I don’t know why, but only just a couple of days ago (on 1 September to be precise) the Dutch judiciary website www.rechtspraak.nl suddenly published a decision by the Court of Appeal The Hague (Kiers-Becking, Schaafsma and Blok) dated 28 February 2017 in a trade mark case which has attracted quite some attention in the Netherlands. It may be - one is never sure about this - that this late publication is caused by the involvement of witches and female spirits who are at the heart of the matter which needed to be decided, be it that also a more tangible and typical Dutch product was involved: cheese (kaas). 

To be honest the trade and the marketing of said product - and this may be true also for the case this post relates to - can hardly be fathomed without reading the great writer Willem Elschot’s 1933 masterpiece ‘Cheese’, the tale and tribulations of shipyard clerk Frans Laarmans, who wants to escape his boring, ultra gray work environment and exchange it for a high end entrepreneurial position, i.e. cheese merchant. A position which is far from high end. His trade goes from bad to worse. ’Say cheese’ is most certainly not a saying which could beg a smile on Laarman’s face, you couldn’t even enforce it if you begged him. Interestingly the novel starts off with a page titled ‘Elements’, ‘Kaas’ (in italics) being one of those elements listed, followed by the different manifestations of ‘Cheese’ in the novel: ‘Cheesedream. Cheesemovie. Cheesebusiness. Cheeseday. Cheesecampaign. Cheesemine. Cheeseworld. Cheeseship. Cheesetrade. Cheeseprofession. Cheesenovel. Cheese-eater. Cheeseball. Cheesetrader. Cheesetrust. Cheesedragon. Cheesemisery. Cheesewill. Cheesefantasy. Cheesewall. Cheesematter. Cheesewagon. Cheesetesting. Cheesetower. Cheesewound.’

Although these manifestations do not play a role in the tale which was taken to Court, after reading the decision of the Court of Appeal (again) and the opposition decision of the Benelux Office for Intellectual Property (Van Hoey, Janssen, Veeze), which was under appeal, I dare to put forward that if Frans Laarman was on the verge of starting a trade mark attorneys or lawyers career, he would certainly think twice, and cast his mind back to his cheese adventure.

On first sight the case looks pretty simple (I don’t use that word often). On 26 January 2015 Dutch based cheese company FanofineFood B.V. (FFF) registered the word mark WITTE WIEVENKAAS and two figurative marks including WITTE WIEVENKAAS as a part, amongst other for cheese products (class 29 and 30).

On 30 March 2015 Levola Hengelo B.V. (Levola) started opposition proceedings against the trade mark registrations before the Benelux Office for Intellectual Property (BOIP). Levola based the oppositions on its 2010 Benelux trade mark registration (in class 29 and 30) for (the word mark) ‘HEKSENKAAS’ (for a proper understanding, it could be translated in English as ‘WITCHES CHEESE’).

Levola puts forward the following grounds: 

(i) the trade mark ‘HEKSENKAAS’ is a trade mark which has inherent distinctiveness and accordingly enjoys a large scope of protection. It is a phantasy name which does not refer to type of goods for which the trade mark is registered nor is it descriptive for said goods.

(ii) the trade mark is well known. Also on that ground the trade mark enjoys a large scope of protection. Because the trade mark is well known a risk of confusion should be assumed more rapidly.

(iii) from a conceptual perspective both trade marks are totally identical. The meaning of the word element ‘KAAS’ is identical (cheese = cheese), while this word element is located at the same place in both trade marks. The word ‘heks/heksen’ (witch/witches) are common words in the Dutch language which have a simple meaning. Levola argues with all sorts of exhibits that ‘WITTE WIEVEN’ (witte = white, wieven = a sort of dialect word for ‘wijven’ = wives) is used as referring to a sort of witch of ghosts of witches. Therefore the element ‘HEKSEN’ and ‘WITTE WIEVEN’ have a similar meaning, albeit a strong conceptual similarity.

(iv)the goods for which ‘WITTE WIEVEN’ has been registered are identical to those for which ‘HEKSENKAAS’ is registered, or at least similar to a high degree. Above that those goods aim at the general, non expert consumer public which has a low level of attention. Accordingly similarity causes confusion more quickly.

(v) the combination of these circumstances is sufficient to accept a risk of confusion.

The BOIP rejects the opposition. After having established that there is not a visual and auditive similarity between the trade marks (which is undisputed by the parties - see above), the BOIP has a closer look at the alleged conceptual similarity between the trade marks. Taking into account the CJ EU decisions in case T-256/04 (Respicur), T-146/06 (Aturion), T-189/05 (Galvalloy) the BOIP puts first that a consumer who perceives a word mark (which consists of more words) will dissect this word in word elements which will have a concrete meaning for him or resemble words he already knows. The invoked trade mark will accordingly be perceived as a composition of the elements ‘HEKSEN’ and ‘KAAS’ and the opposed trade mark as a combination of ‘Witte Wieven’ and ‘KAAS’.

The word element ‘KAAS’ denotes in both trade marks a dairy product made from the curd of coagulated milk, in color and substance variating from its method of preparation, the degree of maturity and the like. Thus the BOIP finds that KAAS describes the type of the goods concerned. The element ‘KAAS’ has a clear concrete and descriptive meaning with respect to the involved goods. Moreover the BOIP remarks that in general the public will not perceive the descriptive element of a combined trade mark as the distinctive and dominant element of the overall impression the trade mark evokes.

Above that the BOIP finds that there is no conceptual similarity between ‘HEKSEN’ and ‘WITTE WIEVEN’. The word ‘HEKS’ refers to someone who has witchcraft, a kind of magician, someone who is able (with help of the devil) to cause disaster to others, a girl which is bright or roguish, or someone who works with magic powers in nature (the BOIP refers for its ‘HEKS’ findings to the famous Dutch dictionary Van Dale’s ‘Groot Woordenboek van de Nederlandse Taal’).

The element ‘WITTE WIEVEN’ however has a different meaning. Citing from the same Dutch dictionary with regard to the word ‘wijf’ the BOIP holds that it means - I translate the Dutch citation - ‘(German mythology) white wives or wieven, sylfees, often regarded as malicious, also a wise woman, inhabitants of holes and other hideaways, which they leave now and then to search for fortune, misfortune or to predict future events and give indications about the whereabouts of stolen or losts goods, white damsels’. Therefore both element are conceptually different. ‘HEKS’ refers to a person with witchcraft, ‘WITTE WIEVEN’ refers to fantasms. That ‘WITTE WIEVEN’ may be linked ‘HEKSEN’ as defended by Levola, does not mean that ‘WITTE WIEVEN’ are witches and thus have an identical meaning conceptually. This is also refuted by the precise meaning of both elements, as derived from the dictionary. Above that the BOIP finds that he who knows the ‘WITTE WIEVEN’ concept shall know that it does not have anything to do with witches, whilst he who doesn’t shall not give any meaning to the concept. With regard to the goods of class 29 and 30 the elements ‘HEKSEN’ and ‘WITTE WIEVEN’ do not have any meaning, contrary to the element ‘KAAS’, thus ‘HEKSEN’ and ‘WITTE WIEVEN’ being the dominant elements. Those dominant elements are different, not similar.

Since the trade marks are different, the BOIP sees no ground for a comparison of the goods, because there can be no risk of confusion if there is no similarity between the trade marks.

Levola disagrees and appeals to the Court of Appeal The Hague, which sets asides the decision of the BOIP and upholds the opposition. The ‘pièce de résistance’ of the Court of Appeal’s decision appears to be the 360 which the Court makes with regard to the similarity between the trade marks.

The Court finds first of all that FFF did not challenge sufficiently well-founded enough that in general the concept/word ‘heksen’ in the Benelux is perceived as denoting female magic creatures who possess witchcraft, which they mainly use for negative purposes, who inspire fear in someone and who live mainly isolated. That witches according to a Wikipedia page - put into evidence by Levola - can also be men, is not enough to remove the general prevailing opinion about witches.

Apparently in appeal Levola came up with pictures, folk tales and a description on Wikipedia, from which the Court of Appeal derives that a part of the public perceives ‘Witte Wieven’ as mainly isolated living female magic creature with mainly malicious intentions, who incite fear and therefore as a kind of witch. 

With regard to the part of the public which perceives the words ‘wit wief’ as meaning a mainly malicious sylph with supernatural gifts, in which ‘wief’ refers to the female sex, the Court of Appeal finds that the meaning of said words is so close by the meaning of the word ‘heks’ that there is also a conceptual similarity: after all both are supernatural magical female figures with a negative connotation, which is such a special category of creatures that even then there is a conceptual similarity, because the public perceives both the witch as the ‘witte wief’ as falling under this exceptional category. [If I am correct the decision however does not contain any evidence for his mindset of the public]

The Court of Appeal ignores the statement of FFF that a large part of the public is not familiar with the words ‘witte wieven’ and accordingly will not attribute any meaning to these words. According to the Court of Appeal to conclude that there is a conceptual similarity it is not necessary that the entire relevant public knows about the meaning of the words used in the involved trade marks (‘HEKSEN’ and ‘WITTE WIEVEN’). It is sufficient - for the Court of Appeal - that only a substantial part of the public - is familiar with the similar meanings. That a part of the public may not know this meaning, does not alter the fact that there can be a conceptual similarity.

However the Court appears to be not sure that this is the argument that FFF wants to put forward. Therefore the Court evaluates also a sort of what if argument: if FFF would have intended to challenge that a substantial part of the relevant public is familiar with the meaning of the words ‘witte wieven’, the Court rejects the argument as unsatisfactory founded, in the perspective of the exhibits on which Levola based its statement about the meaning of the words ‘witte wieven’. The Court also takes into account that the average informed, cautious, normal consumer will attach a meaning to words which belong to the common parlance, except for special circumstances, which FFF did not put forward.

Furthermore the Court finds that Levola has brought forward that also if someone is not familiar with the supernatural, magic characteristics that ‘witte wieven’ have in folk tales, this person at least will know that the words ‘witte wieven’ refers to fictitious female creatures and that these words have a negative connotation, and that both trade marks will be understood - considered in their entirety - as a cheese which is made for, by and/or from a fictitious female creature with a negative connotation. Also in this perspective the Court finds a conceptual similarity, which so holds the Court FFF did not  deny well-founded.

The Courts finally takes into account that the figurative elements added to the Levola trade marks strenghten the conceptual similarity. Especially the bats have this effect, because bats, as Levola put forward undisputed, are perceived as attributes of witches and are associated with doom.

Last but not least the Court also holds that there is a small visual and auditive   similarity, because in both trade marks the word ‘KAAS’ and the letters ‘EN’ - ‘HEKSENKAAS’ - ‘WITTE WIEVENKAAS’ - because both words are plurals, and both prononounced with a dumb e-tone. The Court clarifies that in general there will not readily be a risk of confusion if the only similar element of both trade marks are descriptive and the other elements of the trade marks are not similar (the Court stresses that this is not the case here), but the risk of confusion is different from the question of similarity.

After establishing that the involved goods are identical and/or similar the Court finds no difficulty to find a risk of confusion: the signs are similar, the goods are identical or similar, thus there is a risk of confusion. For the Court all the more, because of the unicity of the HEKSENKAAS trade mark, because - as Levola has put forward undisputed - in the involved classes no other products existed for which a trade mark is used that combines a descriptive sign and a fictitious creature. That FFF put forward that such trade marks are registered is, does not alter this, so concludes the Court without any further clarification.

I don’t know exactly why, and I don’t say that the decision of the Court of Appeal is wrong or right, but I find the findings of the Court with regard to the conceptual similarity a bit of - I think I may say it in this context - witchcraft. I think this is caused by the lack of reflection on the reasoning by the BOIP with regard to the lack of conceptual similarity.

With regard to the ‘ENKAAS’ findings - visual and auditive similarity - one can not deny that this word element is a part of both trade marks, but it needs to be noted that apparently before the BOIP it was undisputed (in confesso) that there was not a visual and auditive similarity at all. But what has been undisputed in first instance, is not necessary from the table in appeal. But it feels a bit as a trouvaille.

Anyway as I postulated in the beginning Elschot’s novel proves - as Frans Laarmans experienced - that cheese can be a product of hardship. This case proves that this also applies for cheese trade marks, especially apparently if some abracadabra is involved.