The “Trademark Use” Debate

Sometimes when legal academics debate each other about some area of theory or doctrine, it doesn’t have much impact on judges and regular lawyers. The current debate about the “trademark use” requirement is not like that. Anyone concerned about free speech and trademark law, or about keyword advertising and related uses of trademarks online, needs to pay attention too. The Iowa Law Review just published an exchange among some of the biggest trademark scholars on just these points. Let me summarize, hopefully in fairly plain English, and also give some links to those important sources that are freely available online.

First, a number of prominent scholars have argued recently that trademark liability requires the use of a trademark “as a mark,” and not in some other capacity. Just what this means is one of the topics of debate, and the various scholars supporting the idea don’t all agree with each other on the point. But essentially they suggest that certain kinds of uses of trademarks as a reference point for one’s product, rather than as an identifier for one’s product, mean that trademark law simply doesn’t apply, by definition. The most obvious example is keyword advertising: if Pepsi buys the keyword “COKE” from search engines, so that an internet user who searched for “Coke” would see a Pepsi advertisement, then by this reasoning that isn’t “trademark use” of “COKE” and so Pepsi cannot be liable for it. (And if Pepsi can’t be liable, then neither can the search engine, because the search engine’s liability is derivative of Pepsi’s, sort of like “aiding and abetting”). The argument goes that Pepsi isn’t pretending to be Coke, just to have some other undefined association with Coke, and that’s beyond the reach of trademark law. [Update: Nor, I should have added, does Pepsi suggest it is sponsored or affiliated with Coke, since that too would give rise to trademark liability.]

The most prominent article advancing this argument is by Stacey Dogan and Mark Lemley. (See Stacey L. Dogan & Mark A. Lemley, Trademarks and Consumer Search Costs on the Internet, 41 HOUS. L. REV. 777, 822 & n.184 (2005)). Others who published work taking a similar position include Margreth Barrett and Uli Widmaier.

The Iowa Law Review exchange arises from a critique of the “trademark use theory” by two other scholars who are also the co-authors of one of the big trademark law casebooks, Graeme Dinwoodie and Mark Janis. In the main article, these two argue that no fair reading of the federal trademark statute supports the idea that such a rule exists, and that even if it did it would be bad policy to categorically remove such disputes from the ambit of trademark law while the technology is still developing. Dogan and Lemley wrote a reply for Iowa, and then Dinwoodie and Janie wrote a reply to their reply.

Now Mark McKenna has written a great new paper which navigates a middle course between these positions (and which he recently presented at this conference). McKenna argues that, while there is a trademark use requirement embedded in federal law, it doesn’t really matter because it is so bound up with the central inquiries of trademark law — determining whether a use confuses consumers as to affiliation or source — that it doesn’t do any good. The questions needed to determine trademark use are too close to the questions needed to determine liability, so it doesn’t do any independent work. I agree with this conclusion that the impact of any trademark use requirement would be very limited (I’m agnostic on whether it exists or not) and I’d reached the same tentative conclusion myself before Mark spelled it out so persuasively.

What does all this mean? Well, for one, I think it means we shouldn’t look to the notion of a threshold trademark use requirement as a meaningful limit on assertions of trademark rights against uses that should be lawful. Instead, other elements of law will need to change. (I am finishing a paper now that argues one of those areas should be the “fair use” doctrine in trademark law.) Another implication of the failure of trademark use is that keyword advertising cases are really trapped in a legal gray area, and we will need to come up with some distinct rules for dealing with such indexical uses of trademarks. Overall, it means more doctrinal work is needed. Oh goody: future employment for trademark scholars and lawyers alike…

14 Responses to “The “Trademark Use” Debate”

  1. At the end of his article Google’s Law, Greg Lastowka also appears skeptical of a “trademark use” requirement being a silver bullet that will foreclose trademark law “regulation” of search results.

    I’ve been wanting to blog about Dinwoodie and Janis’s methodology for some time, since they critique an entirely economic approach to trademark. D&J argue that we can’t shoehorn all the policy concerns raised by TM law into a economically-driven analysis of things like search costs and incentive-balancing. They try to address the concerns of both radical critics of copyright, and “natural rights” types arguments for property-like control of TM’s.

    Anyway, now that you’ve commented on the substance of the dispute so incisively, I feel that I can get into the method!

  2. Thanks, Frank. I am quite skeptical. I’d much rather confront the potential social problems posed by search engine results head-on instead of giving carte blanche to Google under a trademark use theory. But that’s all in the article…

  3. This is an area of TM law that has fascinated me for quite some time. In fact, I had a rather vigorous debate with my TM prof (a student of Dinwoodie). Where the trademark use theory really gets some play is not in the search context but in the keyword advertising context. (for example the GEICO/Google AdWords case). This topic was what I took a stab at writing about for my Advertising Law final paper. It gets real hairy when you consider initial interest confusion doctrine as a basis for secondary liability against players like google.

    In my meager opinion, you either need to have a REAL likelihood of confusion analysis, dealing with CONSUMER confusion (not just anyone’s confusion) AT THE POINT OF SALE (not pre sale confusion, not post sale confusion), OR you have to come down to a more nuanced use based analysis where the use is anaylzed similary to how some courts in the second circuit have analyzed it. In order for a trademark to be used it must 1) be used, 2) in commerce, and 3) used as a mark (i.e., as a source identifier).

    After some debate with my TM prof, I think the BEST solution is to go with consumer confusion rather than torturing the use requirement beyond its ‘utility’. But, this is predicated on consumer confusion retaking the throne of the primary inquiry in trademark infringement analysis (something which courts have not been rigorous about, esp. the 9th circuit, in recent years)

  4. Chris — I think that’s absolutely correct. To put it another way, the only reason that you’d want a trademark use requirement is if you could not get rid of initial interest confusion doctrine. Trademark use is essentially an attempt to over-correct for the way likelihood of confusion has gone off course in recent years. The better solution would be simply to return confusion to its traditional roots.

  5. Nice post Bill, and thanks for the plug. In partial response to Greg and Chris, I agree with you that likelihood of confusion has gone off course. But the problems run much deeper than initial interest confusion. The real problem is that “consumer confusion” is a boundless concept when untethered to actual competition between the parties.

  6. Mark — we might disagree on this, but I’ve felt the the traditional (mid 20th century) LOC analysis seems to tether TMs fairly well to market conditions in a way that courts can generally get good results. But while I’m not in favor of going back to Borden, I’m willing to agree that the broadening of the Lanham Act past source to include “sponsorship or approval,” though, destabilizes things considerably, as does post-sale, pre-sale, dilution theories, etc.

    Is your new paper posted somewhere? I’d like to see where you’re going with this…

  7. Thanks to all for great comments!

    Like Mark, I believe the current tethering of trademark law to a boundless definition of consumer confusion is unworkable — I’d say it’s now bordering on incoherent. But the horse has already left the barn on the definition of confusion.

    The gigantic problem with modern trademark doctrine is that it pretends to be about confusion when really both markholders and legislators are motivated primarily by the desire to protect brand value. Ready identification of a brand with a source is the first component of a valuable brand, but by no means the last or even the most important. The brand’s identity and message have much more content than that. This motivation drove most or all of the various expansions of “confusion” Greg lists — they are flawed because they aren’t really about confusion at all.

    I tend to think the answer lies, not in trying to roll back the clock to any earlier and narrower conception of confusion, but to rethink the rationale of trademarks altogether. That project, however, may take just a little while… In the meantime, I’m advocating stronger defenses to respond to excesses in enforcement on the back end rather than narrowing the scope of protection on the front end, as a mechanism more likely to succeed and also more honest about the nature of the underlying problem.

  8. OK, the real issue as I see it is equitable. We can’t restrict this type of use (paid keywords searches). Google, and other search engines are essentially ubiquitous. To eliminate this type of business arrangement would essentially grant a huge monopoly for TM holders. Could THEY sell the keywords searches?

    I compare these type of uses to nominative use, competitors can use TM’s in their adds to compare their features. Although not completely on point, the analogy is close.

    Further, compare this to Yellow Pages. The Yellow Pages makes a profit by listing companies by name (TM), should this be disallowed as well.

    Although there is currently no good legal reasoning to support my conclusion (thus the struggle of scholars and courts), to find any other way would simply go against equity as our society currently sees it.

  9. Bill> I tend to think the answer lies… in trying… to rethink the rationale of trademarks altogether.

    Well, I’m all for rethinking things, but if we’re going to give up on the notion of consumer confusion & passing off, we probably wouldn’t even need to call whatever new thing we invent “trademark law.”

    We shouldn’t be surprised that trademark holders aren’t most interested in trying to prevent consumer confusion. Trademark holders, like copyright holders and real property owners, often tend to want rights beyond what the law grants them. The duty of the courts is to listen to defendants when plaintiffs overreach and keep the interests of the consumer and society in mind. For some reason, courts have done a poor job of that in the late twentieth century with regard to limiting the expansion of trademark law.

  10. Hi Greg, and thanks for commenting further.

    Of course I agree that some excesses in the expansion of TM scope should be rolled back. So I don’t quite mean “giving up” on consumer confusion as the rationale.

    But marketers and consumers all understand that something called “brand value” exists and that a stringent consumer confusion rationale does not protect it fully. Pragmatically speaking, we are not going to return trademark law all the way back to a place where it no longer covers some of the most valuable assets corporations own. Courts and especially legislators simply won’t go that far. Normatively speaking, I am open to the idea that they shouldn’t do so, at least not completely.

    Thus I don’t really see any viable alternative to a future where trademark law moves forward to a different limiting principle (or set of principles, perhaps including confusion among them to some degree).

    You suggest: “we probably wouldn’t even need to call whatever new thing we invent ‘trademark law.’” True enough. I think Mark McKenna and others might propose calling it “unfair competition” instead, and drawing on those principles for the future. I certainly don’t claim to have good clear answers about what replaces the current system. But I am pretty sure consumer confusion alone won’t work.

  11. Bill — we should chat about it some time, but I think potentially the recent TDRA could deliver a new regime like the one you’re looking for…

  12. Scholarly exchange in action, folks!

    Greg: I agree that the structure of categorical safe harbors in the TDRA has promise, but much of the execution of that idea in the statutory language is horrific. In part that’s because the new language incorporates some concepts, like nominative fair use, that are freighted with lots of problems. Another issue is that we might be more willing to carve out broad exemptions from dilution liability in situations where at least some potential room for traditional infringement liability, in the face of true confusion, might make sense. But, yes, we should talk. Thanks!

  13. I was glad that someone mentioned the TDRA in this context because I was curious what significance, if any, people saw in its language “other than as a designation of source” or Section 33(b)4’s “otherwise than as a mark” with regard to the trademark use debate. On the face of it, it would appear that courts already are determining, at least in some contexts, when something is a trademark use. On the other hand, from the Chewy Vuiton case, it does not appear that the analysis on that point is very clearcut, and a “trademark use” requirement may only introduce another slippery concept into an increasingly arbitrary analytical process. Moreover, I’m not aware of much caselaw elucidating specific evidentiary elements that indicate or establish a designation’s use is “trademark use.”

  14. Trade marks can be used on computers and by statues and papers…but we the people should have a privalige to do Trade marks because not just the big people…. we have people that died we havev important people… we should have a part in saying something or doing something so we the people can make our own trade mark of things… we should have a right in the trade marking we have a saying..