No “Divided Public Opinion” with Regard to Likelihood of Confusion in Trademark Law

Headnote

In determining likelihood of confusion, the governing factor is the opinion of the averagely informed, suitably attentive, and reasonably prudent consumer of the goods or services in question. The presumption of a divided public opinion is therefore not compatible with the term “likelihood of confusion” as a legal concept. A different assessment is only justified as an exception if the opposing marks address different groups of the public, which can be objectively differentiated, for example, the general public and specialized markets or different language groups. In such cases, it is sufficient for the affirmation of an infringement if there is likelihood of confusion in one of the groups of public addressed.

Note

According to the conventional opinion, likelihood of confusion in trademark law is a normative concept. In principle, therefore, the question is not whether actual confusion occurs within the relevant group. This can be seen for example in the fact that likelihood of confusion is presumed to be all the more likely if the older mark is well-known, although in fact it is significantly less probable that a particularly well-known trademark will be confused with a similar third-party mark than that this will occur with regard to a less well-known mark. Nevertheless, a well-known mark – from a normative point of view – earns greater protection against similar third-party marks than a less well-known mark. This is taken into account in the case law at the level of likelihood of confusion.

Normative vs. actual likelihood of confusion

The principle described above, however, does not apply without restrictions. Rather, jurisdiction keeps sometimes arguing with individual circumstances that relate to an actual likelihood of confusion. According to the case law of the German Federal Court of Justice, for example, the possibility of likelihood of confusion under trademark law is excluded if one of the conflicting signs has a concrete textual meaning that is known to the public and that the other sign does not also have.

The “neutralization theory” of the European Court of First Instance also finally relies on the actual point of view of the group of the public being addressed, when it rejects likelihood of confusion in cases in which, for example, the visual similarities that exist are offset by sufficient acoustic differences between the signs. On the other hand, this will not apply if the goods in question are predominantly purchased on the basis of their appearance – another aspect based on actual rather than normative likelihood of confusion.

Previous opinion

In case of a divided public opinion it is sufficient for an infringement if there is a likelihood of confusion only among a part of the relevant public

Accordingly, in the case of a divided public opinion, it was considered sufficient for the breach of trademark for a considerable portion of the group of the public in question to be exposed to likelihood of confusion (see, for example, Federal Court of Justice GRUR 2005, 264, 266 – Das Telefon-Sparbuch, or Federal Court of Justice GRUR 1992, 110, 111 et seq. - dipa/dib).

Realignment of the legal precedentof the German Federal Court of Justice

The Federal Court of Justice is deviating from this previous approach – without drawing sufficient attention to the fact – in its new decision AMARULA/Marulablu and is now of the opinion that the deciding factor is not how a significant portion of the group of the public been addressed assesses the conflicting marks, but rather the point of view of the (normative) average consumer (Federal Court of Justice, WRP 2013, 778, 784 [paragraphs 64 et seq.] - AMARULA/Marulablu). In concrete terms, the decision relates directly to the evaluation of the distinctive nature of the older mark only, but nothing else can apply to the other elements of likelihood of confusion (especially to the assessment of the similarity of the colliding marks). Exceptions from the new rule will apply only if individual groups can be differentiated within the relevant group of the public (e.g., consumers and specialists, residents and foreigners, etc.). In that case, an average level of understanding can then be defined within each group formed in this way and it is sufficient for there to be likelihood of confusion within one of the groups for trademark infringement to be present. Differentiation criteria that are purely subjective such as the presence or lack of certain prior knowledge among consumers are not sufficient for the formation of such groups though.

Consequences

Although the Federal Court of Justice refers, in its new line of argument, to the consumer model developed by the European Court of Justice, it should not be overlooked that there is a certain tension between its strictly normative reliance on the average consumer and the fact that otherwise purely factual circumstances are taken into account – including in European Court of First Instance case law – in the assessment of likelihood of confusion. In addition, the Federal Court of Justice’s new point of view makes it more difficult to act against certain ways of using a mark that are characterized by the intentional creation of an (actual) likelihood of confusion with an older third-party trademark by exploiting a specific existing expectation of a group of consumers that is formed based on a certain prior knowledge. For example: A certain product that only some of the consumers are aware of has initially been sold for a long time under a trademark that is similar to an older mark in such a way that there is likelihood of confusion. The manufacturer of the infringing product then changes its mark in response to a warning such that his previous customers still recognize the old infringing mark as a result of their knowledge of the previous mark and product design, while other outside consumers perceive the new mark differently (e.g., they read the word element of the mark in a different order). As a result, the new overall sign is no longer similar in such a way as to cause confusion with the older mark from the point of view of this outside “average consumer”. In this example, according to the new case law of the Federal Supreme Court, a claim for trademark infringement would probably be rejected, although the mark infringer is actually continuing his previous trademark infringement, at least with respect to the previous purchasers of his product.

Conclusion

As long as the Federal Court of Justice sticks to its new case law, in the future it will have to be expected that the evaluation of likelihood of confusion between two marks will generally be based on the normative point of view of the average consumer. Any actual likelihood of confusion occurring only within certain groups of consumers will be considered irrelevant even if these groups make up a not entirely insignificant portion of the consumers overall. In certain situations this will lead to unsatisfactory results.