“Yes, the law is about words…,” says Ben Chiriboga in writing about the essential skills that lawyers must have. And Ken White noted more recently that “the entire project of the law is about words meaning specific things.” But our problem often is that the law, or lawyers, frequently use unfamiliar or exotic terms that others claim have no more understood meaning than a reference to a “vermicious knid,” and those or other lawyers may overuse a word that they do not seem to actually comprehend. Indeed, understanding and enforcing the rule of law is itself commonly about defining an undefined concept.
Intellectual property law is all the more a refuge for wordsmiths, linguists, and logophiles, as three recent IP-related matters can help illustrate. One is the case of Abitron Austria GMBH v. Hetronic International Inc., recently argued before the United States Supreme Court, and which presented the question of whether one could understand the language of the Lanham Act to apply outside the United States in some manner, i.e. does it have extraterritoriality and how do we define “commerce”? The second is Finjan LLC v. Eset, LLC, a Federal Circuit decision on which Eset is now seeking Supreme Court review; Finjan raises the question of whether a court must accept a patentee’s own expressly-defined claim terms when reviewing a patent. Finally, a third intriguing development is the US Copyright Office’s March 16, 2023 guidance on works containing material generated by artificial intelligence, which raises the question of what words like “author” and phrases like “human creativity” mean.
Let’s look at them in turn.
I briefly mentioned Abitron here recently, but it deserves more attention in the context of defining the boundaries of US trademark laws and just on the notion of defining words. The question presented in that case is “Whether the U.S. Court of Appeals for the 10th Circuit erred in applying the Lanham Act, which provides civil remedies for infringement of U.S. trademarks, extraterritorially to Abitron Austria GmbH’s foreign sales, including purely foreign sales that never reached the United States or confused U.S. consumers,” as noted at scotusblog.com. In brief, as reported by Reuters, Hetronic, an Oklahoma company, makes remote-control systems for cranes and other industrial machinery. Hetronic sued Abitron, which had purchased Hetronic Germany, claiming that Abitron had violated Hetronic’s trademark rights by making and selling Hetronic-branded products with unauthorized parts. Hetronic prevailed at trial to the tune of $114 million in damages, $90 million of which were awarded on the trademark claims. Those damages, however, were largely based on the defendant’s sales of the offending products outside the United States. The trial court and the 10th Circuit Court of Appeals held the damages recoverable because, despite the fact that the sales had occurred outside the United States, Abitron’s infringing activity had a substantial effect on U.S. commerce and “diverted tens of millions of dollars of foreign sales from Hetronic that otherwise would have ultimately flowed into the United States,” as the appellate noted (at Hetronic Int’l, Inc. v. Hetronic Germany GmbH, 10 F.4th 1016, 1046 (10th Cir. 2021).
What the litigants and their counsel battled over was where the boundaries and lines should be drawn under the Lanham Act. The Lanham Act governs federal trademark and unfair competition disputes. It subjects to liability “[a]ny person who shall … use in commerce any … colorable imitation of a registered mark,” 15 U.S.C. § 1114(1) (Section 32), or “[a]ny person who … uses in commerce any” word, false description, or false designation of origin that “is likely to cause confusion… or to deceive as to the affiliation,” origin, or sponsorship of any goods, id. § 1125(a)(1) (Section 43). Notably, the Act defines commerce broadly as “all commerce which may lawfully be regulated by Congress,” id. § 1127, and affords federal courts jurisdiction over all claims arising under it, id. § 1121(a). Though most of the damages the jury awarded to Hetronic flowed from Defendants’ Lanham Act violations, Hetronic argued as a threshold matter that the Act does not apply extraterritorially to their foreign conduct, an argument that they lost in both lower courts.
Central to arguing the issues in Abitron was the notion of definition, which came up at least fifteen different times at the oral argument held before the Supreme Court on March 21, 2023. What did that statutory language mean?
Right off the bat at the Supreme Court, Abitron’s counsel noted that “Hetronic International offers only the text definition of commerce, and this Court has repeatedly rejected the notion that commerce language is enough to extend the law to foreign conduct, even if that language otherwise invokes the full scope of the constitutional commerce power.” Argument Transcript (page 4, lines 4-11). That single sentence illustrates lawyers’ difficult relationship with language, as it couples a concession with what the word commerce means with an argument that the word cannot be understood that way in that context, a position later returned to on rebuttal when counsel argued that “commerce” here must mean “domestic commerce” (page 91, line 13 to 92, line 15), a follow up to the point made in the cert. petition (at 34-35) that the Court has “‘repeatedly held that even statutes that contain broad language in their definitions of ‘commerce’ that expressly refer to ‘foreign commerce’ do not apply abroad.’” And therefore “a fortiori, a statute whose definition of ‘commerce’ at most implicitly could include foreign commerce cannot apply abroad either.”
But there are counterarguments to the suggestion that the law would only have been concerned with uses in domestic commerce. That is because, as the Government’s amicus brief points out (13-14), the Lanham Act permits recovery against any person who “use[s] in commerce” the owner’s mark where such use “is likely to cause confusion, or to cause mistake, or to deceive.” 15 U.S.C. 1114(1)(a), 1125(a)(1). So, while the effect of confusion would need to be felt domestically, the cause could be anywhere. As the Government noted (at 14), “To the extent that petitioners’ sales created a likelihood of consumer confusion in the United States, Congress’s purposes therefore are squarely implicated, even though petitioners’ own conduct occurred abroad.” The INTA amicus brief (at page 4) summarizes this position well, noting:
The Lanham Act may be applied to stop foreign conduct that causes substantial impact in the United States within the focus of the Lanham Act, especially where the defendant is a U.S. citizen. Steele v. Bulova Watch Co., 344 U.S. 280, 285-86 (1952). But this does not mean that the Lanham Act has unlimited worldwide application against foreign actors who are not U.S. citizens. Indeed, the Act reaches foreign conduct only to the extent it causes substantial injury in the U.S. Such foreign conduct that causes substantial impact in the U.S. exists in the boundary between “extraterritorial” and “domestic” application of the law and should be reachable whether or not the Court holds the Lanham Act to have extraterritorial application.
Other amici like Intellectual Property Owners Association (IPO) and American Intellectual Property Law Association (AIPLA) also pushed similar views.
In the end though, the test the court applies is less one of defining terms to be applied as law than of determining causation as to any of the impacts felt in fact, always a more case-specific inquiry, as the ABA’s amicus brief noted (at 15). As a long-ago law review note said, “Any legal definition of causation must raise the question whether all the requirements of the definition are found to be present in the particular case. No sound principle can be laid down by which the judge can always determine this question without infringing on the province of the jury.” Note, Legal Cause In Actions of Tort, 25 Harv L. Rev. 303, 305 (1912). It will be interesting to see where the Supreme Court defines these boundaries and draws the lines in the trademark context.
As we wait to see how the Supreme Court will define the reach of the Lanham Act, we also review what it will do in Finjan LLC v. Eset, LLC, a patent case. As one might imagine, patent law frequently comes down to what particular terms in a patent mean, and because the whole point of a patent is to describe a new invention, existing language may sometimes now fully capture what an invention, or element thereof, really is. Consequently, there is a long tradition in patent law of letting patentees and inventors act as their own lexicographers, as the saying goes:
Although words in a claim are generally given their ordinary and customary meaning, a patentee may choose to be his own lexicographer and use terms in a manner other than their ordinary meaning, as long as the special definition of the term is clearly stated in the patent specification or file history.
[Vitronics Corp. v. Conceptronic, Inc., 90 F.3d 1576, 1582 (Fed. Cir. 1996)]
[T]he claim term will not receive its ordinary meaning if the patentee acted as his own lexicographer and clearly set forth a definition of the disputed claim term in either the specification or prosecution history.
[CCS Fitness, Inc. v. Brunswick Corp., 288 F.3d 1359, 1366 (Fed. Cir. 2002)]
What the saying really means is that a patent applicant can give a word, term, or phrase its own definition, and that definition should be applied to that application or later patent.
The problem in Finjan was two-fold. First, the patent applicant used inconsistent definitions of the term “downloadable.” That in and of itself may have rendered the patent indefinite and therefore invalid under US statutory law, which was the ruling of the district court. But the Federal Circuit on appeal created the second problem, at least as the petitioner sees it. The Federal Circuit provided its own definition for the term “Downloadable,” vacated the indefiniteness finding, and remanded to the district court for consideration of the patent in light of the definition of “Downloadable” that the applicant had never considered or argued. Telling a patent applicant that her invention, or a major element of it, does not fit the applicant’s own definition of the invention or such elements sows confusion among many patent lawyers in the US.
But the real problem here is not with the definition of “Downloadable” by the Circuit court, but failure of the petitioner to understand what being a lexicographer means in this context. Like Samuel Johnson, Noah Webster, James Murray or W.C. Minor, one must earn the title or description lexicographer.As Simon Winchester, author of The Professor and the Madman: A Tale of Murder, Insanity and the Making of the Oxford English Dictionary (1998), pointed out, part of that step is the “end to timidity – the replacement of the philologically tentative by the lexicographically decisive.” That simply never happened in Finjan, as there were multiple and inconsistent cross-references and no specific, concluded definition. Thus nothing decisive, and certainly not, in the language of Vitronics or CCS nothing clearly stated nor clearly set forth. Consequently, it does not seem to violate precedent nor ignore an applicant’s would-be status as lexicographer for the Federal Circuit to have come up with its own definition here. The lexicographic enterprise in Finjan just had not happened yet, as there was certainly no OED or even draft, just, as seen in the book and movie version of the Murray/Minor story, some conflicting and separate slips of paper not yet acted decisively upon.
Finally, let us examine the Copyright Office Guidance referenced earlier. The purpose of that guidance was “to clarify its practices for examining and registering works that contain material generated by the use of artificial intelligence technology,” and the guidance reiterated the “well-established [principles] that copyright can protect only material that is the product of human creativity” and that, “fundamentally, the term “author,” which is used in both the Constitution and the Copyright Act, excludes non-humans.” These are principles that I have commented on her before as far back as 2016 and July 2021 (each on the human author vel non question, and I would note that the issue of whether animals have rights of their own remains one others are bringing to court up to this day, as recently noted here). More recently, such as last June and just this February, this poster blogged here on AI issues and commented elsewhere on artificial intelligence generally and specific to this new guidance.
Central to the guidance is the concept that US law presently requires “authorship” to include human touch or contribution, and works that lack human touch will not qualify. The Copyright Office reaches that conclusion in part because, as the guidance states “the Copyright Act refers to an author’s ‘children,’ ‘widow,’ ‘grandchildren,’ and ‘widower,’—terms that ‘all imply humanity and necessarily exclude animals.” But is that so?
It seems to me that one sees in popular culture references to the “children” and “grandchildren” or at least the “next generation” of non-humans all the time. For instance, the Encyclopedia Britannica entry on the racehorse Sea Biscuit describes him as “Man o’ War’s grandson,” thereby impliedly also labeling Man o’ War as a grandfather. Likewise, Elsie the Cow of Borden’s fame was described in the media as being “with child” before she gave birth to a calf. And gorillas at the London Zoo have been described in the press as both “widowed” and “husbands.” All of these instances, and I am sure there are more, suggest that the US Copyright Office may have too quickly found that such statutory references must automatically be understood to refer only to humans. Just as importantly, or maybe even more importantly, courts have used those words in reference to non-humans, as when the Supreme Court of Minnesota referred to “Traffic Secretary, great-grandson of the famous racehorse Secretariat, “
Further, how much human touch is enough touch is a complicated question, and may be even more difficult a question in mediums like art and music, as David Israelite points out in a well-done article. While noting that “AI is already pushing the boundaries,” he also states that “it’s extremely unclear how these lines will be drawn when the human and AI contributions are more intertwined.” So, copyright cartographers and surveyors will remain somewhat unguided in defining those boundaries.
Not fully at sea though. As the Copyright Clearing House has noted and I have mentioned, the Copyright Office is engaged and listening. The Copyright Office’s AI initiative, “has its own dedicated site, will include public listening sessions in April and May, as well as future notices of inquiry,” as CCH notes. So “[w]e’ll all have to stay tuned to see how the initiative develops and to see what other guidance the Office might release,” according to CCH. Sure, “[i]n the eyes of the Copyright Office, the human hand has the upper hand for now, which should be a relief for practicing artists concerned about AI encroaching on the commissions and freelance market,” as one commentator noted. But, as I noted before, “I think the Kraken comes now—help us,”
It also makes sense that AI work should be copyrightable. That does not mean that a copyright should register in the name of the AI program any more than it should in the name of the camera. The economic realities present suggest that some protected status, whether in copyright itself or in some other form makes sense. I draw the analogy to what happened in the patent field where many long debated whether plant patents were obtainable since the novelty was not created by an inventor but, many would argue, by nature. But giving periods of exclusivity to those who wished to do painstaking, long work of plant cross-breeding and genetically modifying organisms was a necessary incentive to spur investment, as Wen Zhou has discussed. Just as “Congress passed the Plant Patent Act in 1930 as a result of plant breeding and other agricultural efforts, making new plant strains derived from crossbreeding patentable,” Id., some modification of, or supplement to, the Copyright Act to make some level of AI-generated images and texts protected seems an economic inevitability. Exactly where that boundary and those borders will be drawn is unclear.
But, as suggested in our title, defining boundaries is a frequent task in the law, and we must not shrink from defining boundaries in a way that expands our minds and our possibilities.