Thesis Chapter 3 Draft



3.0 | Overview

Under the DoFE words in foreign languages are translated into English in an attempt to determine the level of genericness, descriptiveness, and the potential similarity in meaning and connotation to pre-existing English marks in an effort to determine likelihood of confusion.

Sujata Chaudhri, an attorney at Cowen, Liebowitz & Latman in New York notes that, “much of the United States market place is a melting pot. Consumers speak, understand, and are exposed to brand names in numerous foreign languages”[1]. Because the DoFE is a guideline rather than an absolute rule, it need not apply to every non-English word that appears in a trademark. In Sutter Home Winery, Inc. v. Madrona Vineyards, L.P.[2], in February 2005, the California Northern District Court noted that the DoFE should only be applied, “when an appreciable number of ordinary American purchasers of the particular goods or services in the United States, who speak English as well as the pertinent foreign language, will understand the meaning of the foreign-word mark at issue, and will actually translate that mark into its English equivalent”[3]. Included here are a number of burdens that must be met before the DoFE should be applied. Namely, the ordinary American consumers of those particular goods or services must be identified and it must be identified that this number of consumers constitutes a level that is considered appreciable. The comment is then made regarding whether or not those consumers will be assumed to speak that language (as was discussed in Chapter 2). The next two burdens, however, are the concern of the ensuing chapter. The consumer must first understand the foreign-word mark at issue and be shown to actually translate that mark into its English equivalent. If these two steps do not occur, confusion over a translation or interpretation cannot be rightfully shown. This is the reason that the DoFE is not (or should not be) automatically assumed to apply to every non-English word found in a trademark. The name of the doctrine of foreign equivalents would seem to imply, as Rest maintains above, that words will apply if they are translated into their English equivalents. Yet, courts “have applied the doctrine irrespective of whether the non-English words have precise English translations”[4]. Questions of likelihood of translation have plagued courts as well as the TTAB, often setting the two at odds[5], such as in a disagreement over the likelihood of translation between VEUVE ROYALE and VEUVE CLICQUOT PONSARDIN in French and THE WIDOW in English. It is noteworthy in this case that the court and the TTAB differed in their construction of the American purchaser, with the court referring to the “average American purchaser” and the TTAB referring to an “appreciable number of American consumers” (see Footnote 67). In their 2006 opinion for In re Mark Thomas, dealing with the French term MARCHE NOIR and its confusability with BLACK MARKET, the TTAB uses the terms “ordinary American purchaser”, “average American buyer”, “ordinary American purchaser who is knowledgeable in the foreign language”, “appreciable segment of the population”, and “French-speaking public”, apparently all referring to the same thing[6]. These all, further, contrast with the use of “appreciable number of ordinary American purchasers of the particular goods or services in the United States, who speak English as well as the pertinent foreign language” in the Sutter Home Winery case[7], just a year earlier. As Chaudhri notes, “there seems to be confusion over the description of a relevant consumer under the doctrine of foreign equivalents”[8]. It is for this principal reason that who the relevant consumer in question is must be codified for all cases pertaining to the DoFE. Unless this is done, it is very difficult for courts to determine with any fixed and equitable burden, whether or not the consumer in question would first understand and then actually translate any given trademark.

The TMEP notes that the DoFE only applies to modern, common languages that are familiar to an appreciable number of American consumers[9]. Yet in the cases mentioned above, even with regards to languages such as French, which is the second-most spoken language in the United States after Spanish[10], disagreement can arise as to whether or not consumers are likely to actually translate certain words into English, with the court finding MARCHE NOIR likely to be translated and confused with BLACK MARKET[11], but VEUVE unlikely to be translated and confused with THE WIDOW[12]. Likewise in Spanish, the most spoken language in the United States after English, the court found that the average consumer is unlikely to translate and thus confuse TIA MARIA and AUNT MARY[13].

Further, it is possible for the doctrine of foreign equivalents, despite the suggestiveness of its name, to be applied whether or not a non-English word has a precise or exact translation into English. Frequently, connotation and colloquial re-analysis of certain words and phrases will play a large role in any determination of likelihood of confusion[14]. Simply because a word is encountered as a trademark, however, does not necessarily mean that word will be understood by a consumer or translated appropriately, thus resulting in confusion. Translation may also be more or less likely given certain marketplace considerations, like the sophistication and cost of the product or decision at hand. This chapter will present current thinking on translation and the determinants of word translation as it applies to the DoFE, coupling this with discussion of relevant cases. In doing so, this chapter will address a number of concerns with courts and the TTAB’s often divergent findings regarding likelihood of translation in cases.

3.1 | Psycholinguistic Models of Translation

Jacob Jacoby, Merchants Council Professor of Consumer Behavior at the Stern School of Business at New York University, attempts to address the importance of states of mind in purchasing by noting that, “the principal focus of Sections 43a and 43c of the Lanham Act is on the states of mind — particularly confusion, deception, and dilution — of the relevant public (e.g., prospective purchasers). Other key concepts of trademark law (acquired distinctiveness, secondary meaning, fame and genericism) also refer to psychological states of mind. Yet beyond acknowledging that these concepts refer to psychological states of mind, little has been done to describe how the vast body of relevant psychological literature provides an understanding of the development of and change in these psychological states”[15]. For Jacoby, then, “a unique brand name and cohesive brand identity is probably the single most powerful piece of information for the consumer”[16] based on psychological considerations. It is important always to remember that even though individuals encounter products in the marketplace, that aggregates, or non-insubstantial numbers of people are the primary concern of the law[17].

Nevertheless, it is crucial to identify and establish the perceived identity of the consumer in question because it is the psychological reality of that individual, rather than the objective or intended reality (or in this case, word meaning), that is most determinative in the marketplace. To this end, Jacoby presents the example of an individual being presented with two vertical lines arranged in parallel. Depending on the background and knowledge of the particular individual, this symbol can be translated differently. With regards to the two vertical lines,

“most would interpret this physical stimulus as representing the number eleven. On the other hand, someone accustomed to working with antiquities or signed and numbered lithographic prints might interpret this same objective stimulus as representing the Roman number two. Someone else conversant with the binary system used in computers might interpret two one’s in parallel as representing the binary number four…the external, physical, objective environment is always being interpreted in terms of what the individual already knows. It is not objective reality, but psychologically perceived reality that determines how we interpret and react to the world about us.”[18]

In addition to consumers, not all products can be equivocated. Muehling (1990)[19] notes that, “as a consumer’s perception of product importance increases, so does involvement and the information search activities associated with the product purchase” (pg. 175). That is to say that customers are likely to invest more thought and consideration into purchases that are more expensive or more important to them, or ones in which they are more greatly concerned with the relationship between the consumption of a particular product and the gains and losses that might be incurred by that consumption. To this end, it is reasonable to conclude that consumers would put greater time and energy into translating or otherwise understanding foreign language words as parts of trademarks for things that are expensive or more high-risk purchases, thus lowering the likelihood of confusion, than for products which are more quotidian in nature and more subject to impulse buys.

de Groot and Christoffels[20] cite work by Paradis[21] addressing the neurolinguistic models for bilingualism. Paradis suggests two different models, the use of which may vary depending on the experience of the translator at hand. Model 1, which will be referred to as vertical or conceptually mediated translation involves a strategy wherein, “the source language utterance is decoded until a non-linguistic meaning is abstracted, via phonological, morphological, syntactical and lexical semantic analysis. Subsequently, the abstracted non-linguistic meaning is encoded in the target language via lexical semantic, syntactical, morphological, and phonological processing steps”[22]. This is to say that upon seeing a word in the source language (SL), a consumer will automatically garner resources that will allow them to search in the target language (TL) for the word that is the most conceptually or semantically relevant; the term of ‘best fit’, rather than a direct translation. Paradis also assumed a second strategy of translation, model 2, which will be referred to as horizontal translation or transcoding. Under this model, translation “involves the literal transposition of words or multiword units into their equivalents in the target language. Transcoding may occur at all the various levels in the system (phonological, morphological, syntactical and at the level of lexical semantics). Translation experience and the amount of translation training determine which of the translation strategies, conceptually mediated translation or transcoding, is used most”[23]. Under this model, regardless of real semantic ‘best fit’ or common usage, translators are most likely to search in their memory for words described as the exact equivalent in the TL of the SL word in question. Paradis assumed that especially skilled translators (those who were very well-practiced) exploit this latter route of translation much more frequently than do non-skilled translators. In their own work, de Groot and Christoffels found both translation strategies to occur. If this is assumed, although both methods occur, it is likely that the ordinary, unsophisticated consumer will more readily rely on the conceptually mediated model rather than the model of transcoding and direct translation. Along these lines, de Groot (1992)[24] “developed a model of bilingual lexicosemantic organization, the Conceptual Feature Model (CFM), in which words in each language activate a series of conceptual features. The features activated by one word, for example, “friend”, are not necessarily the same features activated by its translation equivalent, “amigo.” Hence, “friend” may be associated with the concepts [McDonalds] and [honesty], while “amigo” may be associated with the concepts [honesty] and [male]. The difference in the conceptual features linked to each translation-equivalent word could be due to the different contexts in which the words are learned and normally used”[25]. The Conceptual Feature Model can be illustrated as follows[26]:

Marrying these three sets of findings means that the ordinary consumer is assumed to be more likely to search for words that are the most conceptually relevant, rather than direct equivalents to the term in question. Additionally, when this conceptual search takes place, according to the Conceptual Feature Model, the concepts activated by one word are not always necessarily the same for the translation-equivalent in another language.

Duyck and Brysbaert’s findings[27] suggest that word-level variables play a central role in translation, with common, well-confined words that enjoy great overlap amongst languages (such as numbers or colors) resulting in easier translation that more abstract words[28]. The two found that, “Consequently, backward (and for- ward) translation may imply a different amount of semantic activation depending on word-level variables. For instance, because number words (or color words, see the results of La Heij et al., 1996) have a well-confined meaning that (virtually) completely overlaps across languages, they will develop strong L2 form-to-meaning mappings earlier than words that have a more diffuse meaning with language-specific connotations (e.g., abstract words). Consequently, their translation will trigger semantic activation more easily, and earlier in the word-learning process, than other types of words will. Abstract words (e.g., Figure 2, duty) are represented by a more fuzzy set of semantic features and may have slightly different meanings in different languages, yielding smaller semantic translation effects”[29].

Duyck and Brysbaert believed that connection weights differed as a result of word characteristics.

The model above[30] illustrates varying semantic overlap and different lexico-semantic connections, with solid lines representing stronger links than dotted lines. Connections are assumed to be stronger for words with greater form overlap (in this case ball and bal for the English-Dutch bilingual) than for words with less form overlap (in this case duty and plicht for the English-Dutch bilingual). For translation, then, words with large form overlap (cognates) enjoy greater levels of translation, with non-cognates suffering a lower rate of translation. These findings suggest that cross-lingual form overlap ought to be a consideration when looking at non-English words in trademarks. More importantly, Duyck and Brysbaert contend that, “the influence of cross-lingual lexical-form similarity operates on the word level, not at the language level”[31]. What this suggests is that lexico-semantic organization in the brain depends more on word form overlap than on general language proficiency, with Duyck and Brysbaerts’ results showing similar semantic translation patterns in both balanced and un-balanced bilinguals who differed in their language proficiency. Their findings suggest that translation patterns are different for lexically similar languages than for lexically dissimilar languages, regardless of proficiency. Pursuant of this finding, “semantic translation [was] weaker with two more similar language pairs (Dutch-English and Dutch-German), than…less lexically similar languages (Dutch-French)”[32].

David Luna and Laura Peracchios’ research, of the University of Wisconsin-Milwaukee, focuses specifically on the bilingual consumer. They note that, “advertisers seem to agree that it is more effective to advertise in their consumers’ native language than in their second language”[33]. For this reason, it is important to establish linguistically informed justifications for the DoFE as courts and the TTAB attempt to determine under what circumstances an advertiser will actually be able to market a product in consumers’ native language. Crucial to harnessing the buying power of the U.S. Hispanic market, for example, which Luna and Peracchio note is estimated to sit at around $200 billion annually, is the ability to reach out to consumers in Spanish. The Revised Hierarchical Model (RHM) introduced by Kroll[34] is a widely accepted model for concept representations in bilinguals. The RHM suggests that two levels of meaning exist in the bilinguals mind, the word or lexical level and the conceptual or meaning level. Under the RHM, “at the lexical level, each language is stored separately. However, at the conceptual level, there is a unitary system in which words in each language access a common semantic representation or meaning”[35], such that the model appears as follows:


In this way, through the RHM, individuals are presumed to access meaning through their L1, with stronger conceptual associations existing between an individual’s memory (concepts) and their lexical representations in their L1. Kroll noted[37], however, that even for individuals who become fluent in both languages a residual asymmetry still exists in support of this initial model. The L1 enjoys a much greater lexical stockpile than does the L2, with clear vocabulary superiority in most cases. To this end, processing messages in the L2 at the conceptual level, rather than the lexical level, is much less likely than processing the same message in the L1. As it applies to trademarks and branding, then, “the RHM seems to imply that the intuitions of marketers may be correct. If one is trying to reach a Hispanic audience that has Spanish as a first language and is relatively fluent in both English and Spanish, it is better to do so in Spanish. The RHM supports this hypothesis because of the residual effect in conceptual-link strength. Messages in the consumer’s first language are easier to relate to the information stored in the semantic level than messages in the consumer’s second language”[38]. To this end, trademarks in a consumer’s second language (despite fluency) are believed to register more at a lexical level, with the ordinary consumer seeking meaning through a lexical item in the L1 (an equivalent, in this case). This would be more natural (and easier) for the average bilingual consumer than assuming the consumer, when presented with a trademark word in their L2, would directly relate that word to a concept stored in memory, rather than doing so through a lexical item in the L1.

Noriega and Blair (2008)[39] ran tests confirming, “previous studies in the psycholinguistics literature [that] have shown that the choice of language may influence the con- text of thoughts remembered” (pg. 74). They also found, “a positive, significant interaction between language and consumption context, such that when a native-language (Spanish) advertisement referred to a native-language consumption context (dinner), FFHH-related thoughts[40] were more likely to occur” (pg. 81). These thoughts were found to have a positive impact on purchasing by these groups. The effects of language on thoughts, then, may be moderated by the consumption context of the product. Thus, advertising in a minority audience’s native language…will vary across product categories” (pg. 81). Product categories may also have an effect with regards to purpose, with products that are meant for use in the home triggering greater engagement from foreign language speakers than products that are meant for use only in L2 contexts such as school or work[41].

Catford (1965)[42] notes that translation may occur to different extents, with full translation occurring when every piece of L1 our source language text or word is converted into the L2 or target language. Partial translation may occur, however, where some parts of the source language are not translated and are merely incorporated into the target language, though it is not specified how likely this is (pg. 21).

Eight key summaries can be distilled on the basis of the psycholinguistic, translation, and branding studies presented in Section 3.1:

  1. The same symbol (word) may be translated in a number of different ways depending on a person’s background (Jacoby, 1998).
  2. Consumers invest greater thought and consideration into products deemed as more important or more expensive. It is reasonable to conclude that consumers would put greater time and energy into translating or otherwise understanding foreign language words as parts of trademarks for products that are expensive or more high-risk purchases (Muehling, 1990).
  3. Consumers are likely to take words from the SL and translate them into the TL word that is the most conceptually or semantically relevant; the term of ‘best fit’, rather than a direct translation (de Groot and Christoffels, 2006) and (Paradis, 2004).
  4. Semantic features and associations activated by one word are not necessarily the same features activated by its translation equivalent (de Groot, 1992) and (Luna and Peracchio, 1999).
  5. Words representing abstract thoughts, and non-cognate words in one language can generate substantially different concepts than words that are common across all languages, such as colors and numbers  (Duyck and Brysbaert, 2008) and (Luna and Peracchio, 1999).
  6. Words with large form overlap (cognates) enjoy greater levels of translation, with non-cognates suffering a lower rate of translation. Lexico-semantic organization in the brain occurs more at the word level than the language level, regardless of proficiency, resulting in differing translation patterns and standards for lexically similar languages than lexically dissimilar languages (Duyck and Brysbaert, 2008).
  7. Words are organized at two different levels in the brain – lexical and conceptual. To this end, trademarks in a consumer’s L2 (despite fluency) are believed to register more at a lexical level, with the ordinary consumer seeking meaning through a lexical item in their L1 (searching for a learned equivalent) (Luna and Peracchio, 2002) based on (Dufour & Kroll, 1995; Kroll & de Groot, 1997).
  8. The effects of language on thoughts are moderated by the consumption context of the product. Product categories also have an effect, with products more naturally associated with an L1 versus an L2 triggering greater L1 thinking, and vice versa (Noriega and Blair, 2008).

3.2 | Unpacking Equivalence

Language Marketplace, a firm specializing in translation with offices around America and Europe, addresses the question of translation equivalence (TE) in a blog post from December 2010. They note:

“Equivalence, at the abstract level, is a rather necessary and important term in the field of translation studies (Edwin Gentzler 1993 p.58)…

Broek (1981 p.33) also redefines the term equivalence by the concept of “true understanding”. Another scholar M. Mehrach (1997 p.44) also considers equivalence “an impossible aim in translation”. He supports his saying by the idea that no two languages share the same linguistic structures, social or, cultural aspects. Instead, he proposes the use of the term ‘adequacy’ for the ‘appropriate’ translation, that is,” a translation that has achieved the required optimal level of interlanguage communication under certain given conditions.”

Hervey and Higgins (1995 p.75) believe that the principle that a translation should have an equivalence relationship with the source language text is problematic. For supporting their idea, they say that there are three main reasons that an exact equivalence or effect is difficult to achieve. Firstly, it is impossible for a text to have constant interpretations even for the same person on two occasions… Secondly, translation is a matter of subjective interpretation of translators of the source language text. Thus, producing an objective effect on the target text readers, which is the same as that on the source text readers is an unrealistic expectation. Thirdly, it may not be possible for translators to determine how audiences respond to the source text when it was first produced…

Monia Bayar (2007)…in her book To Mean or Not To Mean…distinguishes between formal equivalence, semantic equivalence, cultural equivalence and pragmatic equivalence”[43]

Based on these assertions, and the finding of the courts, equivalence may not be the most appropriate term for the DoFE and may, in fact, be a misnomer in practice, with courts and the TTAB exercising prerogative in refusing registration to marks regardless of whether or not they are found to be literal equivalents in a foreign language.

3.3 | Translation Assumptions in Court and TTAB Decisions

Contrasting decisions were reached for the case of In Re Sarkli, Ltd.[44]. The TTAB originally refused registration to the French mark REPECHAGE because it found a likelihood of confusion with the English SECOND CHANCE, but one year later, on appeal, the US Court of Appeals for the Federal Circuit reversed this finding on the basis that the TTAB erred in its application of the DoFE.


The court found that REPECHAGE was not, in fact, the direct equivalent of the phrase SECOND CHANCE in English, and noted that, “the test to be applied to a foreign word vis-a-vis an English word with respect to equivalency is not less stringent than that applicable to two English words”[46]. They went on to note, “we do not, of course, rule out the possibility that likelihood of confusion may be shown between an English word mark and a foreign word mark which are not exact synonyms, just as two English word marks need not be exact equivalents in meaning to create a likelihood of confusion. But where the only similarity between the marks is in connotation, a much closer approximation is necessary than has been shown here to justify a refusal to register on that basis alone where the marks otherwise are totally dissimilar”[47]. Based on the studies in this chapter, and because SECOND CHANCE represents an abstract concept as opposed to a concrete object or idea, the justification for reversal by the court is correct, but the appellant nevertheless had to go through the process of having their mark refused registration a year prior as a result of an improper finding at the hands of the TTAB with regards to equivalency and likelihood of confusion based on connotation. This finding that confusion is unlikely is appropriate despite the fact that French is the second most widely spoken language in the U.S. after Spanish.

Another French word was at issue in the case Continental Nut Company v. Le Cordon Bleu, wherein the applicant, Continental Nut Company, sought to register the product CORDON BLEU for shelled nuts (it already possessed a trademark for the English BLUE RIBBON), and was opposed by Le Cordon Bleu, who offered gastronomy classes under the name CORDON BLEU.


Both the TTAB and the Court rejected the applicant’s ability to register CORDON BLEU, finding that BLUE RIBBON and CORDON BLEU create different commercial impressions for the American public[49]. Here, however, the court and TTAB did not mention the French-speaking population whatsoever, instead defining relevant consumers as the American public at large (noted above), and “those interested in fine cooking”[50]. This again exposes a contrariety in the treatment of who the relevant population is. If the American population, not fluent in French, is being considered, and the average American is deemed to be familiar with the colloquial usage of the term CORDON BLEU, despite its literal translation (and trademark representation) as BLUE RIBBON, then that is one category of consumers. If, however, the court and TTAB are considering consumers whose L1 is French, then their treatment of CORDON BLEU may be different than the popularized American understanding. Because the effects of language on thoughts can moderated by the consumption context of the product, it is strange that the court equivocated both product types as existing in the same market because they both pertain to food. The mark applicant, Continental Nut Company, sells nuts whilst Le Cordon Bleu is a well-respected culinary arts school[51]. It is natural to assume that the context of a culinary arts school may influence the understanding of CORDON BLEU by consumers, but it is difficult to make the argument the exact same commercial impression would exist consumers in the supermarket buying nuts. Just because the products both vaguely involve food, the consumption contexts vary significantly with the latter context having a significant impact on the understanding of CORDON BLEU that the ordinary consumer may not have while buying nuts.

In the case of CORDON BLEU, the court found a colloquial and popularized understanding of the term more relevant to the case than the literal English translation, BLUE RIBBON. This finding contrasts with the case of In re Thomas[52] involving MARCHE NOIR and BLACK MARKET, yet again in French. Despite pertaining to the same language, here the TTAB was concerned with how the term would be understood by the “French-speaking public” rather than the American public at large, or, as was the case in the CORDON BLEU case, a conceptualized public interested in minerals and jewelry consumption. The applicant, a vender of primarily counter-culture jewelry, argued, “that an analysis of the marks requires more than simply the literal translation of MARCHE NOIR…that when MARCHE NOIR and BLACK MARKET MINERALS are compared in their entireties, the two marks are not at all similar in sound or appearance, and moreover that the connotations of the two marks are different”. To this end, relying upon dictionary entries, the applicant noted, “that the word “marche” in French can be translated to mean “deal” or “dealing.” [They] also submitted a printout of Registration No. 2129644 for the mark THE BON MARCHE pointing out that the translation is listed as “good bargain.” In addition, applicant points to its website showing, according to applicant, that its goods are substantially counter-culture or “Goth” jewelry and accessories. Applicant concludes from the evidence that applicant’s mark can have a connotation of a black/dark deal, a dirty deal, or a “deal with the devil”[53].

The TTAB in this instance found that, “there may be different definitions and meanings associated with the individual words “marche” and “noir” or with “marche” combined with a different word such as “bon” (as in the registered mark THE BON MARCHE). However, none of those other meanings is relevant to a determination of the meaning of the unitary expression MARCHE NOIR”[54]. Therefore, in REPECHAGE and SECOND CHANCE, the court found that although second chance or a redo was one of the connotations that could be gleaned from REPECHAGE, that such a finding did not constitute enough weight to deny REPECHAGE registration. In CORDON BLEU and BLUE RIBBON, it was found, “that CORDON BLUE, while literally translated as BLUE RIBBON, would not be translated by the American public because the two terms create different commercial impressions, CORDON BLEU having been adopted into the English language and acquiring a different meaning than BLUE RIBBON”[55]. Here, the ordinary American or American interested in fine food was considered, whereas in the case of MARCHE NOIR and BLACK MARKET, the French-speaking consumer was considered. For MARCHE NOIR and BLACK MARKET, the former was denied registration because it was found that MARCHE NOIR was a literal translation for BLACK MARKET (much as CORDON BLEU and BLUE RIBBON are literal translations), but the court denied the applicants arguments that MARCHE NOIR could be construed in a variety of ways due to connotations (and even cited the trademark BON MARCHE, where marche is translated by the USPTO as bargain). These three cases, all involving French and English, seem to all place emphasis on differing aspects of the words and cases at hand, indeed, on different populations of speakers altogether. Furthermore, the TTAB noted in In re Thomas, “nor is this a situation where the mark would not be translated because of marketplace circumstances or the commercial setting in which the mark is used…In re Tia Maria, Inc., supra (finding it unlikely that a person who had purchased AUNT MARY’S canned fruits and vegetables from a supermarket would, upon dining at the TIA MARIA restaurant surrounded by Mexican décor and serving Mexican food, translate TIA MARIA into AUNT MARY and then mistakenly assume that both goods and services originated from the same source)”[56]. This line of justification was not cited for Continental Nut Company v. Le Cordon Bleu, though. In that instance, following the logic of In re Tia Maria, Inc., it could be seen as unlikely that a person who had purchased BLUE RIBBON (and, as attempted, CORDON BLEU) nuts from a supermarket would, upon taking culinary and hospitality management classes from LE CORDON BLEU (with the influences there of the ‘decorated chef’ reading of the term, rather than the blue ribbon literal translation) translate BLUE RIBBON (or even the same word, CORDON BLEU) into CORDON BLEU and then mistakenly assume that both goods and services originated from the same source.

It can be concluded from the cases presented thus far that the courts and the TTAB, in their application of the DoFE (which is at odds at times between them, at the expense of consumers and producers) fail to exercise any systematic justification for their decisions that holds up across cases, even for cases involving foreign words in the same language. The DoFE in its application, and in its inability to settle on a specific relevant population of consumers, fails to take into consideration the full complexity of accepted psycholinguistic theories as they relate to how people translate and encounter foreign languages.

3.4 | Fluency Thresholds

An added complication from a linguistic standpoint to questions of whether or not bilinguals will translate words in certain ways is the fact that being fluent or proficient in a language does not necessarily mean that one recognizes all of the words in their L2 or even their L1. As is noted by Luna and Peracchio, for almost all people the L1 enjoys a much greater lexical stockpile than does the L2, with clear vocabulary superiority.

Colin Baker and Sylvia Prys Jones (1997)[57], in the Encyclopedia of Bilingualism and Bilingual Education, note that the term bilingual “includes those who are highly proficient in both languages (across a variety of dimensions) to those whose dual language abilities are underdeveloped in both of their languages. In between are many variations and possibilities” (pg. 74). For those familiar with literature on bilingualism, one major problem becomes “defining precisely what are the thresholds of language proficiency” (pg. 75). There does not exist one clean or universally accepted definition of fluency. Even for speakers who can participate in extended conversations and may be colloquially referred to as being fluent in a language, there might still exist a gulf between a person who learned that language as their L2 versus a native or mother tongue speaker of that language, having it as their L1. The way that individuals who study Spanish in school in the United States encounter Spanish in the marketplace therefore (not a difficult situation to imagine) may be very different from how a mother-tongue speaker encounters Spanish in the marketplace and understands certain words and phrases. Even for individuals who are native speakers of a language, the ordinary consumer (as opposed to the sophisticated consumer) may well lack knowledge of the proper meanings of certain words in their L1. For example, Harley (2006)[58] notes that “For a long time while I was a teenager, I thought the word facetious was related to the word feces – during that time, for me, facetious was a fancy way of saying ‘full of shit.’ I had created a folk etymology” (pg. 93). Folk Etymology is one way of demonstrating that even when people are ‘fluent’ in a language, false etymologies can skew their understanding of the marks that they come into contact with.

[1] (Chaudhri, 2007)

[2] (Sutter Home Winery, Inc. v. Madrona Vineyards, L.P., 2005)

[3] Language taken from (Rest, 2006) also citing (Sutter Home Winery, Inc. v. Madrona Vineyards, L.P., 2005). Italics included by author.

[4] (Chaudhri, 2007, pg. 12)

[5] “The doctrine of foreign equivalents applies to non- English words that consumers are likely to translate into their English equivalents. Palm Bay Imports, 73 U.S.P.Q.2d at 1696. In Palm Bay Imports, the Fed- eral Circuit upheld the T.T.A.Bʼs holding that VEUVE ROYALE for sparkling wine and VEUVE CLICQUOT PONSARDIN and VEUVE CLICQUOT, both for champagne, were likely to be confused. However, the Court reversed the T.T.A.Bʼs holding of likelihood of confusion between VEUVE ROYALE for sparkling wine and THE WIDOW for wine…

In comparing VEUVE ROYALE with VEUVE CLICQUOT PONSARDIN and VEUVE CLICQUOT, the T.T.A.B. had found that “an appreciable number of purchasers” are unlikely to translate the marks into English. Veuve Clicquot Ponsardin v. Palm Bay Imports, Inc., Opp. No. 115,438, 2000 W.L. 21953664 (T.T.A.B. August 4, 2003). However, in comparing VEUVE ROYALE with THE WIDOW, the T.T.A.B. found that “[A]n appreciable number of purchasers in the United States” speak and/or understand French and will translate VEUVE ROYALE as ROYAL WIDOW…

On appeal, the Federal Circuit held that the T.T.A.B. was inconsistent in its application of the doctrine of foreign equivalents because “[a]n appreciable number of U.S. consumers either will or will not translate VEUVE into ʻwidow.ʼ” Palm Bay Imports, 73 U.S.P.Q.2d at 1696. It agreed with the T.T.A.B. that “it is improbable that the average American purchaser would stop and translate VEUVE into ʻwidow.ʼ” Thus, VEUVE ROYALE was not likely to be confused with THE WIDOW.” (Chaudhri, 2007, pg. 13).

[6] (United States Patent and Trademark Office, 2006)

[7] (Sutter Home Winery, Inc. v. Madrona Vineyards, L.P., 2005)

[8] (Chaudhri, 2007, pg. 13)

[9] (Chaudhri, 2007, pg. 13) cites both (McCarthy, 1984) and (United States Patent and Trademark Office, 2010) in noting, “the doctrine of foreign equivalents applies only to modern languages. Thus, words from languages such as Italian, French, Spanish, German, Hungarian and Polish should be translated into English. 2 J. Thomas McCarthy, McCarthy on Trademarks and Unfair Competition § 11.34 at 11-67 and 11-68. The Trademark Manual of Ex- amining Procedures (T.M.E.P.) uses the terms “language familiar to an appreciable segment of American consum- ers” and “modern languages.” T.M.E.P. § 1207.01(b)(vi). Thus, only words in modern languages which are under- stood and/or spoken by the purchasing public must be translated into their English equivalents.”

[10] (U.S. Census Bureau, 2000)

[11] (United States Patent and Trademark Office, 2006)

[12] (Chaudhri, 2007)

[13] (In re Tia Maria, Inc., 1975)

[14] See, for example, (Continental Nut Co. v. Le Cordon Bleu, 1974), as well as (In Re Sarkli, Ltd., 1983), discussed later.

[15] (Jacoby, 1998, pg. 2)

[16] (Jacoby, 1998, pg. 19)

[17] “Acquired distinctiveness, secondary meaning, fame, genericism, confusion and dilution do not apply to single individuals, but to aggregates, typically, the universe of purchasers and prospective purchasers of the product or service offering at issue” (Jacoby, 1998, pg. 22).

[18] (Jacoby, 1998, pg. 36)

[19] (Muehling, 1990)


[21] (Paradis, 2004)

[22] (GROOT & CHRISTOFFELS, 2006, pg. 196)

[23] (GROOT & CHRISTOFFELS, 2006, pg. 198)

[24] (Groot, 1992)


[26] Image (Figure 2) taken from (David Luna, WHAT’S IN A BILINGUAL’S MIND?: HOW BILINGUAL CONSUMERS PROCESS INFORMATION, 1999, pg. 310)

[27] (Woulter Dyuck, 2008)

[28] Luna and Peracchio (1999) refer to this as well, calling it a Language-Specific Schema. For example, using an English schema and a Spanish schema,“for cognates and concrete words, these schemas are very similar -in some instances, maybe even the same. However, in the case of non-cognates and abstract words, it is very likely that the schemas will be substantially different. Thus, if an individual is presented with the word “love”, a set of concepts will be automatically activated which will be very different from the concepts activated when the same individual is presented with the word “amor”. This assertion is consistent with prior research (De Groot 1992; Keatley, Spinks, and De Gelder 1994; Kolers 1963)” (pg. 310).

[29] (Woulter Duyck, 2008, pg. 103)

[30] Taken from (Woutler Duyck, 2008, pg. 103)

[31] (Woulter Duyck, 2008, pg. 110)

[32] (Woulter Duyck, 2008 pg. 111)

[33] (David Luna, 2002, pg. 574)

[34] (David Luna, 2002, pg. 574) cite (Dufour & Kroll, 1995; Kroll & de Groot, 1997)

[35] (David Luna, 2002, pg. 574)

[36] Image taken from (David Luna, 2002, pg. 575). In the model, “connections between words in different languages made at the lexical level are referred to as word associations or lexical links, whereas the connections in memory between lexical representations in either language and the meanings they represent are referred to as conceptual links…the model specifies a stronger lexical link from the individual’s second language (L2) to his or her first language (Ll) than from the individual’s Ll to his or her L2. This is a residual effect from the second-language acquisition process in which individuals begin learning words in their L2 by relating them to words in their Ll. Hence, words in the bilingual individuals’ L2 are closely associated with words in their Ll” (pg. 575). Note: L1 and L2 are used here to refer to the first language an individual has learned, and their second language.

[37] (David Luna, 2002, pg. 575) cites (Dufour & Kroll, 1995; Kroll & de Groot, 1997)

[38] (David Luna, 2002, pg. 576)

[39] (Jaime Noriega, 2008)

[40] Family, Friends, Home, or Homeland

[41] Jaime Noriega, 2008, pg. 82)

[42] (Catford, 1965)

[43] (Language Marketplace, 2010)

[44] (In Re Sarkli, Ltd., 1983)

[45] Image taken from (Repechage Makeup Products, 2010)

[46] (In Re Sarkli, Ltd., 1983)

[47] (In Re Sarkli, Ltd., 1983)

[48] Image taken from (Le Cordon Bleu Launches a New Micro Site for Cooking and Travel Enthusiasts , 2009)

[49] “What does “Cordon Bleu” really mean to the American public and what does “Blue Ribbon” mean? The French term is not so unusual to the American public because it is defined in American Dictionaries. Funk & Wagnalls New Standard Dictionary of the English Language defines that term as “the blue ribbon of the order of the Holy Ghost, the highest order of the old French monarchy”, and as “a person regarded as entitled to a badge of eminent distinction; specif., a first class cook, particularly a woman cook.” Webster’s Third New International Dictionary, 1965, similarly defines the term “Cordon Bleu” and indicates the applicability of said term to a cook of great skill. The term “Blue Ribbon” figuratively refers to an honor or award gained for prominence and, literally, signifies a blue ribbon awarded the first place winner in a competition. See: The American College Dictionary, 1970; Webster’s World Dictionary of the American Language 1966; Webster’s Seventh New Collegiate Dictionary, 1963; Funk & Wagnalls New Standard Dictionary of the English Language, supra; and Webster’s Third New International Dictionary, supra.

On the basis of dictionary definitions, we are of the opinion that “Blue Ribbon” and “Cordon Bleu” would not have the same significance to the American public and that the marks “BLUE RIBBON” and “CORDON BLEU” create different commercial impressions” (Continental Nut Co. v. Le Cordon Bleu, 1974).

[50] (Continental Nut Co. v. Le Cordon Bleu, 1974)

[51] (Le Cordon Bleu Launches a New Micro Site for Cooking and Travel Enthusiasts , 2009)

[52] (United States Patent and Trademark Office, 2006)

[53] (United States Patent and Trademark Office, 2006)

[54] (United States Patent and Trademark Office, 2006)

[55] (United States Patent and Trademark Office, 2006)

[56] (United States Patent and Trademark Office, 2006)

[57] (Colin Baker, 1997)

[58] (Harley, 2006)