(cross-posted from Madisonian.net)

Suppose you’re making or producing a film and you realize that, with the release of a new Tolkien-based movie, The Hobbit: An Unexpected Journey, many consumers are going to be in the mood for purchasing audiovisual content featuring small people and fantastic settings.  You’d like to be able to profit from that enthusiasm, making a movie about small people and monsters and associated elements of the fantasy genre that Tolkien helped to  create.  There have been plenty of these sorts of movies — see, e.g., from the 1980’s: Time Bandits, Willow, The Dark Crystal. So clearly the Tolkien estate doesn’t have exclusive rights to movies featuring magic, elves, fantasy monsters, and small people.

But does it have the exclusive right to movies in that genre that include the word “Hobbit” in the title?  This is, of course, a question of trademark law.  And the preliminary answer to that question is, apparently, yes.  See the reportage here. In short, the makers of a film entitled “Age of Hobbits” were recently ordered not to distribute their film on the basis that consumers could become confused about the origins or sponsorship of the movie, thinking that the film had some relationship with Tolkien’s books, Peter Jackson’s Lord of the Rings films, and the associated “Hobbit” marketing and merchandising empire.

“Hobbit,” if you didn’t know, is a registered trademark — there are currently over 100 registrations you can find on TESS.  I didn’t check every one, but the ones I saw were all registered by the Tolkien estate. The defendants argued that the term “Hobbit” was a generic term now, for rather small individuals, providing the example of Homo floresiensis.  The court was unpersuaded, given that surveys showed that about half of the public now associate “hobbits” with Tolkien’s books, not with small people generally.

Of course, this doesn’t quite settle the question when it comes to film titles, given that the Ninth Circuit allows the titles of artistic works to have some extra breathing room to accommodate First Amendment concerns.  (See Rogers v. Grimaldi, the origin of this interesting doctrine.)  So, provided that the defendant used the mark in a way that was artistically relevant, the court needed to determine if the mark here was explicitly misleading.  If so, the use of the mark could be enjoined.  If not, it should be permitted under Rogers.

Personally, I’m not sure this title was “explicitly misleading,” but I do think the use of the word Hobbit, in the font apparently very similar to the font used in the movie advertising, with a release timed right before the release of the movie, is misleading.  I could easily imagine someone with only a vague familiarity with Tolkien purchasing the movie online by mistake, presuming this was a licensed spin-off.

There is one rather strange thing about this opinion, however.  In the Mattel v. MCA case, Judge Kozinski adopted the Rogers doctrine for the Ninth Circuit and he presented that doctrine (properly,  I think) as a two-part test, stating that titles do not violate the Lanham Act “unless the title has no artistic relevance to the underlying work whatsoever, or, if it has some artistic relevance, unless the title explicitly misleads as to the source or the content of the work.”  (That’s a quote from Rogers.)  The court in this case says…

“Therefore, pursuant to the Ninth Circuit’s formulation of the Rogers test, the Court considers (1) whether the mark has acquired meaning beyond its source identifying function; (2) whether the use of the mark in the title has at least minimal artistic relevance to the underlying work; and (3) whether the title explicitly misleads consumers as to source or function.”

This first one of these is new to me, and apparently based on various things that Judge Kozinski said in Mattel when he wasn’t announcing the rule or explicitly stating the holding.  I generally admire the logic of the Hobbit court’s opinion and particularly the court’s careful treatment of trademark law issues.  But where did this new first Rogers prong come from?

In any event, the court gives the defendant a pass on it, saying that yes, some people do use the term “hobbit” to refer to small people. But then the court leverages that determination to find against the defendant on the (old) first (now second) prong:

“Moreover, Asylum has not pointed the Court to a single case in which a court in this circuit—or any other circuit, for that matter—has permitted a defendant to use the Rogers defense when the term did not in some way relate to the original. As such, the use of the mark in the allegedly-infringing title must be in some way related to the meaning associated with the mark itself, as it is only in these cases where limiting the use of the term would excessively restrict speech.”

So, if I follow this right, because the defendants were talking about “small people” and not Tolkien hobbits, Rogers is unavailable.  But can’t a party use a trademark in a title in a way that is artistically relevant, but not directly nominative?  And it isn’t this in tension with this new first step of the Rogers analysis, which requires some sort of broader public meaning for the term in order for the defendant to prevail?

It seems to me that if you require that the term have a non-trademark meaning as a first prong, then you require that the defendant employ the trademark meaning as the second prong, you’re probably going to find the use is misleading on the third prong, and that will severely narrow Rogers.  (Here the court finds the term explicitly misleading *despite* the fact that it accepts defendant’s claim that the use it makes is not in reference to the plaintiff’s mark.)

Again, I’m not critical of the outcome — I’m just a bit concerned about what this reasoning, if it has legs, could do to the protections accorded to defendants under Rogers.