Assessing the distinctiveness of an EU trademark can be tricky. For an EU trademark to be registrable, it needs to be distinctive in all parts of the EU. If the mark is not distinctive even in one part of the Union (for example, one member state or even one part of a member state), the mark will not be registered.

The first obvious consequence is that the distinctiveness is assessed in all EU languages. If the mark is non-distinctive or descriptive for example in Romanian or Bulgarian languages, it will be refused.

Things get more complicated because descriptiveness could also relate to smaller entities than just states. For example, the EU court decided in 2012 that the Catalan word “ESPETEC” could not be registered as an EU trademark for sausages, because it means “fuet” (a type of Catalan sausage) in Spanish.

“Espetec” is a word that does not have any meaning in any official EU language. Yet, the mark was considered descriptive in a language that is spoken in the EU, i.e. part of Spain. There is a significant population in Spain that speaks Catalan, nearly 10 million.

When assessing the distinctiveness of an EU trademark, you need to be mindful of all the languages spoken in the Union. This includes not only the 24 official EU languages, but also minority languages like Basque and Catalan, or languages of non-member states if these languages are understood in the EU. 

EU languages – these are always considered

The mark is automatically refused if it has a descriptive meaning in any official EU language. This applies regardless of the size or population of the country where the language is spoken.

Non-EU languages – these may be relevant as well

For non-EU languages, registration is refused if there is clear evidence that the language is understood by “non-negligible” part of the relevant public in a part of the EU. Further, this objection applies only to commonly used or understood words. Slang words and dialects of these languages are usually not commonly understood.

The EUIPO does not maintain a list of languages that are understood in non-negligible parts of the EU. That would be admittedly difficult, taking into account that the situations change. For example, just this year there have been approximately 6 million Ukrainian refugees arriving in the EU. There are about 10 EU countries that have a smaller population than that, making Ukrainian (and Russian) more spoken languages than many official EU languages. Further, the Baltic states had more than a million Russian speakers before the war in Ukraine. 

Languages such as Ukrainian and Russian may be relevant in assessing the distinctiveness of an EU trademark, taking into account that they are quite widely understood at least in part of the EU. 

To illustrate this, the EU court refused to trademark медведь in 2017 for meat products. медведь means “bear” in Russian. The refusal was based on the fact that there is a sizeable Russian speaking population in the Baltic states. The EUIPO has also rejected applications that have Chinese or Arabic words. As said, objections will mainly be raised if the descriptive word itself is a common word in that non-EU language. Such is the case with fuet/sausage (Espetec) or bear (медведь). 

So what should you do?

Assessing the distinctiveness of an EU trademark is difficult, because there are no clear rules on which languages and what kinds of words are relevant. This applies in particular to non-EU languages, such as Catalan, Basque or Russian. There seems to be a lot of room for argumentation one way or the other. 

According to Wikipedia, these are the “significant” non-EU languages spoken in the Union: Russian, Luxembourgish, Turkish, Sami languages, Basque, Catalan, Galician, Breton, Romani, Albanian, Serbian, Ukrainian, Occitan, Arpitan, Scottish Gaelic, Scots, Corsican, Sardinian, Armenian, and Georgian.

According to EUIPO’s examination guidelines, the refusal takes place if there is convincing evidence that the word is commonly understood by a non-negligible part of the population in one or more EU member states. This evidence could be something as random as the EUIPO’s examiner’s personal knowledge of the language in question. This makes it very difficult for you (and us) to assess the distinctiveness of a potential EU trademark.

If you purposefully import generic words from the above-mentioned languages to be used as brand names (trademarks) in the EU, you should be particularly cautious. If the imported word is not a common word or is a slang word, it should be registrable. Commonly used everyday words however might run the risk of being detected and refused if they somehow describe the products or services in question. 

It seems to be a matter of luck to some degree whether non-EU language words can be registered as trademarks in the EU or not. Even if it is accepted originally by EUIPO, you should remember that the validity of the registration can be challenged even afterwards. In fact, in the above example mark ESPETEC was refused only after another party challenged EUIPO’s examiners’ decision to accept the application.