©2020. Published in Landslide, Vol. 13, No. 1, September/October 2020, by the American Bar Association. Reproduced with permission. All rights reserved. This information or any portion thereof may not be copied or disseminated in any form or by any means or stored in an electronic database or retrieval system without the express written consent of the American Bar Association or the copyright holder.
The fundamental purpose of our patent laws is to promote innovation by granting inventors a period of exclusivity for their patented inventions. A corollary to that purpose is encouraging patent holders to provide notice of their patent rights to the interested public so that infringement may be avoided.
In an infringement suit, a patent holder can ordinarily seek damages for infringement that occurred as much as six years before the suit was filed. However, in order to recover such presuit damages, the patent holder must prove either (1) it provided the public with constructive notice of its patent rights by marking its patented products with the patent number, or (2) “the infringer was notified of the infringement and continued to infringe thereafter.”1 The Federal Circuit Court of Appeals, which has exclusive jurisdiction over patent appeals, interpreted this seemingly broad “actual notice” provision as requiring that a patent holder provide an “affirmative communication of a specific charge of infringement by a specific accused product or device,” regardless of how the accused infringer may have interpreted a communication about potential infringement.2 The Federal Circuit’s test is significantly more specific than the express language of the statute. It has been criticized as leading to unfair results in certain circumstances, effectively shielding even a willful infringer from liability for presuit damages if a patent holder’s communication to the infringer did not use the right words.
To illustrate the problem with the Federal Circuit’s test, consider the following fact pattern drawn from an actual case: A manufacturer informs a customer that its planned purchase and use of a competitor’s product may “implicate” numerous patents owned by the manufacturer, and hands the customer a presentation that identifies the competitor’s product by name and provides a schematic diagram with patent numbers. Concerned, the customer immediately contacts the competitor, gives the competitor the presentation, and asks what the competitor is going to do about the issue. The competitor promptly engages outside counsel to prepare noninfringement opinions on the identified patents. The competitor’s in-house counsel also confirms, in an internal email, that he interprets the manufacturer’s communication to the customer as an accusation of infringement. On similar facts, applying the Federal Circuit’s test, a district court judge ruled—on summary judgment, no less—that the manufacturer’s communication did not constitute actual notice of infringement, thus absolving the competitor of liability for potentially tens of millions of dollars in presuit damages.3
In recent years, the U.S. Supreme Court has shown an arguably unprecedented interest in patent cases. The type of issue that has seemed to grab the Court’s attention more than any other is the Federal Circuit’s imposition of overly rigid legal standards that exceed the express requirements of the Patent Act4 and unduly restrict trial courts’ discretion in deciding legal issues. The Federal Circuit’s test for determining whether a patent holder satisfied the notice requirement for presuit damages appears to present just such an issue and thus may be ripe for Supreme Court review and reversal.
Actual Notice Requirement for Presuit Damages
Section 287 of the Patent Act encourages patent holders (patentees) to help the public avoid innocent infringement by generally conditioning the ability to collect damages for presuit infringement on providing prior notice of patent rights.5 In particular, patent holders may provide so-called “constructive notice” of their patent rights to the general public by marking their patented products with the relevant patent numbers.6 The statute goes on to say that if a patent holder fails to comply with the marking requirement, presuit damages are forfeited “except on proof that the infringer was notified of the infringement and continued to infringe thereafter, in which event damages may be recovered only for infringement occurring after such notice.”7 This is known as “actual notice.” Importantly, if the patent holder never sold products covered by the patent, or if the patented invention is a process or method for doing something such that there is no “product” to mark, then presuit damages may be recovered regardless of whether any prior notice (actual or constructive) was provided to the infringer.8
Notice of infringement is often a critically important issue in patent litigation, with very significant practical consequences. The Patent Act generally allows a patent owner to collect damages for infringement occurring as much as six years before a suit was filed.9 Forfeiting damages for some or all of that presuit period can dramatically reduce the value of a case for the patentee, or conversely, dramatically reduce the potential exposure for an accused infringer.
The constructive/actual notice regime has been part of U.S. patent law since the enactment of the Patent Act of 1861, and the statutory language governing actual notice has remained largely unchanged in the 150 years since. More than a century ago, the Supreme Court summarized the notice requirements in Dunlap v. Schofield:
The clear meaning of this section is that the patentee or his assignee, if he makes or sells the article patented, cannot recover damages against infringers of the patent, unless he has given notice of his right, either to the whole public by marking his article “patented,” or to the particular defendants by informing them of his patent and of their infringement of it.10
However, the Court did not place any restrictions on how a patentee must meet the actual notice requirement, nor does the legislative history provide any guidance. Until fairly recently, all the law demanded was “proof that the infringer was notified of the infringement.”
Development of the Federal Circuit’s Test
Patent litigation today is relatively mainstream, due in no small part to the advent of the “patent troll” phenomenon. But as recently as the early 1980s, patent litigation was viewed by many as an esoteric endeavor, handled only by highly specialized attorneys with engineering degrees. The results in any given case were unpredictable, and forum shopping was rampant as plaintiffs looked for supposedly “patent friendly” jurisdictions in which to sue.11 It was against this backdrop that Congress created the Federal Circuit in 1982, giving the new court exclusive jurisdiction over appeals in patent cases.12 Consistent with its genesis, the Federal Circuit set about trying to bring a measure of order and increased predictability to patent litigation, largely through the adoption of better-defined legal standards—and in some cases bright-line rules—for district courts to follow when deciding legal issues in patent cases.13 The actual notice requirement for presuit damages is one of many issues the Federal Circuit has tackled over the years.
The Federal Circuit’s test for what constitutes actual notice under § 287 dates back to a 1994 decision, Amsted v. Buckeye Steel, which involved an appeal from a jury verdict that Buckeye had willfully infringed Amsted’s patent and owed nearly $1.5 million in compensatory damages, including damages for presuit infringement.14 The evidence at trial showed that Amsted had sent a letter to Buckeye five years before filing suit that identified the asserted patent and warned Buckeye to “refrain from supplying or offering to supply component parts which would infringe or contribute to the infringement of the patent.”15 The evidence further showed not only that Buckeye knew it was infringing Amsted’s patent at the time it received the letter, but also that Buckeye had interpreted the letter as a notice of infringement. Based on those facts, the district court agreed with the jury’s finding that the letter constituted actual notice of infringement.16
On appeal, the Federal Circuit reversed the decision as to satisfaction of the actual notice requirement. The appellate court explained that “[a]ctual notice requires the affirmative communication of a specific charge of infringement by a specific accused product or device,” so Amsted merely warning Buckeye to refrain from infringing the patent was insufficient.17 The court further instructed that “[i]t is irrelevant, contrary to the district court’s conclusion, whether the defendant knew of the patent or knew of his own infringement”; instead, the inquiry “must focus on the action of the patentee, not the knowledge or understanding of the infringer.”18 Thus, although Buckeye had willfully infringed Amsted’s patent and had interpreted Amsted’s letter as a notice of infringement, the Federal Circuit held that Buckeye was not “notified of the infringement” within the meaning of § 287.
Three years later, in SRI International v. Advanced Technology Laboratories, the Federal Circuit refined its test to clarify that actual notice does not require an “unqualified charge of infringement,” explaining that “[i]t is not controlling whether the patentee threatens suit, demands cessation of infringement, or offers a license under the patent.”19 Instead, the actual notice requirement is satisfied “when the recipient is informed of the identity of the patent and the activity that is believed to be an infringement, accompanied by a proposal to abate the infringement, whether by license or otherwise.”20
More recently, in Arctic Cat v. Bombardier, the Federal Circuit was asked to address the apparent unfairness of a knowing and willful infringer avoiding liability for presuit damages because the patent owner had not provided adequate notice of infringement.21 In response, the appellate court reiterated that “willfulness, as an indication that an infringer knew of a patent and of its infringement, does not serve as actual notice as contemplated by § 287.”22 But in what could be viewed as a retreat from the seemingly extreme view expressed in Amsted—that the infringer’s state of mind is irrelevant—the court advised that “[k]nowledge by the infringer is not enough. Actual notice under § 287 requires performance by the patentee.”23 The court thus appeared to suggest that an accused infringer’s interpretation of a communication by a patent holder could be relevant to the issue of whether that communication constituted actual notice of infringement. Nevertheless, the decision confirmed the continuing vitality of the specific requirements for actual notice articulated in Amsted.24
Supreme Court’s Hostility to Overly Rigid Patent Law Standards
While many have welcomed the Federal Circuit’s efforts to improve the consistency and predictability of lower court decisions in patent cases through the adoption of better-defined legal standards, the U.S. Supreme Court has seemingly made a concerted effort to strike down bright-line rules adopted by the Federal Circuit, particularly where such rules lack solid grounding in statutory language. Indeed, the Supreme Court has reversed a surprising number of Federal Circuit decisions in recent years. Those reversals were often based on the rejection of legal standards that the Supreme Court perceived to be overly rigid, and which unnecessarily (and inappropriately) restricted the scope of discretion available to district court judges.25
A common thread running through these decisions is the Supreme Court’s apparent hostility to the Federal Circuit’s reliance on bright-line rules to resolve multifaceted, often nuanced issues. In the Supreme Court’s view, such issues are more appropriately resolved on a case-by-case basis, in view of the particular facts at hand. The Federal Circuit’s test for what constitutes actual notice within the meaning of § 287 appears to fit this bill.
Even though § 287 broadly requires only that an infringer be “notified of the infringement,” the Federal Circuit’s test requires an affirmative communication by the patentee. Such communication must identify the patent at issue and the activity believed to infringe, and include a proposal to abate the infringement through, for example, a demand to cease and desist, a license offer, or a threat of litigation. Given the contrast between the broad statutory language and the specific requirements imposed by the Federal Circuit, the actual notice test may well be in the Supreme Court’s crosshairs, and perhaps rightly so.
A More Flexible Approach to Actual Notice
Compliance with the actual notice requirement of § 287 appears to be the type of fact-intensive issue that lends itself well to a flexible, totality of the circumstances approach. Indeed, while the specific requirements reflected in the Federal Circuit’s test all seem to be reasonable factors for consideration, one can fairly question whether the statute supports treating any of them as strict requirements, including focusing solely on the actions of a patentee.
For instance, the Federal Circuit has consistently held that the provision of actual notice must be an affirmative act by the patentee. But critics of the Federal Circuit’s test have noted that this requirement appears to be based on a misreading of the Supreme Court’s century-old decision in Dunlap. The specific issue addressed in Dunlap was which party had the burden of proof regarding notice, including both marking and actual notice. In holding that the statute places the burden of proof on the plaintiff, the Supreme Court explained that “[e]ach [of marking and actual notice] is an affirmative fact, and is something to be done by [the patentee].”26 However, in promulgating the initial iteration of its actual notice test in Amsted, the Federal Circuit misquoted Dunlap as saying that notice “is an affirmative act, and something to be done by [the patentee].”27 From that time forward, it was largely taken as a rule that the notice to the infringer had to originate with the patentee, even though the plain language of the statute imposes no such requirement.
For purposes of justifying an award of presuit damages, the critical inquiry should be whether the infringer received notice of its alleged infringement, not who provided that notice. Of course, encouraging patent holders to send such notices themselves is sound policy, as it can aid the recipient in avoiding infringement through, for example, requesting a license. But an infringer should not be able to avoid liability for presuit damages merely because an otherwise sufficient notice of infringement originated with someone other than the patent holder.28
Much the same can be said for the Federal Circuit’s requirement for a specific charge of infringement. The court has tried to eliminate any perceived need for patent holders to threaten an infringement suit (and thereby risk subjecting themselves to a declaratory judgment action) by focusing on whether the communication included any “proposal to abate the infringement,” including a license offer. On its face, the statute imposes no requirements on what a patentee must tell an accused infringer about the potential consequences of continued infringement. Instead, the statute only requires that the infringer be “notified of the infringement.”29 If an alleged infringer continues to infringe after being so notified, it is reasonable to subject the infringer to the risk of having to pay damages for its past wrongdoing, regardless of whether the patentee had threatened suit, demanded that the infringement stop, offered a license, or simply stayed silent about abating the infringement.30
In arguable contrast to the approach taken by the Federal Circuit to date, a totality of the circumstances test could help avoid the inherent unfairness of a knowing infringer being relieved of liability for presuit damages merely because a communication by the patent holder was not sufficiently explicit regarding some aspect of what the Federal Circuit requires for “actual notice.” To be sure, § 287 requires proof that an infringer “was notified of the infringement.” An infringer’s subjective belief that it is infringing should not be sufficient in the absence of some communication by (or on behalf of) the patentee. However, if there was such a communication that the infringer interpreted as a notice of infringement, it is difficult to see why that subjective belief should not be relevant, if not actually controlling, on the issue of whether the infringer was “notified of the infringement.”31
Moreover, it should be possible to apply a more flexible approach to actual notice without detracting from the primary policy objectives of § 287, which include helping the public avoid innocent infringement.32 Patent holders will still be encouraged to give the public notice of their patent rights through marking in order to ensure the availability of damages for presuit infringement. If a patent holder chooses not to mark its patented products for any reason, it is still incentivized to warn others about possible infringement in order to avoid forfeiting presuit damages in the event litigation ultimately ensues. When litigation seems likely and potential damages are high, it would seem to be a rare patent holder indeed who would choose to risk forfeiture of presuit damages through a less than explicit infringement notice. But if, for whatever reason, a communication to an infringer lacks the hallmarks of an explicit notice, a totality of the circumstances test would give judges and juries a greater ability to find that a given communication provided the requisite notice, especially with regard to knowing infringements. In particular, when an infringer interprets an arguably ambiguous communication as a notice of infringement, the policy objectives of § 287 unquestionably have been served.
In terms of how a totality of the circumstances test should be applied, the Federal Circuit’s current test provides a good starting point for a list of factors to be considered in determining whether a given communication constituted actual notice under § 287. Relevant considerations should certainly include whether the communication (1) identified the patent(s) at issue, (2) identified an infringing product or described an activity alleged to infringe, and (3) proposed some form of abatement, including but not limited to a demand to stop the accused activity, an offer of a license, or an express threat of an infringement suit. However, if supported by the evidence, additional factors to consider should include how the recipient interpreted the communication and any actions the recipient may have taken in response to that communication. If the weight of the evidence, evaluated flexibly, supports a conclusion that the recipient was more likely than not “notified of the infringement,” then the patent holder should be entitled to seek presuit damages.
The Federal Circuit’s test for determining whether an infringer received actual notice of infringement appears to be exactly the type of rigid test that the Supreme Court has been prone to reject in recent years. When a patent holder fails to properly mark its patented products, § 287 of the Patent Act expressly requires only proof that the infringer was “notified of the infringement,” yet the Federal Circuit has interpreted that broad statutory language as imposing a number of specific requirements that appear to go beyond what Congress intended. Accordingly, the Federal Circuit’s actual notice test may well be in the Supreme Court’s crosshairs, just waiting for a case with the right set of facts to come along.
1. 35 U.S.C. § 287(a) (emphasis added).
2. Amsted Indus. Inc. v. Buckeye Steel Castings Co., 24 F.3d 178, 187 (Fed. Cir. 1994).
3. See Siemens Mobility Inc. v. Westinghouse Air Brake Techs. Corp., No. CV 16-284-LPS-CJB, 2018 WL 7893901, at *2–4 (D. Del. Dec. 17, 2018).
4. 35 U.S.C. §§ 1 et seq.
5. Id. § 287.
6. See SRI Int’l, Inc. v. Advanced Tech. Labs., Inc., 127 F.3d 1462, 1469 (Fed. Cir. 1997) (“Section 287(a) relates to informing the interested public of the patent status of an article in commerce, and permits either constructive notice by marking the article with the patent number, or actual notice to the infringer.”).
7. 35 U.S.C. § 287(a).
8. Arctic Cat Inc. v. Bombardier Recreational Prods. Inc., 950 F.3d 860, 863–64 (Fed. Cir. 2020).
9. 35 U.S.C. § 286.
10. 152 U.S. 244, 247–48 (1894).
11. See Richard Linn, The Future Role of the United States Court of Appeals for the Federal Circuit Now That It Has Turned 21, 53 Am. U. L. Rev. 731 (2004).
12. See id. at 732.
13. See id. at 732–33.
14. Amsted Indus. Inc. v. Buckeye Steel Castings Co., 24 F.3d 178 (Fed. Cir. 1994).
15. Id. at 186.
17. Id. at 187.
19. SRI Int’l, Inc. v. Advanced Tech. Labs., Inc., 127 F.3d 1462, 1470 (Fed. Cir. 1997).
21. Arctic Cat Inc. v. Bombardier Recreational Prods. Inc., 950 F.3d 860, 862–63 (Fed. Cir. 2020).
22. Id. at 866.
23. Id. at 866–67 (emphasis added).
24. Id. at 864.
25. One of the earliest examples is eBay v. MercExchange, in which the Supreme Court struck down the Federal Circuit’s rule that a prevailing patentee was entitled to a permanent injunction against future infringement absent exceptional circumstances. eBay Inc. v. MercExchange, L.L.C., 547 U.S. 388 (2006). Another example is Halo v. Pulse Electronics, where the Supreme Court rejected the Federal Circuit’s two-part test for willful infringement, which included both objective and subjective prongs, in favor of a more flexible, totality of the circumstances approach. Halo Elecs., Inc. v. Pulse Elecs., Inc., 136 S. Ct. 1923 (2016). Still more examples are Bilski v. Kappos and Mayo v. Prometheus, in which the Supreme Court rejected the Federal Circuit’s heavy reliance on a “machine-or-transformation test” for determining whether a patent claim is directed to patent-eligible subject matter. Bilski v. Kappos, 561 U.S. 593 (2010); Mayo Collaborative Servs. v. Prometheus Labs., Inc., 566 U.S. 66 (2012).
26. Dunlap v. Schofield, 152 U.S. 244, 248 (1894) (emphasis added).
27. Amsted Indus. Inc. v. Buckeye Steel Castings Co., 24 F.3d 178, 187 (Fed. Cir. 1994) (emphasis added).
28. See Lans v. Dig. Equip. Corp., 252 F.3d 1320, 1327 (Fed. Cir. 2001) (“[K]nowledge of the patentee’s identity facilitates avoidance of infringement with design changes, negotiations for licenses, and even early resolution of rights in a declaratory judgment proceeding.”).
29. 35 U.S.C. § 287(a).
30. See Gart v. Logitech, Inc., 254 F.3d 1334, 1345 (Fed. Cir. 2001) (explaining that “[t]he requirement of actual notice is designed to assure that the accused infringer knew of the adverse patent and the alleged infringement during the period in which its liability accrues” (citing SRI Int’l, Inc. v. Advanced Tech. Labs., Inc., 127 F.3d 1462, 1470 (Fed. Cir. 1997))).
31. A number of courts have applied this rationale in support of a finding that a given communication satisfied the actual notice requirement. See, e.g., Wokas v. Dresser Indus., Inc., 978 F. Supp. 839, 846 (N.D. Ind. 1997) (“In addition, it has been held that the notice requirement of § 287 is satisfied where the infringer acknowledges a specific communication to be a notice of infringement.”); Chubb Integrated Sys., Inc. v. Nat’l Bank of Wash., 658 F. Supp. 1043, 1051 (D.D.C. 1987) (“[W]hen one acknowledges . . . that the adversary is claiming infringement, the law most certainly does not compel the patent owner to repeat it more explicitly.”).
32. See Arctic Cat Inc. v. Bombardier Recreational Prods. Inc., 876 F.3d 1350, 1366 (citing Nike, Inc. v. Wal-Mart Stores, Inc., 138 F.3d 1437, 1443 (Fed. Cir. 1998)).