context, see AB Hassle, supra note 1; Pharmascience, supra note 1, and Sanofi-Aventis, supra note 8. These cases reversed on this point Genpharm, supra note 1, which had held at paras 50–51 that infringement by the generic would be established if it was inevitable that the generic’s product would be used for the patented purpose: see Sanofi-Aventis, ibid at paras 8–11, holding that Genpharm was no longer good law on this point and should be confined to its facts. Outside the pharmaceutical context, see also Beloit Canada Ltee/Ltd et al v Valmet Oy (1988), 20 CPR (3d) 1, 18 CIPR 1 (FCA) [Valmet]. Canadian law is similar to US law in this respect,
See e.g. Takeda Pharmaceuticals USA, Inc v West-Ward Pharmaceutical Co, 785 F (3d) 625 (Fed Cir 2015); Sanofi v Watson Laboratories Inc, 875 F (3d) 636 (Fed Cir 2017). but it contrasts with UK law, under which indirect infringement will be established if the supplier knows or should know that some ultimate users will intend to use or adapt the product so as to infringe, even in the absence of any inducement on the part of the indirect party.
The leading UK case is Grimme, supra note 3; and see KCI Licensing Inc v Smith & Nephew plc,  EWCA Civ 1260 at para 53 (summarizing the law); Warner-Lambert Company, LLC v Actavis Group Ptc EHF,  EWCA Civ 556 at para 47 (quoting Grimme).
The policy considerations are substantially more difficult in this context, as there is a direct trade-off between efficient enforcement and overbroad enforcement. When the product supplied is not a staple of commerce, and so is not widely available, preventing one or a few suppliers from providing it to the direct infringers, far from being futile, might well be an effective means, and perhaps the only effective means, of preventing infringement. If the indirect party cannot be stopped from supplying the product in these circumstances, the patent incentive may be fatally undermined, as enforcement against direct infringers may well be impractical. On the other hand, as the Federal Court of Appeal explained in the passage from AB Hassle quoted above, if the generic is enjoined, then the patentee will effectively extend its monopoly to uses that are no longer patented. In the absence of creative solutions, which may or may not be possible,
See Warner-Lambert Co, LLC v Actavis Group PTC EHF,  EWHC 485 (Pat) in which the court ordered the UK National Health Service to send guidance to doctors telling them to prescribe the patentee’s product by brand name when prescribing for the patented use. It is not clear how effective this approach will be in the United Kingdom, nor is it clear whether a similar approach would be legally possible or practically effective in Canada. the problem is one of choosing between two imperfect alternatives.
The law on this point is different in the United Kingdom and Canada because the courts have adopted different solutions to this difficult problem. Canadian courts have consistently given priority to the need to ensure that “the public should not be deprived of its ability to use known products for known uses on the basis of patents for new uses of such products.
”Solvay Pharma Inc v Apotex Inc, 2008 FC 308 at para 200 (describing this as an “arch principle of patent law”). The UK courts, on the other hand, have favoured the need to ensure effective enforcement of the patentee’s exclusive rights.
One might debate which approach strikes the better balance in the case of products with a substantial non-infringing use, but that is not the goal of this article. Canadian law is clearly different from UK law on this point, and I am not suggesting that Canadian law should change. In particular, as discussed supra note 3, the UK approach is sometimes described as imposing liability for “contributory infringement” in this scenario, but that is distinct from the US meaning of the same term.
2.5 No Non-Infringing Use
In the third category, products with no substantial non-infringing use, the policy considerations are different again, and the argument for imposing liability becomes much stronger. Overbroad enforcement is not a concern; if the product has no non-infringing use, the public is not deprived of the ability to use the product for non-infringing purposes. At the same time, effective enforcement of the patent is facilitated, because the patentee can bring a single action against the vendor, rather than multiple actions against the direct infringer. In this case, Australian, UK, and US law all impose liability without the need for inducement, though subject to a knowledge requirement,
See 35 US § 271(c); Patents Act 1977 (UK), c 37, s 60(2); Patents Act 1990 (Cth) s 117(1), (2)(a). while Canadian law, at least formally, does not.
This analysis can also be couched in terms of causation. The argument in favour of imposing liability on the supplier of a product that is used to infringe is that the supply of the product was a cause, and indeed a “but for” cause, of the infringement; but that is true even when the product is a staple. There are always multiple causes of any wrong, and not all are culpable.
A fire in a wastepaper basket is caused “not only by the dropping of a lighted match, but also by the presence of combustible material and oxygen.”
The example is adapted from John G Fleming, The Law of Torts, 8th ed (Sydney, AU: Law Book, 1992) at 193, as quoted in Athey v Leonati,  3 SCR 458 at para 17, 140 DLR (4th) 235. In the case of a staple, the supply of a product is a cause of the infringement in the same sense that oxygen in the air is the cause of a fire when a match is negligently thrown into a wastepaper basket. The oxygen is necessary, but pervasive; if the particular molecules that were consumed had not been used, some other molecules of oxygen would have been available instead. A product with a substantial non-infringing use is akin to the wastebasket itself. Banning wastebaskets might be effective in preventing wastebasket fires, but the public would be deprived of a place to put waste. It would not be entirely futile to ban wastebaskets, or require match-proof lids, as a means of preventing wastebasket fires, but whether it is sensible to do so depends on the balance of fire risk and convenience of waste receptacles. The match itself arguably corresponds to a product with no non-infringing use.
The more general point is that deciding when to impose liability for indirect infringement amounts to deciding when a cause-in-fact of the infringement should be considered a cause-in-law of the infringement, and this involves balancing the desire to prevent infringement against the need to allow legitimate uses.
In summary, the principled justification for imposing, or refusing to impose, liability for indirect infringement depends on the nature of the product. Consequently, when an indirect party has supplied a product knowing it will be used to infringe, there is a principled distinction to be made between a case where the product is a staple, and a case where it is especially adapted for use in the patented invention. That no liability is imposed in the former does not imply that liability should not be imposed in the latter.
With this framework in mind, I will turn to the cases, starting with the historical development of the doctrine of indirect infringement in Canadian law.
3.0Development of the Law
3.1 Copeland-Chatterson and the English Common Law
The modern Canadian law of indirect infringement traces directly back to the seminal 1906 decision of Burbidge J in Copeland-Chatterson v Hatton et al, which established inducement of infringement as a basis for liability and is also said to have adopted into Canadian law some English decisions that excluded any concept of contributory infringement, even when the product supplied by the indirect party could not be used for any other purpose.
Copeland-Chatterson, supra note 16. And see MacLennan, supra note 3 at paras 33, 38 (recapping the development of the law and stating that Copeland-Chatterson adopted the British jurisprudence on this point); and to the same effect see Fox’s Annotation to Slater Steel, supra note 4 at 140. Copeland-Chatterson was accepted and elaborated on in Jackett P’s important 1968 decision in Slater Steel Industries Ltd v R Payer Co.
(1968), 55 CPR 61, 38 Fox Pat C 139 (Ex Ct Can) [Slater Steel cited to CPR]. An influential 1987 article by François Grenier, “Contributory and/or Induced Patent Infringement,” reviewed the relevant early English cases and most of the Canadian cases up to that time, and summarized the law of inducement in essentially the current form.
Grenier, supra note 16.
The English case that was the focus of Copeland-Chatterson, and that is therefore putatively the ultimate source of the Canadian rule that the supply of a product with knowledge that it will be used for an infringing purpose can never constitute indirect infringement, is Townsend v Haworth.
(1875), 48 LJ Ch 770 (Ch), aff’d (1875), 48 LJ Ch 773 (CA) [collectively Townsend] published in Sykes v Howarth (1879), 48 LJ Ch 769 (CA) at 770, n 1 (sub nom Sykes v Haworth) [Sykes v Howarth]. In Townsend the invention concerned the use of two unpatented chemical compounds in combination for preserving textiles. The pleadings alleged that the defendant company sold the chemicals in question to a textile company—the direct party—for the specific purpose of using them for the patented method. It was also alleged that the defendant company agreed to indemnify the purchaser in respect of any patent litigation that might ensue “for the purpose of inducing” the purchaser to buy the compound,
Ibid at 770–71, 773. It appears that the defendants gave directions as to how to infringe, but this is not entirely clear: see ibid at 771. and the decision proceeded on the basis that the direct party would not have purchased but for the indemnity.
Ibid at 772 (noting that the allegation was that the defendant company had sold “to a person who would not have bought of them without an indemnity, and therefore would not have infringed in that way without they had given him an indemnity”). Even so, the defendant’s demurrer was allowed on the basis that these facts, even if proven, could not establish infringement.
Ibid at 773. While the chemicals supplied in Townsend were common chemicals—apparently a commercial staple, and certainly with substantial non-infringing use
Ibid at 771 (describing the compounds as “common substances,” “old chemical compounds” and “perfectly well known”).—the reasoning of Jessel MR did not turn on or even refer to this fact.
See generally ibid at 772. His conclusion was stated in general terms: “[T]he mere making, using, or vending of the elements, if I may say so, which afterwards enter into the combination is not prohibited by the patent.”
However, we must be wary before assuming that Townsend is good authority in Canada, as the result is clearly at odds with modern Canadian law. Townsend was not a case of mere supply with knowledge that the product will be used to infringe. The act of infringement was completed by a direct infringer; the act of the indirect party, in the form of the indemnification, was the “but for” cause of that direct infringement; and the indirect party knew that its influence would result in the infringing act. On its face, then, the defendant in Townsend would be found liable under current Canadian law for having induced infringement.
See the three-part test for inducement set out in Weatherford v Corlac, 2011 FCA 228 at para 162. Indeed, the patentee in Townsend specifically argued that the defendant company “had induced and procured the infringement.”
Townsend, supra note 31 at 771. Although Townsend is consistent with the proposition that the supply of a product with knowledge that it will be used for an infringing purpose can never constitute indirect infringement, that was not the basis for the holding. Rather, Townsend is based on the principle that there can be no liability for anything short of direct participation in the infringing act: “He must be a party with the man who so infringes and actually infringe.”
Ibid at 773 per Mellish LJ.
The facts in Copeland-Chatterson were similar to those in Townsend. The invention was for a combination of a binder with posts and sheets having both closed and open apertures, arranged so as to allow single sheets to be easily inserted or withdrawn.
“Binder and Sheets Therefor”, Can Patent No 51242 (19 November 1895), clms 8–11 (one of the patents at issue in respect of inducement: see Copeland-Chatterson, supra note 16 at 225). The defendant supplied the direct party with sheets adapted for use in the binder, and induced the direct party to purchase by undertaking to indemnify him against any action the patentee might bring.
Copeland-Chatterson, supra note 16 at 241. While the product supplied by the defendant was especially adapted for use in the invention, the facts are nonetheless directly encompassed by the broad principle stated by Mellish LJ in Townsend, and quoted by Burbidge J: “[S]elling materials for the purpose of infringing a patent to a man who is going to infringe it, even although the party who sells them knows that he is going to infringe it and indemnifies him, does not by itself make the person who sells an infringer.”
Townsend, supra note 31 at 773, per Mellish LJ; James LJ ibid similarly acknowledged the indemnification agreement. Both were quoted in Copeland-Chatterson, supra note 16 at 242–43. Nonetheless, Burbidge J found the defendants liable for inducing infringement on this basis.
Copeland-Chatterson, supra note 16 at 246–47.
There is a principled basis for distinguishing the cases, given that in Townsend the products supplied were common chemicals, while in Copeland-Chatterson the product was especially adapted for use in the patented invention.
Ibid at 231 (noting that, “[w]ith regard to the sheets, their distinguishing feature is to be found in their being made or adapted for use in the plaintiffs' binder”). However, as noted, the reasoning in Townsend did not turn on this fact, and accordingly Burbidge J did not distinguish Townsend on this basis. He acknowledged Townsend as “the strongest authority in favour of the defendants and against the plaintiffs that is to be found,”
Ibid at 243. but he did not accept that authority. Burbidge J began his discussion of the law by stating that “[i]t is clear, of course, that it is not an infringement of a patent to sell an article which in itself does not infringe, although it may be so used as to infringe such patent.”
Ibid at 242. He continued: “Going a step further, it is, I think, well settled in England that such a sale is not of itself an infringement although the seller knows at the time of the sale that such article is intended to be used by the purchaser in the infringement of the patent,” citing and discussing Townsend in particular.
Ibid. The contrast between the first statement, made in general terms, and the second, describing English law specifically, suggests that the second statement was meant only as a description of English law, and not accepting it as part of Canadian law. This hint was reinforced when Burbidge J went on to contrast Townsend with various other English patent cases in which liability was imposed on “the person who was not the actual infringer.”
Ibid at 243. The cases discussed were Sykes v Howarth, supra note 31; Innes v Short, supra note 15; and The Incandescent Gas Light Co Ltd v The New Incandescent Mantle Co et al (1898), 15 RPC 81 (QBD) [Incandescent Gas Light Co]. He then turned to English tort law cases imposing liability for procuring a wrong, which were to the effect that liability will be imposed for “all wrongful acts done intentionally to damage a particular person and actually damaging him.”
Copeland-Chatterson, supra note 16 at 245, quoting Lord Lindley in Quinn v Leathem,  AC 495 at 535 (HL) [Quinn v Leathem]. This proposition is evidently much broader than that stated in Townsend, and it is this which Burbidge J accepted, remarking that these general principles of the law “go far I think to remove the difficulty” that had motivated the decision in Townsend.
Copeland-Chatterson, supra note 16 at 245. In effect, Burbidge J declined to follow Townsend on the basis that it was inconsistent with the general tort law of contributory liability.
Burbidge J also disagreed with the policy rationale set out in Townsend, which was explicitly driven by the view that it was necessary to allow the indirect party to support the direct party’s infringement as a way of encouraging challenges to bad patents.
See Townsend, supra note 31 at 771, per Jessel MR (stating, “The notion of any other doctrine being upheld seems to me to be fraught with the most tremendous consequences. It would not only give the patentee with good patents a monopoly, but it would also give patentees with bad patents a monopoly, because they would be able to say, ‘No man shall supply goods which are not protected by the patent with a guarantee to the purchaser to save him harmless from the hands of the patentee.’”). Burbidge J pointed out that an indirect party who wishes to challenge a bad patent by providing an indemnity to induce direct infringement is free to do so, and if it turns out that the patent was invalid or not infringed, the indirect party would not be liable.
Copeland-Chatterson, supra note 16 at 246 (observing that the defendants “may have thought that the claims the plaintiffs were setting up could not be sustained, and if so they had a right to resist them and to indemnify [the direct party]. But they knew of the patent and of the plaintiffs' claims and they took their chances” [citations omitted]). While challenging bad patents is indeed desirable, the reasoning applies equally to a direct infringer,
As Jessel MR effectively acknowledged in Townsend, supra note 31 at 771–72, saying, “And we know what the effect of a threat of attack even by the holder of an invalid patent has as a general rule in deterring persons from purchasing goods so alleged to be an infringement.” and the threat of being held liable if the patent turns out to be valid provides an incentive to attack weak patents, rather than to infringe strong ones.
Thus, far from accepting Townsend, Burbidge J rejected its authority, both in principle and on the facts
See Osborne, supra note 3 at 89, noting, “If I am correct in my understanding, Mr. Justice Burbidge held that Townsend v. Haworth would be decided otherwise if the same facts were presented to-day on properly drafted Pleadings.” He accepted that there could be no liability without actual infringement, but he departed from Townsend as to what might “with propriety” be termed an infringement.
Copeland-Chatterson, supra note 16 at 247. He concluded his analysis with a question that became the basis for the Canadian law of inducement:
In short does not one who knowingly and for his own ends and benefit and to the damage of the patentee induces or procures another to infringe a patent himself infringe the patent?Ibid.
This holds that inducing infringement is infringement, but on its face it does not hold that there is no other act of indirect infringement that might “with propriety” be termed an infringement.
As noted, Townsend itself concerned a common chemical with substantial non-infringing uses. Vaughan Williams LJ’s decision in Dunlop Pneumatic Tyre Co v David Moseley & Sons Ltd is the basis for the proposition that supplying a product with knowledge that it will be used to infringe is not infringement, even if the product has no other use.
(1904), 21 RPC 274 (CA) [Dunlop v Moseley], quoted in MacLennan, supra note 3 at para 37; and see Osborne, supra note 3 at 92, noting, “The decision in Dunlop v. Moseley will probably continue to be followed so that a supplier of a part of a patented combination will not be guilty of infringement even if the part has no other use than in such combination and the supplier knows that it will be so used.” Grenier, supra note 16 at 34, cites, inter alia, “the English authorities” for the proposition that “even if the only use of the article is in a patented combination, the seller will escape liability if its activities fall short of inducing the direct infringer”; but the only English case cited by Grenier actually dealing with a product especially adapted to infringe is Dunlop v Moseley. The basis for Vaughan Williams LJ’s holding was the same as in Townsend: there can be no liability without direct infringement.
Dunlop v Moseley, supra note 58 at 278–79 (stating, “In truth, and in fact, veil it how you like, the Plaintiffs do not complain of any infringement of this Patent in which they say that the Defendants, one or other of them have taken part as actors. All that they complain of is the sale of these covers, which sale, it is said, must have been known to be a sale to a person intending to commit an infringement of one or other of these Patents”), and at 280 (stating that the plaintiffs “do not prove that the Defendants’ have actually infringed,” and that the case would fail even if it were established that the product “could not be used for any other purpose”). Although Burbidge J did not discuss Dunlop v Moseley, he correctly treated it as being decided on the same basis as Townsend.
Dunlop v Moseley was cited by Burbidge J in Copeland-Chatterson, supra note 16 at 242, n 2 as standing for the same proposition as Townsend. The decision of Vaughan Williams LJ in Dunlop v Mosley was also quoted by the Court of Appeal in MacLennan, supra note 3 at para 37 as being “[i]n the same vein” as Townsend. Therefore, he did not accept Dunlop v Moseley any more than he accepted Townsend.
Despite this, in an annotation to Slater Steel in Fox’s Patent Cases, the author, presumably Dr. Fox, stated:
There is, in Canadian law, no doctrine of contributing or constructive infringement such as exists in American law. The doctrine of contributory infringement was expressly repudiated by Burbidge J. in Copeland-Chatterson v. Hatton et al. (1906), 10 Ex. C.R. 224 at 241, when he said:
If the act complained of as a contributory infringement is in fact an infringement, well and good. The court has jurisdiction. But if it is not an infringement, the court has no jurisdiction, and will not acquire jurisdiction by introducing a term that is not to be found in the statute. The question is: Did the [indirect parties], in what they did, infringe the plaintiffs' patent? It is a question of infringement, not a question of contributing to an infringement by some act that falls short of being an infringement.Fox’s Annotation to Slater Steel, supra note 4 at 140.
With due respect, this assertion misreads Burbidge J’s statement. He was not rejecting liability for contributory infringement, while acknowledging inducement—which is no more to be found in the statute than is the term “contributory”—because, at the time, there was no clear distinction between the two.
See supra note 3 and accompanying text. What Burbidge J was saying in this passage is simply that regardless of the label, there is no infringement for anything which could not, “with propriety,” be termed an infringement.
Even if we acknowledge that Copeland-Chatterson did not accept the English common law cases into Canadian law, it is sometimes suggested that Jackett P’s decision in Slater Steel, discussed in more detail below, did.
See MacLennan, supra note 3 at para 33 (suggesting, without directly stating, that Slater Steel also adopted the English case law on this point). While Noël JA in MacLennan remarked on the “meticulous analysis” conducted by Jackett P,
MacLennan, supra note 3 at para 44 it must be said that Jackett P did not actually analyze the English cases, but rather, as he put it himself, “summarize[d] briefly” the effect of the decisions.
Slater Steel, supra note 29 at 71. His discussion was entirely descriptive, consisting largely of long quotations from the cases, with no comment as to whether he accepted any of them as good law in Canada.
Ibid at 71–80. Instead, he concluded by accepting Copeland-Chatterson, and in particular the reference to inducement, as stating the law.
Ibid at 82. Thus, Slater Steel does not accept the English cases any more than did Copeland-Chatterson.
The authority of Townsend, supra note 31, and Dunlop v Moseley, supra note 58, is further undermined by the fact that they are no longer good law in the United Kingdom: see the Patents Act 1977 (UK), c 37, s 60(2), (3). To the extent that they have been independently accepted into Canadian law, the fact that they have been legislatively reversed in England is irrelevant, but that fact does suggest that when interpreting Canadian law, we should not place too much weight on those early English cases, independently of their reception into Canadian law.
In summary, Burbidge J’s decision in Copeland-Chatterson was a very carefully reasoned synthesis of principles of tort and patent law, which deserves to be read as carefully as it was composed. Given the influence of English law on Canadian law at the time, it is not surprising that Burbidge J did not come out and say, “I refuse to follow the leading English patent cases on the issue before me,” but his position to that effect is nonetheless clearly articulated. Burbidge J did not adopt English law to the effect that it is not an infringement to supply a product even if the seller knows that it will be used to infringe (though neither is his opinion inconsistent with that proposition). His point was that the focus in Townsend on whether the indirect party’s acts themselves fell within the claims was misplaced in light of general principles of tort law, and the overarching question should be whether the indirect party’s act might be “with propriety” termed an infringement. It is fair to say that Copeland-Chatterson rejected the hostile approach to indirect infringement shown in the early English cases, and is more consistent than the English cases with the long-established principle that infringement consists of any act that “directly or indirectly” deprives the patentee of the advantage of the patented invention. Consequently, the bounds of indirect infringement in Canadian law should properly turn on the Canadian cases, not on the English decisions.
3.2 Slater Steel
There are relatively few Canadian cases dealing directly with a situation in which the product supplied had no substantial use other than as an element of the patented invention. The first was Copeland-Chatterson itself, which did impose liability, though it did so on the basis of inducement, and without specific reference to the nature of the product. The second was the 1924 case of American Arch Co v Canuck Supply Co.
 3 DLR 567, 30 RL (ns) 449 (Qc Sup Ct) [American Arch cited to DLR]. The patent at issue was for a locomotive fire-box, consisting of a particular arrangement of fire bricks and other components.
Ibid at 570, 573. The defendant was the selling agent for unlicensed manufacturers of fire-brick arches of a type with “one purpose and one purpose only, namely; to become a part of the plaintiff's combination in the practice of the invention.”
Ibid at 576. Duclos J in the Quebec Superior Court considered it “well settled” that this constituted infringement.
Ibid. While the bricks were supplied with instructions that would no doubt have supported a finding of active inducement under current law, Duclos J made nothing of this. The focus of his decision was the nature of the product. Quoting Burbidge J’s statement that inducing or procuring infringement is itself infringement, he remarked, “Much more so if, as in this case, the element furnished by the defendants can have no other use than in the practice of the invention.”
Ibid at 577. Duclos J ignored the English cases discussed in Copeland-Chatterson, and instead relied extensively on US cases. His decision was a clear endorsement of what we would now consider US-style contributory infringement: “the manufacture of an essential part of an infringing structure and its purposeful sale to another person to enable him to make the entire infringing article is infringement.”
Ibid at 578, citing Canada v Michigan Malleable Iron Co, 124 F 486 (6th Cir 1903); and see generally the discussion ibid at 577–78, citing a total of 10 US cases. He clearly regarded this as being entirely consistent with Copeland-Chatterson.
Although it is only a decision of a single judge of the Quebec Superior Court, American Arch is nonetheless significant in that it illustrates the same reluctance to follow Townsend and similar English cases that was shown in Copeland-Chatterson. It also illustrates that, at the time, Copeland-Chatterson was not seen as adopting the view of Vaughan Williams LJ, that the mere sale of a product with knowledge that it would be used to infringe could not constitute indirect infringement, even if the product had no other use. Curiously, even though American Arch deals directly and explicitly with indirect infringement, it was completely ignored by Grenier, who did not even cite it in a footnote, though he must have known about it.
American Arch, supra note 69, was discussed by Osborne, supra note 3, whose article was in turn discussed by Grenier, supra note 16. It has been ignored ever since, perhaps because it was considered to have been overtaken by Slater Steel—though, strictly, they are the same level of court—or perhaps simply because it had not been mentioned by Grenier.
The next decision dealing with a situation in which the product supplied had no substantial use other than as an element of the patented invention is that of Jackett P in Slater Steel, which has been influential both because the law was not fully developed at the time, and because of Jackett P’s reputation as a jurist.
MacLennan, supra note 3 at para 44. The invention at issue in Slater Steel related to a combination achieved by winding preformed armour rods around a stranded electric power transmission line in order to reinforce stress points.
Slater Steel, supra note 29 at 65. The defendants manufactured and sold the rods,
Ibid at 66, 83. which had no other practical use,
Ibid at 67, 83. along with a catalogue containing directions for using the rods in the patented combination,
Ibid at 67, 83–84. to the direct infringers, knowing that the rods were intended to be used, and would be used, to infringe.
Ibid at 84. Jackett P held on these facts that the plaintiffs had failed to establish inducement and thus infringement by inducement.
Ibid at 90. Jackett P proceeded on the assumption that the direct parties did indeed infringe (see Slater Steel, ibid at 86–87), so the first branch of the test was satisfied, and the action failed on the second branch. Consequently, Jackett P did not address the knowledge requirement: ibid at 90.
Slater Steel is particularly important because it is one of the only Canadian cases to refuse to impose liability for indirect infringement, even though the product was especially adapted for use in a patented combination. It is the primary authority cited by Grenier for the proposition that liability will not be imposed absent inducement “even if the only use of the article is in a patented combination.”
Grenier, supra note 16 at 34. However, Slater Steel is not strong authority for that proposition, for two reasons.
First, Jackett P did not hold that there can be no liability absent inducement, even if the product has no non-infringing use. The plaintiff had drafted its pleadings in an attempt to argue US-style contributory infringement,
See Slater Steel, supra note 29 at 69, noting that the allegations in paragraphs 5 and 2 of the statement of claim reflected sections 271(b) (inducement) and 271(c) (contributory infringement), respectively, of the US Act. but the point was dismissed by Jackett P, primarily on the basis that the essential knowledge element had not been alleged.
Slater Steel, supra note 29 at 69–70, noting that it had not been alleged that the defendants knew that the products in question were especially adapted to infringe, as is required under § 271(c) of the US Act. A second problem with the allegation of contributory infringement is that, at the time, it was not clear whether the use of a patented invention by the Crown in right of a province was infringing under s 19 of the Patent Act, RCS 1985, c P-4: see Formea Chemicals Ltd v Polymer Corp Ltd,  SCR 754, 55 CPR 38, and the Canadian Patent Reporter Annotation in Slater Steel, supra note 29 at 63–64. This issue was common to both the inducement and contributory infringement allegations, though Jackett P did not have to address it in light of his conclusion on inducement: see Slater Steel at 90. This was relevant because the parties directly practising the invention were provincial Crown power companies, and if they were not infringers, neither form of indirect infringement could be established. Consequently, he simply never addressed the question of whether contributory infringement in the US sense—namely, supply with knowledge of a product especially adapted to infringe—might constitute infringement in Canadian law.
See Slater Steel, supra note 29 at 70, noting that the US jurisprudence had not been argued to him and stating that, “[i]n the absence of any such information, in my view, the United States jurisprudence cannot be regarded as being of any assistance in construing the provisions of the Canadian statute” [emphasis added]. This is not a rejection of the US position as such; Jackett P simply declined to address a point that had not been argued. Almost the entirety of Jackett P’s analysis, including his discussion of English case law and Copeland-Chatterson, was addressed to the issue of inducement.
See Slater Steel, supra note 29 at 83, tying the discussion of the case law to paragraph 5 of the statement of claim. In concluding his review, Jackett P stated that he accepted the law applied by Burbidge J in Copeland-Chatterson as being the law under the Act “for the purpose of deciding this case.”
Slater Steel, supra note 29 at 82. By this qualification, Jackett P appears to have been leaving the door open to the possibility of US-style contributory infringement, if properly pleaded. Thus, while it is true that Jackett P held that inducement had not been established, and that he declined to impose liability even though the product was especially adapted to infringe, he did not hold that there can be no liability for the supply of a product especially adapted to infringe, absent inducement, as that allegation failed on the pleadings.
With that said, Jackett P’s analysis was certainly not welcoming to any form of contributory infringement. When Jackett P turned to consider inducement, he acknowledged that the product sold by the defendants had no non-infringing use, but evidently regarded this as irrelevant to the question of inducement, as he did not mention it further in his analysis. This raises the second problem with relying on Slater Steel on this point. On the inducement issue, the result and reasoning in Slater Steel are difficult, even impossible, to reconcile with the subsequent Court of Appeal decisions in Windsurfing v BIC Sports
(1985), 8 CPR (3d) 241, 7 CIPR 281 (FCA) (sub nom Windsurfing International Inc v Triatlantic Co) [Windsurfing cited to CPR]. and MacLennan,
Supra note 3. both of which confined Slater Steel to its facts.
The more extended discussion of Slater Steel is found in MacLennan, as Slater Steel was the main basis for the defendant’s argument.
Ibid at para 20. The court noted that the facts in the two cases were “very similar”: both involved a patented combination; the product sold by the defendant was a component that could be used only in the patented combination; it was sold with the knowledge that it was intended for use, and would be used, in the patented combination; and the product was accompanied by directions for working the patented combination.
Ibid at paras 45–46. The court nonetheless held that Slater Steel was distinguishable on the facts.
Ibid at para 47. Noël JA characterized Jackett P’s holding as “relying on his own experience in the business world,” which led him to believe “that it was inconceivable that major companies like B.C. Hydro [one of the direct parties] could have been influenced to commit wrongful acts by mere sellers of pre-formed rods [the indirect party].”
Ibid at para 48. Noël JA stated, “But for this conclusion, which is totally particular to that case, I have difficulty in seeing how President Jackett could have done anything other than conclude that infringement by inducement had been established in accordance with the state of the law that he described.”
Ibid at para 49. As noted by Noël JA, ibid, Slater Steel was similarly distinguished in Windsurfing, supra note 89 at 266–67, in which the court stated that the determination that there was no inducement “was made by inferences drawn from the facts as found.” Indeed, Slater Steel is difficult to reconcile with a host of cases that have imposed liability for supply of a product accompanied by instructions to infringe, regardless of whether the product is especially adapted for use in the invention.
See Abbott Laboratories Ltd v Novopharm, 2006 FC 1411 at paras 40–42, aff’d 2007 FCA 251 at paras 26–27 (primarily based on product monograph); AB Hassle v Genpharm, 2003 FC 1443 at paras 155–56, aff'd 2004 FCA 413 at paras 16–17; Genpharm, supra note 1 at paras 50–51; Grenke v DNOW Canada ULC, 2018 FC 565 at paras 11–14; Hospira Healthcare Corporation v Kennedy Trust for Rheumatology Research, 2018 FC 259 at para 333; Baker Petrolite Corp v Canwell Enviro-Industries Ltd, 2001 FCT 889 at paras 135–39, rev’d on other grounds 2002 FCA 158; Procter & Gamble Co v Bristol-Meyers Canada Ltd (1978), 39 CPR (2d) 145,  FCJ No 812 (FCTD), aff’d (1979), 42 CPR (2d) 33, 28 NR 273 (FCA); Varco Canada Ltd v Pason Systems Corp, 2013 FC 750 at paras 253–55; Glaston Services Ltd Oy v Horizon Glass & Mirror Ltd, 2010 FC 1191 at paras 88–91; Uview Ultraviolet Systems Inc v Brasscorp Ltd, 2009 FC 58 at para 237; Aventis Pharma Inc v Apotex Inc, 2005 FC 1381 at paras 46–69. Noël JA’s pointed statement suggests that the Court of Appeal was not entirely persuaded by Jackett P’s reasoning; it is more accurate to say that Slater Steel was confined to its facts, than distinguished on them. The factual matrix, which was indeed central to Jackett P’s holding, made it unnecessary to formally hold that Slater Steel was wrongly decided. But Slater Steel is such an important case on this issue that it is worth probing more deeply, and, with respect, Jackett P’s own reasoning does not allow Slater Steel to be distinguished on the facts.
While Jackett P’s reasoning is not entirely easy to follow, the key, as emphasized by Noël JA, was his conclusion that “none of [the direct parties] would decide knowingly to do something that it regarded as illegal.”
The key passage in Slater Steel, supra note 29, is at 89–90, quoted in MacLennan, supra note 3 at para 48. Jackett P continued by saying, “It is even less probable that they could be ‘induced’ or ‘procured’ against their own judgment to do something illegal by a ‘one man show’ such as that operated by the [indirect party].” This point was evidently crucial as it was repeated three times in the same paragraph, in slightly different ways. The implication is that because infringement is illegal, the indirect parties therefore could not have induced infringement. However, even accepting the premise that the direct parties could not have been induced to do something illegal, it does not follow that they could not be induced to infringe, because they might not have known that the act in question constituted infringement (and infringement does not require intent or knowledge). In particular, Jackett P clearly held that the direct parties must have known of the patents,
Slater Steel, supra note 29 (stating, at 89, “I find it impossible to believe, in the absence of clear evidence, that [the plaintiffs] did not, in some manner, bring the existence of their patent rights to the attention of their potential customers,” and, at 90, “I find it most improbable that the plaintiffs failed to make their patent rights known”). and he was evidently of the view that the direct parties must have believed—wrongly, as it turned out—that what they were doing was legal, either because the patents were invalid or because the direct party had a licence or did not need one.
Ibid at 89 (listing the reasons that the direct parties might have believed that the acts found to have infringed were nonetheless legal). The fact that the direct parties could not have been induced to do something that was illegal is irrelevant when the issue was whether they were induced to do something that they believed to be legal.
Moreover, if we suppose, contrary to Jackett P’s finding, that the direct party did not know of the patent, then inducement is also plausible for the same reason, because, again, the indirect party would not believe the act in question was wrongful.
Consequently, even on the unusual facts of the case, the holding in Slater Steel is contrary to the current law as set out in MacLennan. With due respect, despite Jackett P’s prominence as a jurist, Slater Steel was not his finest moment. It is a clear example of motivated reasoning, as is highlighted by his entirely improper use of his own experience to establish a key evidentiary point.
There is an interesting question as to what might have been motivating Jackett P. On the facts, the patentee in Slater Steel, supra note 29, would have had a practically effective claim against the direct infringer, and we might speculate that Jackett P was of the view that the patentee should have brought an action against the large direct party, rather than against the smaller indirect party: see Grenier, supra note 16 at 35, saying, “one cannot escape noticing” that in several of the cases in which no liability was imposed, “the patentee had, apart from his remedy against the alleged inducer, a practical alternative: the direct infringer was easily identified and it was a simple matter to bring him to court,” while in the two leading cases imposing liability, “it was practically impossible to sue the end users.” Although there is some logic to this view, the counterargument is that if the indirect party has in fact done something wrongful, it is not relevant that another party might also have acted wrongfully, and it should not be for the court to tell the patentee which of two wrongdoers it may pursue. Perhaps for that reason, the courts have never explicitly pursued this line of reasoning, and it is no more than speculation to suggest that this was what motivated Jackett P. In any event, even if that was Jackett P’s motivation, it does not allow us to distinguish MacLennan, supra note 3, as the same might be said in that case.
Another important decision bearing on the issue of supply of a product with no non-infringing use is the Court of Appeal decision in Windsurfing.
Supra note 89. The invention at issue was a sailboard, which is a combination of known components.
Ibid at 265. The defendant was selling a kit of complete parts, with rudimentary instructions for assembly into an infringing whole.
Ibid. The Court of Appeal held that the defendant infringed:
I think it beyond dispute that the only inference to be drawn from the voluminous evidence in this case is that the respondent knew and intended that the ultimate purchaser would utilize the sailboard parts for the assembly of a usable sailboard which, upon assembly, would infringe the appellants' patent. It thereby became a party to such infringement, in my view.Ibid at 266. Infringement was not addressed at trial as the patent was held to be invalid. The finding of invalidity was reversed on appeal, and the Court of Appeal chose to address infringement itself.
While the case was concerned with a kit, this reasoning is not specific to a kit. On its face, this passage says that the sale of the product with the intent and knowledge that it would be used for an infringing purpose establishes infringement, without any requirement of active inducement.
See also ibid at 265: “The respondent clearly is not selling parts. It is selling parts for the purpose of making a sailboard. Without assembly there can be no sailboard. Without assembly there can be no purpose in a purchaser buying the unassembled parts since, unassembled, they cannot be used for the purpose for which they are purchased, that is, to sail. To suggest that a patent infringement suit can be successfully avoided by selling parts as components of a kit in contradistinction to their sale assembled is, in my view, errant nonsense” [emphasis in original]. While this is directed specifically at a kit, the logic is that without infringement, there is no point to the purchase. This logic is equally applicable to a single part especially adapted for infringement. The court’s reasoning may be compared with the holding in Heaton-Peninsular Button Fastener Co v Eureka Specialty Co, an early leading US case on what is now known as contributory infringement, in which the defendant sold unpatented staples that could be used only in a patented method of fastening buttons to shoes:
It is the knowledge that the staples made and sold by defendants are to be used for the purpose of infringing, coupled with the active intent that they shall be so used, which, in a case like this, constitutes contributory infringement.77 F 288 at 297 (6th Cir 1896). note that the case concerned “staples,” in the sense of a bent piece of wire, not “staples” in the sense of a basic commodity.
The emphasis in both cases is on the defendant’s knowledge and intent that the product sold be used to infringe. That the product in both cases was especially adapted to infringe is important, because this allows the inference of intent to be drawn from the sale of the product alone, in contrast to a case in which the product has a substantial non-infringing use.
Certainly, there are many references to inducement throughout the Windsurfing decision—the court concluded, for example, that the defendant “by its conduct induced and procured purchasers” to infringe the patent,
Windsurfing, supra note 89 at 268. but the conduct was simply the supply of kits, along with the rudimentary instructions. Recall that 1985 was still early in the development of the law. Copeland-Chatterson and Slater Steel were the main cases of note, and the court distinguished Slater Steel on its facts.
Ibid at 266–67. Windsurfing was not applying a developed body of law on inducement; rather, it was developing the law of inducement by deciding what might properly, or “with propriety,” be termed infringement. So far as active inducement is concerned, Urie J remarked that the supply of a product with only an infringing use, “in my view, is inducement even where the printed instructions are limited to the extent disclosed in the evidence in this case.”
Ibid at 265–66. This was not to say that the printed instructions were essential to the finding of inducement, but rather that the supply of a kit spoke so strongly to inducement that more detailed instructions were not necessary. Thus, Windsurfing cannot be seen as endorsing the view that there can be no liability for contributory infringement.
This article began with the statement by the Court of Appeal in MacLennan that Canadian courts “have consistently held that selling a component intended to be incorporated in a patented combination (or process) without anything further, does not constitute an inducement to infringement, even where this component cannot be used for any other purpose.”
MacLennan, supra note 3 at para 33 [emphasis added]. This statement has two parts: the basic proposition that selling a component intended to be incorporated into a patented invention does not, without anything further, constitute indirect infringement; and the emphasized phrase to the effect that contributory infringement does not constitute indirect infringement. The court in MacLennan cited Copeland-Chatterson, Slater Steel, the Court of Appeal decisions in Beloit v Valmet and AB Hassle, and the Trial Division decision in Permacon Quebec Inc v Les Enterprises Arsenault & Freres Inc, as well as Grenier’s article, as support for its statements on indirect infringement.
Ibid, citing Copeland-Chatterson, supra note 16; Slater Steel, supra note 29; Valmet, supra note 20 at 15; Permacon Quebec Inc et al v Les Enterprises Arsenault & Freres Inc et al (1987),  2 FC 179, 19 CPR (3d) 378 at 384–85 (FCTD) [Permacon]; AB Hassle, supra note 1 at para 18; and Grenier, supra note 16.
Canadian courts have indeed consistently held that selling a component intended to be incorporated into a patented invention does not, without anything further, constitute indirect infringement. Valmet and AB Hassle were evidently cited by MacLennan solely for this point; in both cases, the court declined to impose liability for indirect infringement, but the product supplied had a substantial non-infringing use, and the question of contributory infringement was not addressed.
In AB Hassle, supra note 1, the patent was for a new use for a known pharmaceutical, and it is clear that the compound had substantial non-infringing uses. The court held that the sale of a product with knowledge that it would inevitably be used for an infringing purpose did not in itself constitute indirect infringement, and the court did not address or cite any cases on the issue of the sale of a product with no non-infringing use. In Valmet, supra note 20, the invention was for a combination, and the parts supplied by the indirect party were described as being “well known”: Valmet, supra note 20 at 14. The court summarized the law, at 14, in the passage quoted supra at note 20, citing Windsurfing and Slater Steel, but the statement of the law did not purport to be definitive (there “seems” to be only two exceptions). Valmet therefore adds nothing to the law related to contributory infringement beyond those two cases. (The court also cited Incandescent Gas Light Co, supra note 49, as supporting the proposition that liability might be imposed on a joint infringer, but, again, this is not directed at contributory infringement.)
The additional proposition that the supply of a product especially adapted to infringe cannot constitute the “further” element necessary to establish indirect infringement, is not so well established. While MacLennan itself of course stated a rule against contributory infringement, it did not apply that rule on the facts.
The patented invention in MacLennan was a combination of saw tooth and tooth holder, for use in the forestry industry. The saw tooth as such was not patented.
MacLennan, supra note 3 at para 4. Individual saw teeth are commonly broken during harvesting operations, and the advantage of the patented invention was that it allowed the easy replacement of individual teeth.
Ibid at para 6. The defendant sold replicas of the patentee’s teeth, which had no use except as part of the patented combination, and distributed a price list with an equivalency table, indicating which series number of the original tooth its teeth were designed to replace.
Ibid at para 7. While the teeth sold by the defendant were especially adapted to infringe, the court’s holding that the defendant was liable was expressly not based on contributory infringement, but rather on inducement, with the price list and equivalency table in particular constituting the inducement.
Ibid at para 40: see also para 13 (describing the accepted three-part test for inducement) and para 42 (finding inducement). Thus, on the facts, the court’s statement of the rule against contributory infringement was strictly obiter, since the court held the indirect party liable due to inducement, on the basis of what might fairly be described as a “relatively trifling additional fact.”
See the prescient remark by Osborne, supra note 3 at 92–93, suggesting that when the defendant supplied a product with no non-infringing use, despite the English authority against contributory infringement, “perhaps a relatively trifling additional fact pointing in the same direction is all that a plaintiff will require to establish his cause of action. I do not imagine that our Courts will be astute to protect those who profit from the invasion of a legitimate monopoly.”
The court in MacLennan stated that Copeland-Chatterson “adopted the British jurisprudence that was the source of this rule.” With due respect, this is not correct; as discussed above, Burbidge J declined to accept the restrictive British jurisprudence on indirect infringement, and instead followed the British tort law cases imposing liability for “all wrongful acts done intentionally to damage a particular person and actually damaging him.”
Copeland-Chatterson, supra note 16 at 245, quoting Lord Lindley in Quinn v Leathem, supra note 50. Burbidge J never held, or suggested, that nothing but inducement could “with propriety” be termed an infringement.
Copeland-Chatterson, supra note 16 at 247. American Arch, decided shortly afterward, clearly considered Copeland-Chatterson to be consistent with imposing liability for contributory infringement.
Grenier’s article, cited by MacLennan, does state that “even if the only use of the article is in a patented combination, the seller will escape liability if its activities fall short of inducing the direct infringer,”
Grenier, supra note 16 at 34. citing Slater Steel as authority. However, while inducement was directly at issue in Slater Steel and the court refused to impose liability for the supply of a product accompanied by instructions to infringe, as discussed above it is clearly not authority for a general proposition that such an act cannot constitute indirect infringement, outside of its particular facts. Similarly, while Slater Steel refused to impose liability for the supply of a product especially adapted to infringe, it is very difficult to see how it can be authority for a general proposition that such an act cannot constitute indirect infringement, particularly given that contributory infringement was not pleaded and thus not directly addressed. The Court of Appeal in MacLennan cited Slater Steel, but only as giving a “history of the evolution of the jurisprudence,” and not as authority in itself.
MacLennan, supra note 3 at para 33. It might be suggested that the court was implicitly approving Slater Steel, and merely highlighting the review of the jurisprudence found therein; but given that the court went on, only a few paragraphs later, to distinguish Slater Steel on its facts, the better reading is that this careful language foreshadowed the court’s concerns with the soundness of that decision.
The only other Canadian case cited by Grenier in support of the rule against contributory infringement was the decision of the Federal Court Trial Division in Saunders v Airglide Deflectors Ltd,
(1980), 50 CPR (2d) 6,  FCJ No 609 (FCTD) [Airglide]. but it is also weak authority for this proposition, in part simply because the relevant facts are not clear, as Grenier acknowledged.
Grenier, supra note 16 remarked at 32, that Airglide “is difficult to comment upon since insufficient facts are disclosed in the reasons with respect to inducement.” With that said, it is not unlikely that the product supplied was especially designed for use in the invention.
The invention was a wind baffle for tractor-trailer vehicles, placed at a specific position so as to divert the airflow. The product supplied was the wind baffle itself. It is not clear either from the decision, or the patent at issue (“Apparatus for Reducing Linear and Lateral Wind Resistance in a Tractor-Trailer Combination Vehicle”, Can Patent No 809545 (1 April 1969)), whether the baffle was known, and only the position was new, or whether the baffle itself was specially designed for the invention. Even if we assume that to be the case, Airglide remains weak authority, because it was pleaded that several of the defendants, who were not held liable, had provided instructions as to how to mount the deflector for use in the patented combination.
Airglide, supra note 123 at 27–28. Consequently, as with Slater Steel, it is difficult to reconcile this decision with Windsurfing and MacLennan, along with the many other decisions holding that the supply of a product accompanied by instructions to infringe constitutes inducement.
See the cases cited supra note 96.
This leaves Permacon as the clearest example of a Canadian case actually refusing to impose liability for the supply of an unpatented product with no use other than in the invention,
Permacon, supra note 112. The patent was for a retaining wall composed of concrete blocks, “with convex dihedral protrusions,” which would allow the wall to be set up by stacking the blocks without mortar (at 381). While the decision does not say expressly that the blocks were especially adapted for use in the invention, that certainly appears to have been the case. with American Arch, neglected by both MacLennan and Grenier, to the contrary.
American Arch, supra note 69. In Slater Steel, supra note 29 at 82, Jackett P cited American Arch without discussing it, saying, “In the view I take of the matter, for the purposes of this case, no good purpose will be served by summarizing the other Canadian cases referred to by the plaintiffs.” The other Canadian case cited, but not discussed, was Gillette Safety Razor Company of Canada, Ltd v Mailman,  Ex CR 54, 1932 CarswellNat 7, rev’d on other grounds  SCR 724,  1 DLR 8. Although various leading English cases on indirect infringement were discussed in Gillette, it was ultimately decided on the basis of direct infringement, as the part supplied by the defendant (a razor blade of a particular design) was separately claimed.
In any event, the point here is not to count up trial decisions one way or the other but to illustrate that Canadian courts have not consistently held that there can be no liability for the supply of a product with no non-infringing use, absent inducement. There are only two trial decisions on point (or three if we include Airglide) and they are not consistent. There are no appellate decisions at all actually refusing to impose liability for the supply of a product especially adapted for use in a patented invention. The view that Canadian law has not traditionally recognized liability for contributory infringement appears to rest primarily on a misreading of Copeland-Chatterson.
The previous section of this article was concerned with the details of the cases. It is now time to step back and take a broader view. In Monsanto, the Supreme Court stated that the central question regarding infringement is, “did the defendant, by his acts or conduct, deprive the inventor, in whole or in part, directly or indirectly, of the advantage of the patented invention?”
Monsanto, supra note 7 at para 44, quoting Saccharin, supra note 7. It is now very well established that the answer to that question is “yes” when the indirect party has induced infringement. We can ask the same question in respect of contributory infringement: if the defendant knowingly supplies a product especially adapted for use in infringement of the patent, with no significant non-infringing use, and that product is in fact used to infringe, did the defendant thereby deprive the inventor, “in whole or in part, directly or indirectly, of the advantage of the patented invention”? Surely the answer to this question is “yes.” This is not to say that Monsanto changed the law regarding indirect infringement so as to recognize contributory infringement.
See Apotex Inc v Nycomed Canada Inc, 2011 FC 1441 at para 27, aff’d 2012 FCA 195 at para 3, noting that Monsanto, supra note 7, was not concerned with indirect infringement in any form, and so should not be taken to have changed the law in that respect. On the contrary, this general principle has long been recognized.
The primary authority cited in Monsanto, supra note 7, in support of this proposition is Saccharin, supra note 7, which predates Copeland-Chatterson, supra note 16; see also Beloit Canada Ltd v Valmet-Dominion Inc,  3 FC 497, 73 CPR (3d) 321 at 334–35 (FCA) [VDI]; and Sykes v Howarth, supra note 31 at 773 (referring to the prohibition in the letters patent on “either directly or indirectly” practising the invention as a justification for imposing liability for indirect infringement). Burbidge J had framed the fundamental question in essentially the same way in Copeland-Chatterson, when he pointed out that the grant of a patent gives the patentee the exclusive right to make, sell, and use the invention, and then stated that “it does not appear to me to be going too far to hold that any invasion or violation of that right is an infringement of the patent.”
Copeland-Chatterson, supra note 16 at 246–47.
As discussed above, MacLennan expressly stated a rule imposing liability solely based on the supply of a product with no non-infringing use. Nonetheless, the court did impose liability, albeit on the basis of inducement, and we might well wonder what the outcome would have been if the defendant had supplied the parts without an equivalency table, perhaps relying on the expertise of the purchasers to select the right tooth, or if there was only one type of tooth so that no table was needed. It is hardly clear that omitting an express statement of equivalency would be enough to exculpate the defendant.
The question will come to a head if the courts are faced with a product especially adapted for use in the patented invention, with no instructions or other acts that would be a hook to constitute inducement. For example, in Genpharm Inc v Procter & Gamble Pharmaceuticals Canada, Inc, the patent related to a new use of a known medicine. It had been sold in 200 mg pills for the old use, but the new use required a particular intermittent dosage regimen of 400 mg pills. To facilitate the necessary dosage regimen, the patentee sold the medicine in a kit containing blister packages of 400 mg pills for a two-week cycle with placebo pills where necessary. The generic proposed to sell its product with the same 400 mg pills in the same 14-pill blister pack. In addition, the product monograph implicitly instructed the infringing use by using a bioavailability comparison with the new use rather than the old use.
Genpharm, supra note 1 at para 31; and see the summary in Pharmascience, supra note 1 at paras 40–41. This constitutes indirect infringement,
As discussed supra note 20, Genpharm was overruled on an important point of law, but see Pharmascience, supra note 1 at paras 40–41, 59 holding that Genpharm was nonetheless correct in the result. and it is consistent with an inducement theory on the basis that the product monograph constitutes instructions to infringe. What would the result be on the same facts, absent the reference to the new use in the product monograph? Would a court hold that there is no indirect infringement despite the sale of the product in a blister pack especially adapted for the infringing use?
As another example, suppose the patent covers a paper-making machine having its rolls in contact with each other, and the defendant makes a machine with the rolls separated, but with an adjustment that has no other purpose than to enable the user to bring the rolls together, thus infringing. Would a court hold that this did not constitute indirect infringement if the machine were sold without instructions as to how to make the adjustment, in reliance on the expertise of the purchasers to recognize the possibility?
This example is loosely based on Knight v Gavit, 14 F Cas 765 (Cir Ct Pa 1846), cited by Osborne, supra note 3 at 77.
The answer, I suggest, is to be found in Windsurfing. It might be argued that Windsurfing can be explained on the basis of inducement theory, on the view that the rudimentary assembly instructions constituted the specific inducement to infringe; but it is quite untenable to suppose that the defendant in Windsurfing would have escaped liability had it omitted the instruction sheet. As discussed above, the reasoning of the court clearly turned directly on the supply of components that were especially adapted to infringe. Accordingly, in its summary of the law in Valmet, the Court of Appeal did not consider Windsurfing as turning on inducement, but rather cited it as support for a separate branch of indirect infringement that arises when the vendor “sells all the components of the invention to a purchaser in order that they be assembled by him.”
Valmet, supra note 20 at 14; see also VDI, supra note 132 at 335 (quoting this passage); and Varco, supra note 96 at para 260 (citing Valmet and Windsurfing for the proposition “that simply separating the parts of a device and shipping the parts for later integration and assembly does not avoid the liability for infringement”). Windsurfing does certainly stand for that proposition, but that is a summary of the facts, not a statement of principle.
What, then, is the principle?
One might say that the principle is indeed confined to a kit, on the view that selling a kit is in substance the same thing as selling the entire combination.
See United Telephone Co v Dale (1884), 25 ChD 778 at 782–83, quoted by Vaughan Williams LJ in Dunlop v Moseley, supra note 58 at 280. That is not satisfactory. Recall that indirect infringement is, in principle, not founded on any different basis from direct infringement. The question is always about what might “with propriety” be termed infringement, so it is always true that an established act of the indirect infringement is considered tantamount to direct infringement.
The better view is that Windsurfing stands for the proposition that the supply of the product can itself constitute the necessary inducement. The court dismissed the argument that evidence of direct infringement was required as “specious”:
To suggest that a person purchasing components, the only known use for which are for assembling to provide the purchaser with what he obviously desires — a sailboard — has not been persuaded to [assemble the parts] by the holding out of the desired result by both the manufacturer and the vendor thereof, stretches credulity to its limits.Windsurfing, supra note 89 at 265.
This is a clear statement that the supply of the kit itself induces the direct infringement. Exactly the same logic applies when what is supplied is a part with no non-infringing use: all that is needed is to change the words “components” to “a component,” and the reasoning would remain the same. If indirect infringement is a question of what might properly be termed the cause of the infringement, it is difficult to see any difference between the supply of a kit and the supply of a part especially adapted to infringe. This contrasts with a product with substantial non-infringing uses, where supply of the product alone, without any additional instructions, might never result in infringement.
The paragraph quoted above continued with the core rationale for imposing liability, which bears repeating:
I think it beyond dispute that the only inference to be drawn from the voluminous evidence in this case is that the respondent knew and intended that the ultimate purchaser would utilize the sailboard parts for the assembly of a usable sailboard which, upon assembly, would infringe the appellants' patent. It thereby became a party to such infringement, in my view.Ibid at 266. Infringement was not addressed at trial because the patent was held to be invalid. The finding of invalidity was reversed on appeal, and the Court of Appeal chose to address infringement itself.
The general principle stated in this paragraph, then, is that a party who supplies a product, the only known use for which is in the patented invention, knowing and intending that it be so used, thereby becomes a party to the infringement.
In summary, the holding in Windsurfing that the defendant was liable for indirect infringement did not, on its face, turn on the supply of the rudimentary instructions. And while it did turn on the supply of a kit, the principle expressed cannot be confined to a kit. The most straightforward reading of Windsurfing is that the supply of a product with no non-infringing use can itself constitute inducement sufficient to establish indirect infringement. This proposition is clearly correct in principle and is consistent with the basic test for infringement set out in Monsanto. Whether the indirect party supplies a kit of all the components, especially adapted to be assembled by the purchaser into the patented invention, or a single component, especially adapted for use in the invention, which the purchaser uses to complete the patented invention, the indirect party is depriving the patentee of the advantage of the invention. This principle also makes perfect sense from a policy perspective, as there is no risk of overbroad enforcement when the product supplied is especially adapted to infringe.
The only difficulty arises in reconciling this analysis with the statement in MacLennan of a rule against contributory infringement, and in particular the statement that “the sale of a component of a patented combination, even if the component has no use other than working the patented combination, is not sufficient to establish infringement by inducement.”
MacLennan, supra note 3 at para 40. In making this statement, the court in MacLennan relied on what was, with respect, a misinterpretation of Copeland-Chatterson as adopting British law into Canada.
Ibid at para 38. In fact, Copeland-Chatterson is entirely consistent with basing inducement on the supply of a product especially adapted to infringe, as American Arch illustrates.
Moreover, the court in MacLennan did not discuss Windsurfing at all on this point.
The only reference to Windsurfing in MacLennan was ibid at para 49, noting that Windsurfing had also distinguished Slater Steel on its facts. Apart from that statement in MacLennan, the two cases are entirely consistent and indeed harmonious; both held on the facts that a defendant who had supplied a product especially adapted to infringe, along with rudimentary instructions for use in an infringing manner, was liable as an infringer, and neither evinced any sympathy for a party who supplied a product especially adapted to infringe, nor suggested any reason in policy why liability should not be imposed.
Canadian law does not formally recognize liability for the supply of a product that has no purpose other than for use in a patented invention, in the absence of some form of inducement. In MacLennan, the court remarked that this rule “may seem questionable at first glance.”
MacLennan, supra note 3 at para 34. At second glance, the rule seems clearly sound when the product is a commercial staple, and defensible when it is not a staple but is nonetheless capable of substantial non-infringing use; but the rule remains questionable from a policy perspective when the product has no non-infringing use. The fundamental argument against imposing liability, as pointed out by the court in AB Hassle, is a concern for overbroad enforcement, and the desire to ensure that the public is not prevented from using the product for unpatented uses. This concern is absent when the product has no other use.
And indeed, Canadian courts have never articulated a policy rationale for the rule that selling a product with knowledge that it will be used to infringe does not constitute infringement, even if the part has no non-infringing use. The rule has become established largely on the basis that Copeland-Chatterson accepted the British cases into Canadian law—a view that does not withstand a careful reading of that decision. There are very few cases that actually apply the rule, and none at the appellate level. While the Court of Appeal in MacLennan held that the supply of a product especially adapted to infringe is not sufficient to establish inducement, this proposition was strictly obiter, the cases it cited do not establish that rule as a matter of Canadian law, and, most important, that statement is at odds with the holding in Windsurfing.
Windsurfing was the first appellate decision in Canada to deal with liability for the supply of a product especially adapted to infringe. The central holding is that such supply constituted the inducement necessary to impose liability for indirect infringement. This holding in Windsurfing is sound both as a matter of the policy concern relating to overbroad enforcement, and the long-standing legal principle that the patentee should not be deprived, directly or indirectly, of the advantage of the patented invention.
When a court is squarely faced with a case in which the defendant has supplied a product with no substantial non-infringing use, knowing and intending that it be used in the patented invention, and the product is in fact so used, the court should hold, on the authority of Windsurfing, that the supply of a product especially adapted to infringe is, in itself, sufficient inducement to ground liability for indirect infringement.