Much work has been done on the theoretical foundations of trademark law generally, but very little on trademark registration specifically (at least in the U.S.). The reason is that, for most of the last fifty years, courts have been telling us that, with a few exceptions, registration really doesn’t matter. Courts evaluate the validity of an unregistered mark under essentially the same standards as registered marks, and they use the same likelihood-of-confusion analysis to determine infringement.
But it turns out to be hard to maintain a rule that registration means nothing when the Lanham Act clearly was intended to create some substantive rights that did not previously exist. It’s also difficult to ignore the elaborate regulatory apparatus the PTO has constructed to evaluate applications to register – one that includes detailed rules about the format in which a mark is claimed and the goods and services are described, and that provides for administrative proceedings to oppose or cancel registrations. Why would any of that exist, and why would companies spend so much time and money dealing with registration, if it was meaningless?
So, not surprisingly, registration does sometimes matter to courts – indeed, in its recent B&B Hardware decision, the Supreme Court described it as significant. But how is it significant, and when? As Rebecca Tushnet wonderfully demonstrates in her terrific new article Registering Disagreement: Registration in Modern American Trademark Law, there is no consistent answer to that question, because trademark law has no theory of registration.
We should start with a little background here, because registration’s insignificance was not inevitable. It is, rather, the result of changes courts have made to trademark and unfair competition law over the last half century. Once upon a time, registration was clearly understood to be procedural. Only certain signs (“technical trademarks”) were federally registrable, and only federally registrable marks could be enforced in federal court. But registrations did not create trademark rights – those were always understood to arise through use, and registration simply acknowledged those rights and provided a federal enforcement mechanism. Parties that lacked a federal registration might have had a cause of action for unfair competition, but that was a common law claim (not federal) with additional proof requirements.
After Congress passed the Lanham Act, however, courts began assimilating unfair competition law into trademark law by equating registered and unregistered marks. They did so (without any clear statutory authority) because they wanted to create a federal cause of action for unregistered marks – which is to say that they denied registration’s significance in cases in which the question was whether unregistered marks should have some lower status than registered marks. Put simply, courts assimilated unfair competition law into trademark law by leveling up – by giving to unregistered marks essentially the same status as registered marks. And they did so without giving much consideration to the policy reasons behind the particular requirements of registration.
The project was enormously successful – so much so that, with the exception of priority and the geographic scope of rights, registration is largely irrelevant to civil litigation. Parties spend significant resources crafting trademark applications and resolving disputes over registrability even though the resulting register has almost nothing to say about the scope of any party’s rights. Even incontestability, probably the most controversial concept when the Lanham Act was drafted, doesn’t often make much difference, simply shifting the focus of a case from validity (where incontestability can be dispositive) to infringement (where it makes virtually no difference).
And yet, the question of the significance of registration and the registration process keeps popping up, precisely because it’s hard to believe registration is actually so inconsequential. The problem, as Tushnet persuasively argues, is that registration seems increasingly a conceptual misfit; trademark law’s modern emphasis on confusion over everything else makes it that much harder to attribute distinctive importance to registration. Because courts tend to think of trademark law nearly exclusively as a tool for eliminating consumer confusion, they conceive of its ends in empirical terms, focusing on the contexts in which consumers actually encounter marks. But registration isn’t about consumer understanding. The registration system is motivated primarily by nation-building and market regulation goals that fit uneasily with an empirical, consumer-confusion orientation. As Tushnet argues, “even if trademark is a matter of consumer protection, trademark registration is a matter of industrial policy, contributing to a national unified market by providing incentives for registration, including nationwide priority over other users.” (P. 10.)
The consequence has been deep disagreement over the role of registration, and a remarkable lack of clarity. For example, decades of post-Lanham Act case law hadn’t worked out the extent to which decisions of the PTO and its administrative tribunal the Trademark Trial and Appeal Board (TTAB) have preclusive effect in civil litigation. While the standards used in both settings often look (at least superficially) the same, those standards are applied in very different ways. Thus, when the Supreme Court took up the question in B&B Hardware, it insisted that registration is indeed significant and that TTAB decisions can have preclusive effect. But it then limited the practical significance of that conclusion, acknowledging that preclusion will not apply in “some or even many cases” because the TTAB so often approaches questions differently than do courts. That result seems plausible enough, given the differences between the approaches in the TTAB and courts. But left entirely unanswered is why the approaches differ so much. If courts recognized the primary business regulation function of registration, they might well put more emphasis on registration status and the content of registrations despite registration’s disconnect from consumer understanding.
Perhaps even more remarkably, it remains unclear some seventy years after passage of the Lanham Act what effect refusal or cancellation of a registration has on the protectability of a mark. On the one hand, unfair competition doctrine once existed precisely to offer some protection for unregistrable subject matter. Thus, historical practice suggests that bars on registration have nothing to say about the availability of claims based on unregistered rights. On the other hand, when courts assimilated unfair competition into trademark law, they eviscerated the important differences between trademark infringement and unfair competition more generally, so it’s hard to see anymore why unregistered marks should get (mostly) the same treatment as registered marks but not be subject to the same limitations. On that reasoning, in its recent In re Tam decision, the Federal Circuit suggested that marks that are unregistrable under section 2(a) likely are unprotectable.
This state of affairs seems intractable, and Tushnet candidly acknowledges that there are good arguments both for making registration more substantive, such that the system does more to “manage relationships between businesses regardless of consumer confusion or nonconfusion,” and for making registration purely procedural. (P. 5.) Indeed, she admits that “none of the tensions in current law can be entirely resolved to favor only one side.” (Id.)
But that doesn’t stop Tushnet from offering a number of ideas about how a system that took registration more seriously might look. Her discussion here is characteristically thoughtful and nuanced. One of the things I appreciate most about the paper is the way Tushnet demonstrates that taking registration more seriously would not necessarily mean broader protection; indeed, it might provide some new tools for limiting the scope of trademark rights – a conclusion many readers might find counterintuitive.
In the end, consideration of any of the possible solutions requires frank discussion of political choices that mostly have been excluded from conversation about trademark law. How do we rank priorities in nation-building (or global economy-building) against concerns about consumer interests? Surely the answer isn’t that we pretend that those interests are the same. But resolution of that issue is critical to thinking about the importance of registration in at least some contexts. In my view, this paper is the beginning of a sustained inquiry into the role of registration in various settings, and the paper is going to be cited heavily. Not because it has all the answers, but because it asks all of the right questions. It’s an excellent piece of work.