Monday, May 13, 2024

In a Split Decision, TTAB Reverses Mere Descriptiveness Refusal of ARBATA for Tea Despite the Word's Lithuanian Meaning

The USPTO refused to register the mark ARBATA for "tea" on the ground of mere descriptiveness, relying on the doctrine of foreign equivalents: the word "arbata" means "tea" in Lithuanian. Applicant MJ Cobalt, LLC argued that the doctrine should not apply because Lithuanian is not a "common" language in the United States. The panel majority agreed, finding it "very unlikely that the ordinary American consumer of tea would 'stop and translate [ARBATA] into its English equivalent.'" The dissent criticized the majority's "sole emphasis on [a] statistical analysis" and its reliance on precedents dealing with likelihood of confusion rather than descriptiveness or genericness. In re MJ Cobalt, LLC, Serial No. 97673097 (May 9, 2024) [not precedential] (Opinion by Judge Christopher C. Larkin, dissent by Judge Angela Lykos).

The Board began with the issue of "the applicability of the doctrine of foreign equivalents to the Lithuanian language."

“Under the doctrine of foreign equivalents, foreign words from common, modern languages are translated into English to determine . . . descriptiveness. . . . When it is unlikely that an American buyer will translate the foreign mark and will take it as it is, then the doctrine of foreign equivalents will not be applied.” Palm Bay Imps., Inc. v. Veuve Clicquot Ponsardin Maison Fondee En 1772, 396 F.3d 1369, 73 USPQ2d 1689, 1696 (Fed. Cir. 2005).

The "ordinary American purchaser" includes purchasers knowledgeable in English as well as the pertinent foreign language. In re Taverna Izakaya LLC, 2021 USPQ2d 1134, at *2-3 (TTAB 2021).

The Board found that Lithuanian is clearly a modern language; "unlike a 'dead' language such as Latin, Lithuanian is currently spoken in Lithuania and elsewhere." But is it a "common" language in the United States?

The TMEP instructs examining attorneys to provide evidence of the percentage or number of United States consumers who speak the language in question. (Section 1207.01(b)(vi)(B)). Here, it was the applicant who provided census data: the October 2015 U.S. Census Bureau table showed that between 2009-2013, fewer than 40,000 persons in the United States spoke Lithuanian at home.

The number of Lithuanian speakers in the U.S. population is at least one order of magnitude smaller than the percentages of speakers of Spanish, French, and Russian in the United States in the cases discussed in TMEP Section 1207.01(b)(vi)(B), and it is many times smaller than the number of speakers of modern languages found to be “common” in other cases.

The Board therefore found that Lithuanian is not a “common” language in the United States, "making it very unlikely that the ordinary American consumer of tea would ‘stop and translate [ARBATA] into its English equivalent.'" Malhotra, 128 USPQ2d at 1102 (quoting Palm Bay Imps., 73 USPQ2d at 1696). It is far more likely that such a consumer will simply "take [ARBATA] as it is . . . ." Palm Bay Imps., 73 USPQ2d at 1696."

The Examining Attorney also contended, alternatively, that, even if not translated, "the term ARBATA is known in English-speaking circles," and that "upon a Google or other Internet search the term ARBATA would be found and would be used in reference to tea." In other words, it is unnecessary to apply the doctrine of foreign equivalents because "arbata" has an independent meaning as a word in English. The evidence, however, consisted mostly of foreign websites, and was unpersuasive.

The Examining Attorney has not shown why U.S. consumers of tea, a staple beverage in the United States, are likely to search for and be exposed to websites of tea companies located in Lithuania and Hong Kong.

And so, the refusal was reversed.

Dissent: Judge Lykos agreed that ARBATA, without translation, is not merely descriptive of tea. She disagreed, however, with the majority's refusal to apply the doctrine of foreign equivalents. Judge Lykos asserted that the panel majority was wrong in "[p]lacing the sole emphasis on this statistical analysis" because it "represents a departure from our precedent governing the application of the doctrine of foreign equivalents in the context of mere descriptiveness and genericness refusals."

This is because the underlying policy rationale is different and arguably stronger for applying the doctrine of foreign equivalents in the context of mere descriptive and genericness refusals as compared to likelihood of confusion refusals. Such refusals pertain to the validity of a trademark as opposed to consumer confusion.

In rejection the majority's statistical approach to the meaning of the word "common," Judge Lykos reviewed many precedents and explained: "The key point is that the doctrine applies to mere descriptiveness and genericness refusals for common foreign languages even if the general public is unfamiliar with the language."

The doctrine of foreign equivalents should not be applied “mechanically.” In re Isabella Fiore, LLC, 75 USPQ2d 1564, 1569 (TTAB 2005). In this case, it is undisputed that Applicant’s single word mark ARBATA is the Lithuanian word for “tea;” that the identified goods consist of “tea;” and there is no other meaning in English of the word “arbata.” The record shows that the United States is home to the largest diaspora of individuals of Lithuanian descent outside of the country of Lithuania. Under these circumstances, where the validity of a proposed mark is implicated, the Examining Attorney’s mere descriptiveness refusal should be affirmed.

Read comments and post your comment here.

TTABlogger comment: Quite a clash of opinions! Well worth the read. Is it time for an en banc decision by the Board on this issue?

Text Copyright John L. Welch 2024.


At 10:18 AM, Anonymous Anonymous said...

Sanity at the TTAB! Hallelujah! 0.012% of people in the USA speak Lithuanian. It is beyond me how anyone could argue that it is a "common" language. Judge Lykos is right in arguing that this decision breaks precedent with earlier decisions, but so many of those decisions were wrong.

It would be very nice if there was actually precedent and guidance on what is a common foreign language under this doctrine.

Also, I'm amused how they differentiate this from a case in which the foreign word is from a dead language like Latin. If I recall, haven't they applied the doctrine of foreign equivalents to Latin marks in the past?

At 11:17 AM, Anonymous Anonymous said...

So 40,000 Lithuanian speakers isn’t enough to apply the doctrine of foreign equivalents but 1,600 people with the surname "Remo" and only five people being named BELUSHI in the U.S. is enough to make the mark primarily merely a surname. Seems like these don't line up.

At 9:59 PM, Blogger Brian M. Taddonio said...

Not having read the decision yet, I agree with the Dissent - a word is a word at least regarding descriptiveness. It should not be removed from the available lexicon just because the percentage of speakers is relatively low. I wonder if applicants have truly run out of options other than to employ "unique" languages for ordinary words..."ke a leboga" ("Thank you" in Botswana).

At 9:37 AM, Anonymous Anonymous said...

@Brian, a line has to be drawn somewhere because at some point so few people are aware of the descriptive meaning that the consuming public as a whole would view the mark as being a fanciful mark. Should the doctrine of foreign equivalents apply when the language is spoken by a dozen people in the USA, such as maybe an obscure tribal language?

At 10:59 PM, Blogger Brian M. Taddonio said...

I have since read the decision. I appreciate the Majority's opinion on page 18, starting with "The per se rule" and ending with "cases discussed above." It recognizes that the dissent's argument "may be wise as a matter of policy," but it is not the TTAB's position to create such change that is best left to the legislature or the primary reviewing courts.


Post a Comment

<< Home