Pepsi sued defendants (“Sahni”) for selling bottle and can safes made from actual containers, including cans of Pepsi, Diet Pepsi, Mountain Dew, Sierra Mist, Aquafina, Cheetos, Doritos, and Fritos, all of which are Pepsi brands. The court accepted that these were all famous trademarks. (Sierra Mist and Aquafina? Really?) Pepsi licenses its marks for use on many products, including novelty and promotional merchandise, and maintains strict quality and “good taste” standards. For more on the merchandising right, see Stacey Dogan & Mark Lemley, The Merchandising Right: Fragile Theory or Fait Accompli?, 54 Emory L.J. 461 (2005).
The point of the safes is to look and even feel like ordinary Pepsi products, and the court found that Sahni had achieved this goal. In order to make the Pepsi bottle safes, Sahni used “unidentified liquids” to simulate the look of Pepsi drinks. Some of the bottle safes have removable caps, so one could mistakenly drink from them. The court found that consumers who do so are likely to believe that Pepsi is responsible. By contrast, Sahni doesn’t remove the snack foods from their canisters. As a result, the court found, consumers might eat the remaining stale snack food and mistakenly blame Pepsi for their “foul taste.” (Oh, come on. Unless Sahni modifies the “sell by” date, this is a silly argument.) Moreover, Sahni’s modification process creates sharp edges where can safe lids screw into the can bodies. Nor does Sahni use Pepsi’s “thorough rinsing process to guard against contamination and ensure product safety.”
After listing the confusion factors, the court concluded that consumers would mistakenly believe either that Pepsi was the source of the safes or that Pepsi authorized them. “Even remotely connected products” may generate sponsorship confusion. Moreover, confusion was likely not only among safe purchasers (who are actually least vulnerable, especially if Sahni were to add a disclaimer), but among people exposed to the post-sale product. I presume, however, that deceived burglars were not in the relevant consumer class, since their confusion could not harm Pepsi.
The court also found Sahni liable for federal and state dilution because “Sahni uses the marks on goods commonly associated with the concealment of illicit narcotics.” I had no idea; when I went surfing, the sellers, even the ones who sold other drug-related merchandise, uniformly emphasized that most burglars only spend a few minutes in a house, and so the safes are useful for protecting money and jewelry. Pepsi's allegations are, however, supported by the fact that this site put the safes under "Stonerware Pot Items," along with herb grinders, digital pocket scales, and pot leaf-shaped cookie cutters.
The court found Pepsi entitled to treble damages and attorneys’ fees because this was an exceptional case, though it didn’t explain why. The parties apparently agreed that Sahni should pay $15,000.
This does not seem to have been an aggressively litigated case on the defendants’ behalf, and Pepsi has prevailed in other actions against similar safe makers. As a professor, I have the advantage of being able to speculate, and here I wondered whether there might not be a type of functionality at work: the safes, to be effective, must appear to be ordinary objects. Making up a fake brand of soda, aside from posing surprisingly difficult name clearance issues, would make the safe less credible and more easily found out. Especially security-conscious people might not want to have just one bottle or can, to avoid having it stand out, so being able to put the safe with other unaltered goods could improve the quality of the product (though also improving the chances a guest might inadvertently attempt to drink it). If that’s true, the use of an actual branded product might in this case be functional. (Other possible soda bottle functionality discussed here.) The best answer, it seems to me, is that it is possible to make so-called diversion safes in a variety of configurations, including non-branded ones like sugar canisters, candles, and wall outlets – though genuine branded goods seem more prevalent at the security-product websites I visited.
As a side note, I wonder whether a book safe made by cutting a hole in an actual book would infringe the derivative works right under the 9th Circuit’s rule. For a book transformed by deletions, see A Humument.
No comments:
Post a Comment