The exclusions in §2 that arguably don’t go to core trademark policy (though this too is debatable) are those for immoral, scandalous, or disparaging marks; flags/coats of arms; names/signatures/portraits of living persons/deceased presidents with living spouses without written consent; geographic indications (GIs) on wine or spirits identifying someplace other than their origin; and primarily geographically deceptively misdescriptive terms. Some of these exclusions are closer to source significance than others, and it might be worth noting that the “core” exclusions are pretty well mixed in with the non-core ones, so that “deceptive” is listed right in between “immoral” and “scandalous.” Arguably it’s all congressional policy about what ought to serve as a mark. (See the policy reasoning in Renna v. County of Union, arguing that governments ought not to have access to ordinary trademark remedies, given their First Amendment implications.)
But anyway, the NAFTA amendments might seem to be a really obvious place to look for congressional policy about the relationship between registration and protectability. Congress intended—before California Innovations gutted the change—to switch geographically deceptively misdescriptive marks from registrable to unregistrable. Did it also intend to make them unprotectable under §43? As my research assistant pointed out, in retrospect this seems like a really obvious question. And yet, as I confirmed with Professor McCarthy, there seems to have been no consideration of that question. Perhaps this is related to the fact that most of our treaty partners operate more registration-based systems, and weren’t attuned to the fact that the US now offers essentially the same protection to registered and unregistered marks.
Relatedly, our NAFTA commitment required us to provide a remedy to persons harmed by the use of primarily geographically deceptively misdescriptive terms: “Each party [United States, Mexico, Canada] shall provide, in respect of geographical indications, the legal means for interested persons to prevent: (a) the use of any means in the designation or presentation of a good that indicates or suggests that the good in question originates in a territory, region or locality other than the true place of origin, in a manner that misleads the public as to the geographical origin of the good....”
Congress did not amend the Lanham Act to implement this provision, while it amended §2 to deal with registration. Presumably, the assumption was that false advertising law covered the situation already. Did Congress just not notice that materiality is a requirement under §43(a)(1)(B) (as it should be under §43(a)(1)(A))? The use of geographically misleading terms is therefore not unlawful unless the misleadingness is material. That’s probably not what our trading partners wanted, but it’s what they got—both for §43(a)(1)(B) and for §2, after California Innovations. (I say that materiality was not supposed to be required because part of the theory behind protecting all GIs is that different places should be encouraged to develop reputations for specific qualities. Protection should enable such reputations to develop even if they don’t exist now and therefore aren’t material now. There’s other language that can be used to specify GIs with an existing reputation.)
The best that can be said, I think, is that Congress wasn’t really thinking that much about the details, and so the NAFTA amendments don’t help us much in figuring out how we should think about the modern relationship between §2 and §43.