Deven Desai & Sandra Rierson, The Genericism Conundrum, forthcoming in the Cardozo Law Review: This interesting paper argues that only uses in a selling/purchasing context ought to count in the determination of whether a mark is generic: that is, only when competitors actually use the term to advertise their products, or when consumers actually ask sellers for “a coke” and mean any type of soda. Uses in newspapers like “google that question” or noncommercial conversation like “can I have a kleenex?” would simply not count as evidence of genericity, since they can equally evidence hybrid meaning: a commercial, brand-specific meaning used to make purchases and a noncommercial, conversational meaning used to communicate with others.
Though the paper doesn’t say so, given the types of evidence used to show genericity now, marks would rarely be found generic (Chocolate Fudge Soda-type marks would be the exception; in essence, the proposal restores the idea that some marks are so descriptive that they are the “common descriptive name,” but the slide of fanciful marks into genericity would be less likely). The benefit from the low-protectionist perspective is that trademark owners might stop harassing newspapers and ordinary users, and could abandon their “education” campaigns that try to restrict language by telling people never to use marks as nouns. This also ties into the “trademark use/use as a mark” debate: The authors argue that newspaper uses of trademarks as shorthand for a product class or an action associated with a product is not “use as a mark” because it is not a sales context, and therefore trademark owners have no legal grounds for their threatening letters anyway. As they point out, it is a dubious practice to consider newspaper and dictionary definitions in the genericity determination, say that trademark owners are responsible for policing such uses, and then give them no cause of action against newspaper/dictionary misuse. Since, however, such a cause of action would be a bad idea, the obvious solution is to no longer use such uses as evidence of genericity.
Among other points worthy of discussion: The paper makes several empirical claims in support of its proposals, but cites no evidence for them. For example, the authors state that, when trademark owners invest a lot of money to create secondary meaning, they usually succeed; therefore, a finding of genericity if the manufacturer guessed wrong about the status of the name can lead to a lot of actual (“de facto”) confusion and inefficiency. Given that most product launches fail, I’d love to see the evidence that spending money is enough to create secondary meaning.