Case C-163/16, Louboutin – shoe soles, signs solely comprising a shape, and judicial cobblers

If a red shaped-sole gives substantial value to a shoe, then is a trade mark registration invalid because EU trade mark legislation prohibits the registration of signs where shape gives substantial value to the goods?

Christian Louboutin is a shoemaker. He has registered rights under Benelux law over the shoes he makes. Consequently, he can exert his rights against companies that trade in Holland.

Mr Louboutin took the view that a Dutch chain of shoe shops, trading as Van Haren, was selling a type of shoe which looked remarkably similar to one of the shoes he had designed. Consequently, he believed the Dutch shop (and its website) was infringing his Benelux rights.

However, Van Haren denied doing anything wrong. It took the view that Louboutin had no legal rights under Benelux law – for despite Louboutin having registered his mark, the mark was not legally valid. That is to say, Van Haren relied on an important exception to Benelux law which does not allow the registration of trade marks where a sign is ‘a shape … which gives the product substantial value’. Thus, without a valid legal right, there could be no question of infringement by Van Haren.

The success of Van Haren’s legal ‘invalidity’ argument rested on the Benelux Convention on Intellectual Property (Trade Marks and Designs).

Article 2.1(2) of the Benelux Convention states:

Signs capable of constituting Benelux trade marks[:]

2. However, signs solely comprising a shape which is imposed by the very nature of a product, which gives the product substantial value, or which is necessary for obtaining a technical result, may not be regarded as trade marks.

In essence, Van Haren believed that Louboutin’s mark was but a sign comprising of a shape that gave the product substantial value and so was not capable of being registered as a mark.

Louboutin disagreed with Van Haren’s interpretation of the Benelux Convention and consequently sued Van Haren. The dispute was heard by judges at the District Court in The Hague.

At The Hague District Court
Having heard the parties, the Dutch judges indicated that they were minded to make a preliminary reference to the CJEU. Legal uncertainty flowed from the correct interpretation of Article 2(1)(2) of the Benelux Convention and the corresponding provision in the EU’s trade mark Directive (First Council Directive 89/104/EEC relating to trade marks (OJ 1989 L40/1), as codified by Directive 2008/95/EC (OJ 2008 L299/25)).

The particular problem was the wording of Article 3 of the EU’s codified Directive, which in in the English version states:

Grounds for refusal or invalidity
1. The following shall not be registered or, if registered, shall be liable to be declared invalid:

(e) signs which consist exclusively of:
(i) the shape which results from the nature of the goods themselves;
(ii) the shape of goods which is necessary to obtain a technical result;
(iii) the shape which gives substantial value to the goods;

Thus, the judges understood Van Haren to claim that Louboutin’s trade mark was invalid in so far as it was the shape of Louboutin’s shoe which gave it substantial value.

The judges understood Louboutin also to be focusing on the wording of (e). Mr Louboutin pointed out that the provision talked of signs which ‘consist exclusively’ of the shape giving substantial value of the goods. Louboutin pointed out that that was not the case here. Louboutin’s mark comprised for one part the form, and for another, the use of a particular shade of red.

Nor was Louboutin keen on a preliminary reference being made to the CJEU, which it considered premature. If questions were to be referred, then Louboutin thought it appropriate for the Dutch court to ask 16 questions of the CJEU! In contrast, Van Haren thought one intricately drafted question might suffice.

The three judges at the District Court in The Hague decided to ask just one question. The first half of their question reflects the wording initially proposed by Van Haren.

The second half of the Dutch court’s question seems to be a cleaned-up and rewritten version of the second half of Van Haren’s originally-proposed question. Namely, Van Haren had also relied on the EU’s ‘recast’ version of the trade mark directive, 2015/2436 of 16 December 2015 to approximate the laws of the Member States relating to trade marks.

Article 4 of the recast Directive will prevent the registration of ‘signs which consist exclusively’ of: ‘the shape, or another characteristic, which gives substantial value to the goods’.

From the Article 4 addition of the phrase ‘or another characteristic’, Van Haren inferred that the existing drafting of the EU’s trade mark Directive was unclear. Consequently, Van Haren wondered whether the exception in Article 3(1)(e)(iii) of the trade mark Directive would not only cover signs which had been applied to the good at a specific position but also signs where the good merged or coincided with (a component) of the good, such as colours.

The Dutch judges decided that the element of ‘position’ was not relevant to the dispute, but they incorporated the essence of Van Haren’s various concerns into the question they decided to ask the CJEU.

Question Referred
My unofficial translation of the question asked by the District Court in The Hague reads:

Where the German, English and French versions of Article 3(1)(e)(iii) of Directive 2008/95/EC use the words ‘Form’, ‘shape’ and ‘forme’, is the Dutch word ‘vorm’ to be understood as limited to the good’s three-dimensional characteristics such as the contours, measurements and volume (expressed in three dimensions), or rather does this provision also cover other (non three-dimensional) characteristics of the good such as colour?

Could the question in Louboutin be an example of judicial ‘cobblers’? The Dutch court’s question refers to the word ‘shape’ in various language versions, and suggests that the Dutch version should be ‘limited’. Yet it is not clear from the order of reference why the Dutch court has adopted this drafting.

Is there a language problem between the Benelux Convention and the Directive? No. The authentic language versions of the Benelux Convention are both Dutch and French. The Dutch language versions of both the Convention and the EU Directive use the same word, ‘vorm’. The same can be said for the French versions of both pieces of legislation. Thus, there is no terminological inconsistency between the Benelux Convention and the EU legislation.

Is there a legal-translation problem that is internal to EU law? The Dutch judges do not identify one. Though their question is premissed on ‘limiting’ the scope of the Dutch word ‘vorm’, they do not explain this premiss. They do not provide a dictionary definition of the Dutch word ‘vorm’. Nor do the Dutch judges indicate what they understand by the German, English and French words of ‘Form’, ‘shape’ and ‘forme’. Nor do the Dutch judges identify what if any legal consequences flow from those foreign-language terms.

In the absence of any explanation as to why the Dutch version of the EU Directive is either broader than the other language versions and should therefore be ‘limited’, is it correct to infer that the Dutch court’s question simply houses a false premiss?

Perhaps the Dutch judges were just looking for an excuse to make a reference to the CJEU because of the broader policy issue at stake. Namely, Van Haren had claimed that the form of the good here ‘merged’ or ‘coincided’ with a colour property of the good (red). Because they merged, Van Haren was arguing that the mark should similarly be deemed to be invalid.

In that context, the Dutch judges identified the key problem of policy. If Louboutin’s mark were indeed to be deemed valid, then the effect of that finding would destroy competition. Competitors providing the public with goods that had valued characteristics, such as a red sole to a lady’s high-heeled shoe, would be prevented from entering the market. Nor would the barrier be but a temporary phenomenon: a mark was not subject to any temporal limitation. Thus, if the mark were valid, then that would create a situation which seemed to be at odds with the policy consideration that underpinned the ‘shape/value’ exemption in trade mark law.

Perhaps the CJEU will rejig the question asked by the judges in The Hague District Court?

Readers of EU Law Radar who are interested in EU trade mark law on shapes and colours are referred to the CJEU’s earlier judgments in Hauck, Apple and Oberbank.

The above information was first posted on EU Law Radar as part of the comment to another Benelux-EU law preliminary reference, Case C-169/15, Montis Design – EU copyright and Benelux design formalities, a game of musical chairs?

Update – 3 June 2016
Perhaps the CJEU has indeed rejigged the question asked by the judges in The Hague District Court because the mention of the allegedly-disperate language versions of the international legislation and EU legislation has been erased.

According to the website of the UK’s Intellectual Property Office, the Dutch court has asked:

Is the notion of ‘shape’ within the meaning of Article 3(1)(e)(iii) of Directive 2008/95/EC to approximate the laws of other Member States relating to Trade Marks limited to the three-dimensional properties of the goods, such as their contours, measurements and volume (expressed three-dimensionally), or does it include other (non three-dimensional) properties, such as their colour?

Update – 25 January 2017
If a shoe sole becomes protectable as a trade mark, then attention should be paid to another preliminary reference from the Dutch courts about accessing evidence which is capable of establishing the infringement of an intellectual property right; see further, Case C-644/16, Synthon – resisting its impounded documents being inspected by a rival.

Even if a shoe sole does not not become protectable as a trade mark, the makers and importers of luxury leather footwear are already litigating several anti-dumping cases; see further, Case C-631/16, Timberland Europe – puts its boot into retroactive anti-dumping Regs.