Bob Brauneis on geographic marks

Bob Brauneis, Geographic Marks and the Protection of Competitor Communications, GW IP Colloquium, Apr. 10, 2006

Professor Brauneis and his colleague Roger Schecter have written a paper that was the subject of his presentation, which, among other things, had great accompanying visuals. Marks with geographic components are everywhere – Nantucket Nectars, Philadelphia Cream Cheese, Casablanca Fans, Pepperidge Farm Geneva Cookies, California Pizza Kitchen, Wall St. Deli, Bank of America, Toyota Tacoma – the list goes on. There’s a trend to register these marks immediately, rather than requiring a secondary meaning showing, for example TRENTON for computers (New York owner), CHICAGO for water coolers (Canada), and SANTA FE for barbecue grills (Georgia). An ITU for SWISSCELL batteries (New Jersey) was allowed last December.

The central thesis: the liberal grant of immediate registration gives too little weight to competitors’ interest in communicating about where goods come from. Swissbatteries.com, for example, sells Swiss-made batteries and uses SWISSBATTERIES.COM as its brand. The basic problem is that we don’t want great similarity in marks, which increases consumer search costs and harms producers. At the same time we want comparison to be easy – we need to see certain goods/services as substitutes or partial substitutes for each other. If two producers make Swiss watches, they both need that term to communicate.

We could deal with the problem by adjusting rules about the scope of trademark or eligibility for trademark. One scope rule: descriptive fair use might protect the competitor’s communication. KP Permanent, though, held that the degree of confusion may be relevant to a fair use defense, so you can still have a problem – “Swiss made” prominently enough on the package could create enough likelihood of confusion with SWISSCELL that it would go to the jury, creating barriers to truthful competition.

The alternative is restricting geographic marks through eligibility rules, but they’ve been getting progressively weaker. Brauneis reviewed the history since 1946, when geographic marks with secondary meaning were allowed registration by the Lanham Act. We moved first from the rule that a word generally recognized as a place name by the public could only be immediately registered if it was arbitrary as applied to the goods (delayed registration if the word could conceivably indicate geographic origin), to a rule that required a good-place association before immediate registration would be refused. Brauneis argued that this was already a mistake – we should look at what competitors, now or later, could communicate and might find more difficult to communicate if registration were granted.

How could there be a gap between a goods-place association and competitor interests? Certain shifts in consumer understanding are foreseeable and extremely easily accomplished. The foundational In re Nantucket case offers a good example – is it impossible to communicate that shirts were made in Nantucket by putting the term on a shirt? In other words, are consumers irrevocably commited to perceiving “Nantucket” as a metaphor for a tony summer resort? Brauneis said no; rather, lots of consumers aren’t ready to form an opinion.

This problem is tied to how things are produced and traded in an inpersonal global economy where we don’t generally know where our possessions were made. Consumers’ understanding is malleable. Geographic terms are often rich in metaphorical meaning, but one can communicate both metaphor and reality at once (SILICON VALLEY GRAPHICS indicating both actual location and the quality of being tied into the latest advances in computer imaging). So how easy would it be to convert a place name into a clear signal of geographic origin? NANTUCKET NECTARS is a convenient example; the bottle features a picture of an island and a map of same, along with a statement that the drinks were “born” in Nantucket and “embody the wholesome quality of the island.” The TTAB said that, in context, the mark was descriptive.

This possibility for meaning switch exists even without an applicant’s label; the SWISSCELL ITU application came without packaging. The TTAB allowed the application because there was no association of batteries with Switzerland. Then, after the registration issued, the packaging showed up, adding a Swiss flag and red color which increased the geographic linkage.

The authors' conclusion is that there should be no immedate registration if competitors exist in a location designated by applicant’s brand name, and the burden should be on the applicant to show this. If the applicant doesn’t, registration has to await secondary meaning.

But things got worse: In 1993, NAFTA implementation separated geographically descriptive from geographically deceptively misdescriptive marks, forever barring registration of the latter, and the Federal Circuit’s bizarre In re California Innovations decision said that therefore only materially deceptive marks could be considered geographically deceptively misdescriptive. Result: immediate registration for CALIFORNIA INNOVATIONS for bags not from California, whereas registration would have been delayed for California-made bags.

It’s easy to criticize California Innovations (yeah, easy because it’s a really bad decision!). Brauneis wants us to attend to the overall result. The current standard is that if a word is generally recognized as a place name, you get immediate registration if the examiner can’t prove a goods-place association or if there’s a false association the examiner can’t prove material; registration is delayed if there’s a true goods-place association; registration is forever denied if it’s a deceptive association.

NAFTA and TRIPS require parties to refuse registration of misleading geographic indications and to allow private parties to challenge their use, but GIs are a defined term. Only where the particular quality, reputation or other character of the goods is attributable to geographic origin is there a GI. States must ban misleading use (1) when you can trace a particular quality to the place named by the mark or (2) when the region has a reputation for the particular good in question, as with Swiss watches. This is close to a materiality standard, though there need be no showing of actual influence on consumer purchases. Some things that might influence consumers are excluded, such as a desire to support or boycott a particular regime.

The authors’ proposal would ban registration of terms generally recogizned as place names under the circumstances set out in the treaties and delay registration if applicants failed to show absence of competition in that place. The obvious question is why make “absence of competition” the standard rather than, say, “absence of potential future competition”? Afghanistan might get a semiconductor industry someday, even if it lacks one now, so why should AFGHAN SEMICONDUCTORS be registrable immediately (or at all)? I note in this regard that the web pages for Nantucket shirt stores Brauneis showed as part of his argument that the meaning of NANTUCKET for shirts can be geographic showed shirts sold in Nantucket, but made no representation that the shirts had been made there.

The presentation was very useful for me; though GIs are the hot topic right now, the appropriation of geographic terms by specific firms remains important, and has implications for how we should think about GIs too.