作者:蔡億達律師(本會副秘書長)

Generally, the likelihood of confusion relates to the distinctiveness of the marks in issue. The lower the distinctiveness, the harder the likelihood of confusion to establish. A mark with a fanciful or coined term is per se distinctive. On the other hand, a mark with only descriptive, generic, or highly suggestive terms is less distinctive, i.e., a weak mark. However, an originally weak mark may become distinctive through use. The question is will the determination of the likelihood of confusion be affected by how the mark becomes distinctive? Here come the cases.

OATLY case, namely OATLY AB v Glebe Farm Foods Limited [2021] EWHC 2189 (IPEC), provides a detailed process for determining whether the likelihood of confusion exists between weak marks. The 味全總舖師 case in Taiwan shares similar facts with the OATLY case, but the verdict is completely different. The differences in the two cases are worth further discussion. Therefore, this article will focus on the comparison of the two cases and analyzes them for our readers.

The OATLY case is briefly introduced as follows. OATLY is a trademark owned by OATLY AB registered for dairy substitutes and oat-based drinks. OATLY products have a high market share in the UK, which establishes a highly distinctive character and reputation for the “OATLY” brand. OATLY AB took Glebe Farm Foods to the court, alleging that Glebe had infringed the “OATLY” trademark by selling oat milk called “PUREOATY”. The marks use by both parties shows as below.

The Intellectual Property Enterprise Court (the UK court), which is a specialist court part of the Business and Property Courts of the High Court of Justice, listed guidance with regard to the approach to assessing a likelihood of confusion as follows:

  1. The likelihood of confusion must be appreciated globally, taking account of all relevant factors.
  2. The likelihood of confusion is assessed by the average consumer of the named goods who, although being reasonable and prudent, does not compare trademarks side by side and often rely upon her imperfect recollection to purchase. In other words, she perceives a trademark as a whole, getting an overall impression, and does not analyze its various details.
  3. In a general matter, the visual, aural and conceptual similarities of the marks must be compared by their overall impressions. However, if all other components of a complex trademark are negligible, we may make the comparison solely by the dominant elements. Moreover, if an element corresponds to a registered trademark, it may retain an independent distinctive role in a composite mark even though it does not constitute a dominant element of the mark.
  4. A lesser degree of similarity between the goods may be offset by a greater degree of similarity between the marks, and vice versa;
  5. The lower the distinctiveness, the harder it is to establish similarity and vice versa. There is a greater likelihood of confusion if the registered trademark has a highly distinctive character, either per se or by use of it. If the similarity between the respective trademarks is a generic element in relation to the goods for which they are registered, that points against there being a likelihood of confusion.
  6. If the association between the marks causes the public to wrongly believe that the respective goods come from the same or economically-linked undertakings, there is a likelihood of confusion.

Based on the guidance listed before, the UK court compared “OATLY” with “PUREOATY” and found that there is no visual similarity. Indeed they share the same word “oat”. However, the marks have different lengths in letters, and also the additional word “PURE” at the start of the mark detracts the average consumer significantly from the visual similarity. There is no aural similarity for the same reason.

As to the conceptual similarity, OATLY AB argued that “PURE” is merely a descriptive word and thus the consumer will only focus on “OATY”, which is highly similar to the plaintiff’s “OATLY” trademark. The UK court disagreed, stating that “OATLY” is also descriptive in relation to a generic word “oat”, and the suffix “ly” does not change the literal meaning of the word. On this basis, any conceptual similarity between the marks is based on their common reference to “oat”, not to “OATLY”. Namely, the conceptual similarity of “OATLY” and “PUREOATY” is attributable to their descriptiveness. The UK court cited several precedent cases which emphasized the difficulty in establishing any likelihood of confusion if the similarity lies in a commonality of descriptive elements. In Reed Executive Plc v Reed Business Information Ltd [2004] EWCA Civ 159 at [83]–[84] the Court of Appeal said:“…where you have something largely descriptive the average consumer will recognize that to be so, expect others to use similar descriptive marks and thus be alert for detail which would differentiate one provider from another.”

Generally, the overall impression of a mark is mainly created by its dominant elements, except the dominant elements are descriptive. Under this circumstance, the consumer will pay more attention to any details which may help her to distinguish between brands. In such cases, subordinate elements of a mark will have more weight than they would have been in general cases with regard to the determination of the likelihood of confusion. For these reasons, the UK court gave more weight to the word “PURE” and found that there is no likelihood of confusion between the “PUREOATY” and “OATLY”, accordingly rejecting OATLY AB’s claim.

There is a very similar case named 108年度行商訴字第70號行政判決 determined by the Intellectual Property and Commercial Court in Taiwan (the Taiwan court). Here goes the fact.

The plaintiff owns a “總舖師” trademark registered in dumplings. The intervener filed an application for a mark “味全總舖師” sought to be registered in dumplings. The plaintiff filed an opposition to the registration for similarity, and the Taiwan Intellectual Property Office (TIPO) ruled against the opposition. The case went to the Taiwan court.

The non-dispute facts go as follows.

  1. “總舖師” is a common honorific for lead chefs and thus a descriptive word in food-related goods. On application, the plaintiff disclaimed the entire “總舖師” in order to be registered.
  2. The marks share the same dominant element “總舖師” and the same category of goods.
  3. In Taiwan, “味全” is a famous company name and also a well-known trademark which has a long and profound history in the field of food and drink.

The issue here is how much weight the court should give to the word “味全” in the determination of the likelihood of confusion.

TIPO argued that there is no likelihood of confusion because “味全” mark has a distinctive character and national-wide reputation. However, the Taiwan court ruled in an opposite way. “總舖師” establishes its highly distinctive character by use for decades. Both parties’ marks contain the same dominant element“總舖師”. Although the alleged infringing mark “味全總舖師” consists of five words while the alleged infringed trademark“總舖師”consists of three, “味全” is the name of a company and has no semantic meaning with “總舖師”. Consumer may divide the alleged infringing mark into two parts: “味全” and “總舖師”. Thus, high similarity exists. Accordingly, the Taiwan court ruled to revoke the registration of “味全總舖師”.

In the “OATLY” case and the “總舖師” case, the alleged infringed trademarks are both descriptive words and were weak at the beginning, but they became distinctive through use for decades. The alleged infringing marks are both composite mark sharing the same dominant element of the alleged infringed trademarks. Compared with “PUREOATY”, “味全總舖師” has two advantages in the judgment of the likelihood of confusion. First, the word “總舖師” had been disclaimed, while “OATLY” had not. Second, “味全” is a fanciful word and a well-known trademark, while “PURE” is merely a descriptive word. However, the verdicts go the opposite way. The main difference is that the UK court gave less weight to the dominant element “OATLY” due to its descriptive nature, while the Taiwan court gave more weigh to “總舖師” due to its high distinctiveness obtained by use.

As mentioned above, higher distinctiveness means a higher likelihood of confusion. But what if the dominant element of the trademark is weak, i.e., it is generic, descriptive, or highly suggestive of the named goods, but it afterward becomes highly distinctive through uses? How much weight should we give to the dominant element? That is the key question in the two cases. The UK court’s decision and the process of reasoning are more thoughtful. Despite the high market share of“OATLY” brand, the UK court chose to give it less weight for two reasons: the connection of “OATLY” and “PUREOATY” lies in their common reference to “oat”, not to “OATLY”; and the consumer will pay more attention to subordinate elements which may help her to differentiate brands when the dominant element “OAT” is descriptive. Likewise, the federal court in the US also held that the descriptive nature of the shared wording weighed against a finding of similarity. Although it does not mean that the descriptive words play no role in creating confusion, the public at large do rely more on the non-descriptive portions of marks to distinguish them, See, e.g., In re Bed & Breakfast Registry, 791 F.2d 157, 159 229 USPQ 818, 819 (Fed. Cir. 1986) and Nat’l Data Corp., F.2d at 1060, 224 USPQ at 752.

In these decisions, we may also see a concern about public policy and fair trading–to avoid a trademark owner monopolizing a generic term in the named goods, i.e.,“oat” in oat milk. The same concern had been fully discussed in Patent and Trademark Office v. Booking.com B. V., 591 U.S. ___ (2020). In this case, the Supreme Court in the US held that a generic term (here, “Booking.com”) is eligible for registration if it would convey to consumers a source-identifying characteristic. The USPTO argued that if the Supreme Court allowed a generic term to be registered, it would hinder competitors to use the term. The Court responded “in assessing the likelihood of confusion, courts consider the mark’s distinctiveness. The weaker a mark, the fewer are the junior uses that will trigger a likelihood of consumer confusion…When a mark incorporates generic or highly descriptive components, consumers are less likely to think that other uses of the common element emanate from the mark’s owner…And even where some consumer confusion exists, the doctrine known as classic fair use protects from liability anyone who uses a descriptive term fairly and in good faith to describe her own goods.” Arguably “Booking.com” has a worldwide reputation, but this fact only grants registration for the term, not exclusive uses of it. So the answer to the question above is very clear.

The determination of the existence of a likelihood of confusion is never an easy job because “likelihood”, “similarity” and “confusion” are legal concepts as uncertain as “reasonable”. Without a clear standard established by consistent judgments, it’s hard for practitioners to accurately predict the outcome, not to mention lay persons. The standard need to be established by the court through a detailed, deliberate process of reasoning, thereby reducing disputes and safeguarding the rights and interests of the trademark owners. Therefore, these cases introduced above are worth taking as reference for practitioners and the Intellectual Property and Commercial Court in Taiwan.

The Determination of the Likelihood of Confusion between Weak Trademarks-Based on the OATLY Case in the UK and the 味全總舖師 Case in Taiwan