Aveda Corporation v Dabur India Ltd

On 20 May 2010 Dabur India Ltd (“Dabur”) applied to register the trade mark DABUR UVEDA in respect of the following goods:

“Class 3

Bleaching preparations and other substances for laundry use; cleaning, polishing, scouring and abrasive preparations; soaps; perfumery, essential oils, cosmetics, hair lotions; dentifrices; deodorants for personal use.

Class 5

Pharmaceutical and veterinary preparations; sanitary preparations for medical purposes; dietetic substances adapted for medical use, food for babies; plasters, materials for dressings; material for stopping teeth, dental wax; disinfectants; preparations for destroying vermin; fungicides, herbicides; foods and beverages which are adapted for medical purposes; air deodorising preparations.”

The application was subsequently opposed by Aveda Corporation (“Aveda”), primarily on grounds raised under section 5(2)(b) of the Trade Marks Act 1994 which implements Article 5(1)(b) of European Parliament and Council Directive 2008/95/EC of 22 October 2008 to approximate the laws of the Member States relating to trade marks (codified version) (“the Directive”).

Aveda relied upon three earlier registrations of the trade mark AVEDA:

It will be appreciated that it is Aveda’s registrations in Classes 3 and 5 that are of most relevance to the opposition. It is not necessary for the purposes of this judgment to set out the Class 3 specification of the first registration in full, but it included “cleaning, polishing, scouring and abrasive preparations”.

Dabur required Aveda to prove use of the first two registrations. The third registration was too recent for proof of use to be required. At the hearing before the hearing officer, Aveda contended, and Dabur did not dispute, that it had proved use in relation to the following goods in Class 3 (emphasis added):

“Hair care preparations, namely shampoo, cream rinses, hair spray and hair gel, colour rinses, liquids for permanent waves; skin care products, namely moisturising creams, lotions, cleansing lotions; lotions, beauty masks and astringent face creams for cosmetic purposes; cosmetics, namely lipsticks, eye shadow, foundations, rouges, face powder, eye pencils, lip pencils and mascara; toiletries for personal use, namely body cleansing lotions, eau-de-toilette; perfumery and eau de Cologne, massage oils and lotions; soaps; essential oils; skin creams, skin lotions, body oils and body lotions for softening and moisturising the skin; cleaning, polishing, scouring and abrasive preparations for use on the body.”

Curiously, Aveda also contended, and Dabur did not dispute, that it had proved use in relation to “scented candles” in Class 4 even though those goods were not covered by either of Aveda’s first two registrations.

Aveda accepted that it was unable to prove use in relation to goods in Class 5 extending beyond the Class 5 specification of the third registration, namely “dietary and nutrition supplements, vitamins”.

At the hearing, Aveda sought permission to adduce a further witness statement which it had served very shortly beforehand. The hearing officer Mark Bryant refused to grant permission on the ground that the evidence was “of minimal, if any, relevance”.

The hearing officer dismissed the opposition for reasons given in a written decision dated 22 August 2012 (O/318/12). Aveda has appealed against that decision. At the hearing of the appeal, however, Aveda only pursued the appeal in respect of certain goods covered by Dabur’s application. Aveda did not object to the application proceeding to registration in respect of the following goods (emphasis added):

“Class 3

Bleaching preparations and other substances for laundry use; household and industrial preparations, namely, cleaning, polishing, scouring and abrasive preparations, none for use on the body.

Class 5

Veterinary preparations; sanitary preparations for medical purposes; disinfectants; food for babies; plasters; materials for dressings; material for stopping teeth, dental wax, preparations for destroying vermin, herbicides.”

The hearing officer’s decision may be summarised as follows. First, he set out the Trade Mark Registry’s standard summary of the principal decisions of the Court of Justice of the European Union concerning Article 5(1)(b) of the Directive or its equivalents, namely Case C-251/95 SABEL BV v Puma AG [1997] ECR I-6191, Case C-39/97 Canon Kabushiki Kaisha v Metro-Goldwyn-Meyer Inc [1998] ECR I-5507, Case C-342/97 Lloyd Schuhfabrik Meyer & Co GmbH v Klijsen Handel BV [1999] ECR I-3819, Case C-425/98 Marca Mode CV v Adidas AG [2000] ECR I-4861, Case C-3/03 Matrazen Concord GmbH v Office for Harmonisation in the Internal Market [2004] ECR I-3657, Case C-120/04 Medion AG v Thomson Sales Germany & Austria GmbH [2005] ECR I-8551 and Case C-334/05 Office for Harmonisation in the Internal Market v Shaker de L. Laudato & C SAS [2007] ECR I-4529:

“(a)

the likelihood of confusion must be appreciated globally, taking account of all relevant factors: Sabel BV v Puma AG;

(b)

the matter must be judged through the eyes of the average consumer of the goods/services in question; Sabel BV v Puma AG, who is deemed to be reasonably well informed and reasonably circumspect and observant - but who rarely has the chance to make direct comparisons between marks and must instead rely upon the imperfect picture of them he has kept in his mind: Lloyd Schuhfabrik Meyer & Co. GmbH v Klijsen Handel BV;

(c)

the average consumer normally perceives a mark as a whole and does not proceed to analyse its various details: Sabel BV v Puma AG;

(d)

the visual, aural and conceptual similarities of the marks must therefore be assessed by reference to the overall impressions created by the marks bearing in mind their distinctive and dominant components: Sabel BV v Puma AG;

(e)

a lesser degree of similarity between the marks may be offset by a greater degree of similarity between the goods, and vice versa: Canon Kabushiki Kaisha v Metro-Goldwyn-Mayer Inc:

(f)

there is a greater likelihood of confusion where the earlier trade mark has a highly distinctive character, either per se or because of the use that has been made of it: Sabel BV v Puma AG:

(g)

in determining whether similarity between the goods or services covered by two marks is sufficient to give rise to the likelihood of confusion, the distinctive character and reputation of the earlier mark must be taken into account: Canon Kabushiki Kaisha v Metro-Goldwyn-Mayer Inc;

(h)

mere association, in the sense that the later mark brings the earlier mark to mind, is not sufficient for the purposes of section 5(2): Sabel BV v Puma AG;

(i)

further, the reputation of a mark does not give grounds for presuming a likelihood of confusion simply because of a likelihood of association in the strict sense: Marca Mode CV v Adidas AG and Adidas Benelux BV;

(j)

but 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 within the meaning of the section: Canon Kabushiki Kaisha v Metro-Goldwyn-Mayer Inc;

(k)

assessment of the similarity between two marks means more than taking just one component of a composite trade mark and comparing it with another mark; the comparison must be made by examining each of the marks in question as a whole, which does not mean that the overall impression conveyed to the relevant public by a composite trade mark may not, in certain circumstances, be dominated by one or more of its components: Medion AG v. Thomson Multimedia Sales Germany & Austria GmbH;

(l)

it is only when all other components of a complex mark are negligible that it is permissible to make the comparison on the basis of the dominant element: Shaker di L. Laudato & C. Sas v OHIM.”

Secondly, he compared the respective goods, concluding that some of the goods in Dabur’s specification were identical to goods covered by Aveda’s specifications in respect of which use had been proved or was not required, some were either highly or reasonably similar and some were not similar. In this regard the following passages in the decision are of importance for the purposes of the appeal:

“Dabur’s Class 3 goods

…

28.

In respect of Dabur’s cleaning, polishing, scouring and abrasive preparations, Mr Bartlett submitted that such terms include goods for use on the body and, consequently, the terms are in respect of identical goods to those of Aveda. I remain unconvinced by this submission. The natural meaning attributed to Dabur’s unqualified term is likely to be one of describing household or industrial preparations rather than those for use on the body. This perception is amplified by the use of all four adjectives together in the same term. There is nothing before me to illustrate that the terms scouring preparations and abrasive preparations are terms used to normally describe preparations for use on the body (even where a preparation may actually work by scouring or abrading part of the body). These terms are more readily associated with preparations that have household or industrial applications. The inclusion of such terms in the overall term cleaning, polishing, scouring and abrasive preparations has the consequence of leading the reader (and consumer) to believe that the overall term relates to household and/or industrial preparations and not preparations for use on the body. Such a finding appears to be consistent with the guidance provided in Beautimatic and Thomson Holidays. The guidance provided in and [sic] Avnet was in respect to services rather than goods, but even if it is considered that the guidance has equal application to the consideration of goods, I do not see my finding as being inconsistent with that guidance. In summary, I find that Dabur’s cleaning, polishing, scouring and abrasive preparations are not similar to Aveda’s goods, as identified by Mr Bartlett.

…

32.

Finally, I consider the similarity of Dabur’s dentifrices with Aveda’s goods and services. These are pastes or powders for cleaning the teeth. Whilst such goods can, therefore, strictly be described as “cleaning preparations”, I nonetheless find that there is no similarity with Aveda’s cleaning … preparations in Class 3. The reason for this is the same as set out in my paragraph 28 above.

Dabur’s Class 5 goods

33.

Mr Bartlett submitted that Dabur’s Pharmaceutical […] preparations includes medicated skin care preparations and that these share a high similarity with Aveda’s skin care products in Class 3. I concur with this submission. These respective goods may share the same nature (for example, being a cream for application directly to the skin), and may also have an overlap of trade channels with these respective goods being sold in the same type of retail outlet and if not on the same, then at least in closely adjacent parts of the outlet. Their intended purpose of improving the condition of the skin, is also the same, even though, they may act in a different way. Consequently, I conclude that these goods share a reasonably high level of similarity.

34.

Mr Bartlett argued further that Dabur’s preparations for destroying fungicides could be in the form of medicated skin preparations and, as such, they are also very similar to Aveda’s skin care products. I am not convinced by this rationale. I do not believe that such an interpretation should be given to the term preparations for destroying fungicides. Rather, the term will ordinarily be understood as referring to products not designed for use on the body. Consequently, and for the same reasons as set out in paragraph 28 above, I conclude there is no similarity with Aveda’s goods.”

Thirdly, he considered the attributes of the relevant average consumer, saying in relation to the Class 3 goods at [40]:

“… Such goods are relatively regular purchases, with the latter purchased as part of regular shopping for grocery and general provisions. Consequently, such goods are not bought with the greatest of care, but nonetheless, for goods such as cosmetics, more consideration may be involved than in respect of everyday purchases. The purchases are likely to involve mainly visual considerations, but I do not discount that aural considerations may also be relevant.”

Fourthly, he compared the respective marks. He noted that it was common ground that neither mark had any conceptual meaning. Comparing the marks as a whole, he concluded at [47] as follows:

“In summary, I have found that the respective marks share a low level of visual similarity, a moderate level of aural similarity but that they are neither conceptually similar nor dissimilar. This combines to give the marks a moderate level of similarity overall.”

Fifthly, he assessed the distinctive character of Aveda’s trade mark, finding that it had a “high level of inherent distinctive character”. He also accepted that there had been significant use of that mark in respect of beauty products and services, but due to the inherent distinctiveness of the mark he did not consider the resulting enhancement to the distinctive character decisive.

Finally, he assessed the likelihood of confusion. In this regard, he noted that Aveda contended, relying upon BULOVA ACCUTRON Trade Mark [1968] FSR 336 and Medion v Thomson, that UVEDA was confusingly similar to AVEDA and that the addition of DABUR did not prevent there from being a likelihood of confusion. He rejected this contention for reasons which he expressed at [57] as follows:

“… I am not convinced by Mr Bartlett’s approach. I have already found that the DABUR and UVEDA elements of Dabur’s mark share equal prominence in its mark. I have particular regard for the guidance in MEDION (as referred to by Mr Bartlett) and also in Matratzen Concord GmbH v OHIM Case T-6/01 [2002] ECR II-4335 and Matratzen Concord GmbH v OHIM Case C-3/03P. However, in the current case, it is my view that the word UVEDA does not have a dominant role as referred to in these cases. It has no more than an equal dominance with the word DABUR. When considering the marks as a whole, it is my view that the addition of the word DABUR and the different first letter in the words UVEDA/AVEDA are sufficient to avoid a likelihood of confusion, even where identical goods are involved. ”

Aveda’s grounds of appeal can be summarised as follows:

It is common ground that the correct approach to an appeal of this kind is that stated by Robert Walker LJ (as he then was) in REEF Trade Mark [2002] EWCA Civ 763, [2003] RPC 5 at [28]:

“In such circumstances an appellate court should in my view show a real reluctance, but not the very highest degree of reluctance, to interfere in the absence of a distinct and material error of principle.”

Aveda’s first and most important ground of appeal is that the hearing officer failed correctly to apply the guidance of the CJEU in Medion. Before considering this in detail, I think it is worth making a preliminary observation. As I have recorded, the hearing officer directed himself by reference to the Registry’s standard summary of the principles established by the CJEU’s leading decisions. He cannot be criticised for doing so. I have used the summary myself in a number of judgments, and it was cited with approval by Kitchin LJ in Specsavers International Healthcare Ltd v Asda Stores Ltd [2012] EWCA Civ, [2012] FSR 19 at [52]. Nevertheless, it must not be forgotten that it is only a convenient summary. There are cases in which it is necessary to look in more detail at aspects of the CJEU’s jurisprudence. This is such a case.

In Medion v Thomson Medion was the owner of the German registered trade mark LIFE for leisure electronic devices. Thomson marketed such products under the sign THOMSON LIFE. Medion claimed that this was an infringement. The Landgericht (Regional Court) Düsseldorf dismissed the claim on the ground that there was no likelihood of confusion. The Oberlandesgericht (Higher Regional Court) Düsseldorf referred the following question to the Court of Justice:

“Is Article 5(1)(b) of [the Directive] to be interpreted as meaning that where the goods or services covered by competing signs are identical there is also a likelihood of confusion on the part of the public where an earlier word mark with normal distinctiveness is reproduced in a later composite word sign belonging to a third party, or in a word sign or figurative sign belonging to a third party that is characterised by word elements, in such a way that the third party’s company name is placed before the earlier mark and the latter, though not alone determining the overall impression conveyed by the composite sign, has an independent distinctive role within the composite sign?”

In its referring judgment the Oberlandesgericht explained that, according to the case law of the Bundesgerichtshof (Federal Court of Justice), which was based on the “Prägetheorie” (theory of the impression conveyed), in order to appreciate the similarity of the signs in issue, it was necessary to consider the overall impression conveyed by each of the two signs and to ascertain whether the common component (there LIFE) characterised the composite sign (THOMSON LIFE) to the extent that the other component (THOMSON) was largely secondary to the overall impression. There would be no likelihood of confusion if the common component merely contributed to the overall impression of the composite sign. It did not matter whether the sign corresponding to the trade mark still had an independent distinctive role in the composite sign.

In answering the Oberlandesgericht’s question, the Court of Justice began by recapitulating the principles it had already laid down with regard to the assessment of likelihood of confusion (paragraphs (a)-(g) of the Registry’s summary). It then stated:

“29.

In the context of consideration of the likelihood of confusion, assessment of the similarity between two marks means more than taking just one component of a composite trade mark and comparing it with another mark. On the contrary, the comparison must be made by examining each of the marks in question as a whole, which does not mean that the overall impression conveyed to the relevant public by a composite trade mark may not, in certain circumstances, be dominated by one or more of its components (see Matratzen Concord, paragraph 32).

30.

However, beyond the usual case where the average consumer perceives a mark as a whole, and notwithstanding that the overall impression may be dominated by one or more components of a composite mark, it is quite possible that in a particular case an earlier mark used by a third party in a composite sign including the name of the company of the third party still has an independent distinctive role in the composite sign, without necessarily constituting the dominant element.

31.

In such a case the overall impression produced by the composite sign may lead the public to believe that the goods or services at issue derive, at the very least, from companies which are linked economically, in which case the likelihood of confusion must be held to be established.

32.

The finding that there is a likelihood of confusion should not be subject to the condition that the overall impression produced by the composite sign be dominated by the part of it which is represented by the earlier mark.

33.

If such a condition were imposed, the owner of the earlier mark would be deprived of the exclusive right conferred by Article 5(1) of the directive even where the mark retained an independent distinctive role in the composite sign but that role was not dominant.

34.

This would be the case where, for example, the owner of a widely-known mark makes use of a composite sign juxtaposing this mark and an earlier mark which is not itself widely known. It would also be the case if the composite sign was made up of the earlier mark and a widely-known commercial name. In fact, the overall impression would be, most often, dominated by the widely-known mark or commercial name included in the composite sign.

35.

Thus, contrary to the intention of the Community legislator expressed in the 10th recital in the preamble to the directive, the guarantee of the earlier mark as an indication of origin would not be assured, even though it still had an independent distinctive role in the composite sign.

36.

It must therefore be accepted that, in order to establish the likelihood of confusion, it suffices that, because the earlier mark still has an independent distinctive role, the origin of the goods or services covered by the composite sign is attributed by the public also to the owner of that mark.

37.

Accordingly, the reply to the question posed must be that Article 5(1)(b) of the directive is to be interpreted as meaning that where the goods or services are identical there may be a likelihood of confusion on the part of the public where the contested sign is composed by juxtaposing the company name of another party and a registered mark which has normal distinctiveness and which, without alone determining the overall impression conveyed by the composite sign, still has an independent distinctive role therein. ”

It can be seen that paragraph (k) in the Registry’s summary accurately reflects what the Court of Justice said in Medion v Thomson at [29], which addresses what the Court of Justice described at [30] as “the usual case”. Paragraph (k) does not reflect what the Court of Justice went on to say at [30]-[37] about the impact of a sign which forms part of a composite sign, but which has an independent distinctive role, on the likelihood of confusion.

Medion v Thomson has been referred to in many subsequent cases, but for present purposes it is only necessary to refer to six. I shall take them in chronological order.

In Novartis Seeds BV’s Application [2006] ETMR 82 Novartis applied to register the word CANTO. The Registrar refused the application on the ground of conflict with an earlier registration for ERIC CANTONA CANTO for identical goods. Geoffrey Hobbs QC sitting as the Appointed Person distinguished the case from BULOVA ACCUTRON and Medion v Thomson on the ground that the average consumer would perceive the earlier mark as a doubling up of a name and its nickname rather than a house mark plus invented word. Accordingly, he did not consider that the word CANTO performed an independent distinctive role in the composite mark ERIC CANTONA CANTO.

In DRILLISCH ALPHATEL Trade Mark Application [2009] ETMR 27 the applicant applied to register “Drillisch ALPHATEL” as a trade mark in Germany. The opponent opposed the applicant on the ground of conflict with four earlier registrations for the trade mark ALCATEL for very similar goods and services. The opposition was dismissed on the ground that there was no likelihood of confusion. The opponent’s appeal to the Bundespatentgericht (Federal Patent Court) was dismissed. Part of the Court’s reasoning was that the opponent had failed to prove that its trade mark had acquired an enhanced distinctive character: see paragraph [7] of the ETMR report. More importantly for present purposes, the Court held that the respective signs were only very slightly similar for reasons which it expressed as follows:

“9

a) As to their overall impression, which in principle is to be given priority in assessing the risk of confusion between trade marks, irrespective of the priority in time of the signs being compared (cf. Medion AG v Thomson Multimedia Sales Germany & Austria GmbH (C-120/04) [2005] E.C.R. I-8551; Sabel v Puma), the marks being compared immediately differ visually, aurally and conceptually because of the addition of ‘Drillisch’ in the contested mark; moreover there are significant visual differences between the contested marks, which are merely protected as word marks, and the marks relied on, nos 1 139 897 and IR 525 160, because of their graphical format.

10

b) Contrary to the understanding of the opponent, the contested mark is also not characterised solely by the ALPHATEL component.

11

aa) Insofar as the opponent relies on the judgment of the European Court of Justice (ECJ) in Medion on this point, and argues that the ALPHATEL component has independent distinctive effect, it misunderstands what is said in that judgment. As the ECJ expressly emphasises, in fact (cf. cited above at [30]), the situation where individual elements of a composite mark are independently distinctive is an exception from the general principle of assessment by reference to the overall impression, or by reference to the impression conveyed by individual component parts of the mark ‘in a particular case’ only when the contested mark is made up of a composite of the old mark and the company name of the proprietor of the new mark. The present case does not involve any such adoption of the old sign, however, because the ALPHATEL and ALCATEL components of the mark are not identical. There is no reason for an extension of the case law of the ECJ on ‘adoption’ (which, anyway conceptually speaking, is not present in this case) of a sign which is merely similar to the old mark and the addition of the company name of the proprietor of the contested mark, because of the clear wording of the judgment on the rule-exception-relationship, so that on this question no preliminary ruling will be obtained from the ECJ under art.234 EC nor any appeal on a point of law permitted under s.83(2) of the Trade Mark Law (Germany) .

12

bb) Insofar as the opponent further considers that the ‘Drillisch’ part of the contested mark is less significant because it is the producer's name, it also misunderstands, fundamentally, that the case law of the Federal Supreme Court does not permit such an inference to be drawn (cf. Mustang BGH [2004] GRUR 865, 866). For the field of telecommunications, which was the field it principally considered, the Federal Supreme Court has moreover even accorded the producer's name significance as one of the factors conveying an impression (cf. T-INNOVA/Innova BGH [2003] GRUR 70, 73), so that in relation to the goods and services sector at issue here, it is to be assumed that in the eyes of the public to whom the mark is directed the ‘Drillisch’ component of the mark not only is not of secondary importance, but in fact represents the element which actually conveys an impression. It is not necessary to decide whether it is still possible to maintain differentiation by sector in the wake of the Medion v Thomson judgment (critical of this, Ströbele/Hacker, Markengesetz, 8th edn (2006), §9 para.289), because in the opinion of the ECJ—differing in this respect from the Federal Supreme Court—the producer's name always dominates a sign made up of more than one part (cf. Medion v Thomson at [34]), so that even when the more recent Supreme Court case law of the ECJ is used as a basis for decision making, the ‘Drillisch’ component of the mark in the contested mark has at least contributory if not dominant significance in the impression conveyed by the contested mark.”

It can be seen from this reasoning that the Federal Patent Court held that Medion v Thomson did not extend to the case where the composite sign incorporated a sign which was merely similar to, rather than identical with, the earlier trade mark.

In OHIM v Shaker Shaker applied to register the following figurative mark as a Community trade mark:

The application was opposed by Limiñana y Botell, SL on the basis of an earlier Spanish registration for word LIMONCHELO. The OHIM Opposition Division upheld the opposition, essentially because it considered that the dominant element of the mark applied for was the word LIMONCELLO which was confusingly similar to LIMONCHELO. The Second Board of Appeal dismissed Shaker’s appeal. The Court of First Instance (now the General Court) allowed an appeal by Shaker. On an appeal by OHIM, the Court of Justice set aside the judgment of the CFI and remitted the case back to that court. The Court of Justice held that the CFI had misapplied Article 8(1)(b) for the following reasons:

“35.

According to further settled case-law, the global assessment of the likelihood of confusion, in relation to the visual, aural or conceptual similarity of the marks in question, must be based on the overall impression given by the marks, bearing in mind, in particular, their distinctive and dominant components. The perception of the marks by the average consumer of the goods or services in question plays a decisive role in the global appreciation of that likelihood of confusion. In this regard, the average consumer normally perceives a mark as a whole and does not proceed to analyse its various details (see SABEL, paragraph 23; Lloyd Schuhfabrik Meyer, paragraph 25; Medion, paragraph 28; Mülhens v OHIM, paragraph 19; and order in Matratzen Concord v OHIM, paragraph 29).

…

37.

In the present case the Court of First Instance, at paragraph 49 of the judgment under appeal, noted the case-law mentioned in paragraph 35 of the present judgment according to which the global appreciation of the likelihood of confusion must be based on the overall impression created by the signs at issue.

38.

However, it stated in paragraph 54 of the judgment under appeal that, if the trade mark claimed was a complex mark which was visual in nature, the assessment of the overall impression created by that mark and the determination as to whether there was a dominant element had to be carried out on the basis of a visual analysis. It added that, in such a case, it was only to the extent to which a potentially dominant element included non-visual semantic aspects that it might become necessary to compare that element with the earlier mark, also taking into account those other semantic aspects, such as for example phonetic factors or relevant abstract concepts.

39.

On the basis of those considerations, the Court of First Instance, in the context of the analysis of the signs at issue, firstly held that the mark for which registration was sought contained a dominant element comprising the representation of a round dish decorated with lemons. It then inferred, in paragraphs 62 to 64 of the judgment under appeal, that it was not necessary to examine the phonetic or conceptual features of the other elements of that mark. It finally concluded, in paragraph 66 of the judgment, that the dominance of the figurative representation of a round dish decorated with lemons in comparison with the other components of the mark prevented any likelihood of confusion arising from the visual, phonetic or conceptual similarities between the words ‘limonchelo’ and ‘limoncello’ which appear in the marks at issue.

40.

However, in so doing, the Court of First Instance did not carry out a global assessment of the likelihood of confusion of the marks at issue.

41.

It is important to note that, according to the case-law of the Court, in the context of consideration of the likelihood of confusion, assessment of the similarity between two marks means more than taking just one component of a composite trade mark and comparing it with another mark. On the contrary, the comparison must be made by examining each of the marks in question as a whole, which does not mean that the overall impression conveyed to the relevant public by a composite trade mark may not, in certain circumstances, be dominated by one or more of its components (see order in Matratzen Concord v OHIM, paragraph 32; Medion, paragraph 29).

42.

As the Advocate General pointed out in point 21 of her Opinion, it is only if all the other components of the mark are negligible that the assessment of the similarity can be carried out solely on the basis of the dominant element.”

It can be seen that paragraph (l) in the Registry’s summary accurately reflects what the Court of Justice said in OHIM v Shaker at [42]. In my view it is clear that in that paragraph the Court of Justice was merely adding a slight clarification to what it had said in Medion v Thomson at [29], which it had repeated in the preceding paragraph. This is nothing to suggest that it intended to qualify in any way what it had said in Medion v Thomson at [30]-[37]. Nor did it have any occasion to do so, since OHIM v Shaker was factually a different kind of case to Medion.

In Rousselon Frères et Cie v Horwood Homewares Ltd [2008] EWHC 881 (Ch), [2008] RPC 30 Horwood was the proprietor of registrations for JUDGE SABATIER and STELLAR SABATIER. Rousselon applied for a declaration of invalidity on the ground of conflict with several earlier registrations, one of which was of the word SABATIER on its own. The hearing officer dismissed the application. Warren J allowed the appeal on the ground that the hearing officer had failed correctly to apply Medion v Thomson.

In his judgment Warren J said this:

“90.

The Court in Medion was clear in its rejection of any precondition to a finding of likelihood of confusion that the overall impression produced by the composite sign be dominated by the part of it which is represented by the earlier mark. In the present case, such a precondition would, of course, mean that there is no likelihood of confusion because, on Mr Foley's findings of fact, SABATIER was not the dominant element of HH's Marks. Although as noted in paragraph 23 above, the actual answer given by the Court related to the absence of such a precondition, it is clear from the decision that there will be a likelihood of confusion if the earlier mark retains an independent distinctive role in the later composite mark.

91.

In this context, paragraph 34 of the Judgment is of great importance in stating that this independent distinctive role will be present in the examples given. There is no material difference between those examples and the present case, (see paragraph 93 below). It may be that, in some circumstances, an earlier mark may not retain any independent distinctive role, a possibility which is consistent with the language of paragraph 31 (‘……..it is quite possible…’) and paragraph 31 (‘….the overall impression produced …..may lead….’). But that would surely be an exceptional case. As it is said in paragraph 36 ‘ in order to establish the likelihood of confusion, it suffices that, because the earlier mark still has an independent distinctive role, the origin of the goods or services covered by the composite sign is attributed by the public to the owner of that mark’. Thus where the earlier mark is a word mark, and that earlier mark forms part of a composite mark itself comprising words (whether with or without a further device), it may be difficult, or so it seems to me, to resist the conclusion that the earlier mark does indeed retain an independent distinctive role.

92.

Mr Arnold submits that Medion is simply a particular application of the ECJ's earlier case-law which I have already mentioned. I agree with that, as does Mr Vanhegan. But it is a particular application which shows how the ECJ regards a composite mark which includes an earlier mark which retains an independent distinctive role. Of course, the actual test which must not be lost sight of remains whether, because of similarity of marks, there is a likelihood of confusion. The court in Medion does not in any way abandon that test. It does not say that a composite mark which contains an earlier mark will always carry a likelihood of confusion; rather, it says that there is such likelihood if the earlier mark has an independent distinctive role. However, the assessment whether the earlier mark does have such a role is, I consider, something which has to be decided before answering the question whether there is a likelihood of confusion. To elide the question of an independent distinctive role with the question of the likelihood of confusion runs the risk of answering the former question by first providing an answer to the latter. That would be to answer the question ‘Is there a likelihood of confusion’ with the answer ‘Yes, if there is a likelihood of confusion’ when the answer should be ‘Yes, if the earlier mark retains an independent distinctive role’.”

In this passage Warren J came quite close to saying that, if the earlier mark retains an independent distinctive role within the composite sign, then it must follow that there will be a likelihood of confusion. I do not think he went quite that far, however, since he recognised that the ultimate test remains whether there is a likelihood of confusion. In any event, as counsel for Dabur submitted, what the Court of Justice said at [37] of Medion v Thomson is that there “may be” a likelihood of confusion in those circumstances.

In Joined Cases T-5/08 to T-7/08 Société des Produits Nestlé v OHIM [2010] ECR II-1177 the applicant applied to register three Community trade marks consisting of the words GOLDEN EAGLE and an image of a red coffee mug placed on coffee beans. Nestlé opposed the applications on the grounds of conflict with various registrations consisting of a similar image of a red coffee mug on coffee beans for identical goods. The Opposition Division refused the oppositions finding the respective marks to be dissimilar. The Board of Appeal upheld the decision. Nestlé (represented, it may be noted, by the distinguished German trade mark lawyer Alexander von Mühlendahl) appealed to the General Court. The General Court annulled the decision of the Board of Appeal. It found that Nestlé’s mark and the corresponding element in the composite mark were similar and that there was some, albeit slight, overall visual and conceptual similarity between the marks. Accordingly, the Court concluded that the Board of Appeal should therefore have assessed the likelihood of confusion.

In this context the General Court held at [60]:

“It must also be pointed out that, according to the case-law, where a composite mark is composed by juxtaposing one element and another mark, that latter mark, even if it is not the dominant element in the composite mark, still can have an independent distinctive role therein. In such a case, the composite mark and that other mark may be held to be similar (see, to that effect, Medion, paragraph 28 above, paragraphs 30 and 37). It must be observed that Medion, paragraph 28 above, concerned a situation in which the earlier mark had been reproduced identically in the later mark. Nevertheless, it must be held that, in the event that the earlier mark is not reproduced identically in the later mark, it is still possible that the signs at issue are similar on account of the similarity between the earlier mark and one element of the later mark which has an independent distinctive role.”

It does not appear, however, that the General Court’s reasoning was based on a finding that the image of a red coffee mug on coffee beans had an independent distinctive role in the marks applied for. Rather, its reasoning was based on an overall assessment of similarity.

In Case T-569/10 Bimbo SA v OHIM [2012] ECR II-0000, [2013] ETMR 7 the applicant applied to register BIMBO DOUGHNUTS as a Community trade mark. The application was opposed on the basis of an earlier Spanish registration for DOGHNUTS covering identical and similar products. The opposition was successful. The applicant’s appeal to the Board of Appeal was dismissed. The applicant appealed to the General Court. The General Court agreed with the lower tribunals that there was a likelihood of confusion and dismissed the appeal.

A key part of the General Court’s reasoning was as follows:

“96.

According to the case-law, where goods or services are identical there may be a likelihood of confusion on the part of the public where the contested sign is composed by juxtaposing the company name of another party and a registered mark which has normal distinctiveness and which, without alone determining the overall impression conveyed by the composite sign, still has an independent distinctive role therein (Case C-120/04 Medion [2005] ECR I-8551, paragraph 37). There may also be a likelihood of confusion in a case in which the earlier mark is not reproduced identically in the later mark (see, to that effect, Joined Cases T-5/08 to T-7/08 Nestlé v OHIM – Master Beverage Industries (Golden Eagle and Golden Eagle Deluxe) [2010] ECR II-1177, paragraph 60).

97.

In this case, the ‘doughnuts’ element, which is almost identical to the earlier trade mark, has an independent distinctive role in the mark applied for. Indeed, contrary to what is claimed by the applicant, that element is not devoid of distinctive character but on the contrary has average distinctive character for the part of the relevant public which is not familiar with English. Furthermore, since the ‘doughnuts’ element is wholly meaningless for that consumer, the mark applied for, BIMBO DOUGHNUTS, does not form a unitary whole or a logical unit on its own in which the ‘doughnuts’ element would be merged. The part of the relevant public which is not familiar with English will not be able to understand the sign at issue as meaning that the goods concerned are doughnuts produced by the undertaking Bimbo or by the proprietor of the trade mark BIMBO.”

It should be noted that, although the General Court held that DOUGHNUTS was “almost identical” to DOGHNUTS, it did not proceed on the basis that the average consumer would perceive it as being identical. An appeal to the CJEU against the decision of the General Court is pending as Case C-591/12 P.

Turning to the present case, the following facts are relevant to the issue presently under consideration:

Against that background, Aveda’s trade mark attorney argued in summary as follows:

Counsel for Dabur argued in summary as follows:

Counsel for Dabur’s second submission (although not his first submission) is supported by DRILLISCH ALPHATEL. One of the leading commentaries on German trade mark law states, however, that Medion v Thomson can apply where the earlier mark and the component of the sign are not identical, but only similar. In this context the authors refer to DRILLISCH ALPHATEL as “unzutreffend”, i.e. wrong: see R. Ingerl and C. Rohnke, Markengesetz (Beck, 2010) §14 at paragraph 1135. They are supported by Nestlé v OHIM and Bimbo v OHIM, particularly the latter.

I must apply the decisions of the General Court. In addition, however, I consider that the General Court is correct on this point. My reasons are as follows.

Although the decision in Medion v Thomson does not in terms extend to cases in which the composite sign incorporates a sign which is similar to, rather than identical with, the trade mark and some of the Court of Justice’s reasoning would not apply to such a case, I consider that the underlying logic is equally applicable.

I entirely accept the basic proposition which the Court of Justice has repeated many times, namely that the assessment of likelihood of confusion must be made by considering and comparing each of the signs as a whole. As the Court of Justice recognised in Medion v Thomson, however, there are situations in which the average consumer, while perceiving a composite sign as a whole, will recognise that it consists of two signs one or both of which has a significance which is independent of the significance of the composite whole. Thus when the well-known pharmaceutical company Glaxo plc acquired the well-known pharmaceutical company Wellcome plc, the average consumer of pharmaceutical goods confronted with the composite sign GLAXO WELLCOME or GLAXOWELLCOME would perceive the significance of both the whole and its constituent parts and conclude that this was an undertaking which combined the two previously separate undertakings (see Glaxo Group Ltd v Glaxowellcome Ltd [1996] FSR 388). The essence of the Court of Justice’s reasoning in Medion v Thomson is that an average consumer of leisure electronic products confronted with the composite sign THOMSON LIFE could perceive both the whole and its constituent parts to have significance and thus could be misled into believing that there was a similar kind of connection between the respective undertakings.

As Mr Hobbs’ decision in Novartis Seeds shows, this can only occur in circumstances where the consumer perceives the relevant part of the composite sign to have significance independently of the whole. In that case Mr Hobbs did not think that the average consumer would perceive CANTO to have significance independently of ERIC CANTONA CANTO. On the contrary, he considered that the average consumer would perceive ERIC CANTONA CANTO as a unit in which only ERIC CANTONA had independent significance.

In my view the principle which I have attempted to articulate in paragraph 45 above is capable of applying where the consumer perceives one of the constituent parts to have significance independently of the whole, but is mistaken as to that significance. Thus in BULOVA ACCUTRON the earlier trade mark was ACCURIST and the composite sign was BULOVA ACCUTRON. Stamp J held that consumers familiar with the trade mark would be likely to be confused by the composite sign because they would perceive ACCUTRON to have significance independently of the whole and would confuse it with ACCURIST.

On that basis, I consider that the hearing officer failed correctly to apply Medion v Thomson. He failed to ask himself whether the average consumer would perceive UVEDA to have significance independently of DABUR UVEDA as a whole and whether that would lead to a likelihood of confusion. Accordingly, it is necessary for me to consider the matter afresh. Having regard to the matters set out in paragraph 39 above, I think there can be little doubt that the average consumer who was familiar with AVEDA beauty products would be likely to be confused by the use of DABUR AVEDA in relation to identical goods. In particular, there would be a strong likelihood that the average consumer would think that it indicated some connection between DABUR and AVEDA. In my judgment it makes little difference that the second word in the composite mark is UVEDA rather than AVEDA. As the hearing officer rightly accepted, UVEDA is both visually and aurally very close to AVEDA. The human eye has a well-known tendency to see what it expects to see and the human ear to hear what it expects to hear. Thus it is likely that some consumers would misread or mishear UVEDA as AVEDA. (Indeed, not only did the hearing officer himself write AVEDA instead of UVEDA at [43], but also the Intellectual Property Office’s database of past decisions currently records the contested mark as being “Dabur AVEDA”.)

While this leads to the conclusion that the hearing officer should have found that there was a likelihood of confusion at least in relation to goods covered by Dabur’s application which are identical to Aveda’s goods, it remains to be determined whether there is a likelihood of confusion in relation to goods which are similar. Before addressing this question, I must consider Aveda’s other grounds of appeal.

Aveda’s second ground of appeal is that the hearing officer was wrong not to admit its late evidence. This evidence consisted essentially of some print-outs from the internet showed how Dabur was using DABUR UVEDA. Since the evidence was tendered to prove two points which were not in dispute before me (namely those set out in sub-paragraphs 39(iii) and (iv) above), however, Aveda does not need this evidence.

Aveda’s third ground of appeal is that the hearing officer failed to apply the interdependency principle (paragraph (e) of the Registry’s summary). In support of this ground of appeal Aveda’s trade mark attorney pointed out that there is no reference to this principle, nor any overt sign of its application, in the part of the decision where the hearing officer considered likelihood of confusion. Furthermore, he argued that this principle strongly supported a finding of likelihood of confusion given the hearing officer’s findings that (i) AVEDA was highly distinctive, (ii) some of the goods were identical and (iii) overall there was moderate similarity between the marks. Against that, counsel for Dabur pointed out that the hearing officer stated at [51] that he was applying the global approach and that it was inconceivable that an experienced hearing officer would have failed to take such a well-known principle into account. I have to say I am somewhat doubtful that the hearing officer did properly apply this principle, but it is not necessary for me to reach a conclusion on this point in the light of my conclusion with respect to the first ground of appeal.

Aveda’s fourth ground of appeal is that the hearing officer incorrectly assessed the attributes of the average consumer. It suffices to say that I am unpersuaded that he made any error in this respect.

Aveda’s fifth ground of appeal is that the hearing officer erred in concluding that some of the respective goods were not similar. Under this heading some minor criticisms were made of the hearing officer’s decision which I do not regard as coming near to demonstrating an error of principle. More significantly, however, Aveda contends that the hearing officer’s assessments at [28], [32] and [34] were flawed for the following reasons.

Aveda’s trade mark attorney argued that the hearing officer’s reasoning at [28], which was to the effect that “cleaning, polishing, scouring and abrasive preparations” did not include, and were not even similar to, “cleaning, polishing, scouring and abrasive preparations for use on the body” was perverse. In my judgment, however, there is a more fundamental problem with it, which is that it is inconsistent with his earlier acceptance that Aveda had demonstrated use of “cleaning, polishing, scouring and abrasive preparations for use on the body” as a sub-category of “cleaning, polishing, scouring and abrasive preparations”. That acceptance necessarily led to the conclusion that “cleaning, polishing, scouring and abrasive preparations” were identical to “cleaning, polishing, scouring and abrasive preparations for use on the body” in the sense that the former encompassed the latter. But the hearing officer’s reluctance to accept that conclusion indicates that he ought to have questioned the premise.

What lies behind this is that Aveda accepts that it had made no use of its trade mark in relation to household or industrial cleaning, polishing, scouring and abrasive preparations during the relevant period, but contends that it had used the mark in relation to cleaning, polishing, scouring and abrasive preparations for use on the body. More specifically, Aveda relies upon use in relation to exfoliants, body polishers and dermabrasives.

I reviewed the correct approach to the construction of the specification of goods and/or services of a registered trade mark in Omega Engineering Inc v Omega SA [2012] EWHC 3440 (Ch) at [21]-[33]. In short, the words used in the specification should be given their natural and usual meaning. They should neither be given such a broad interpretation that the limits of the specification become fuzzy, nor strained to produce a narrow meaning.

I have reviewed the law in relation to partial revocation for non-use in Stichting BDO v BDO Unibank Inc [2013] EWHC 418 (Ch) at [51]-[62]. In short, the tribunal is required to arrive at a fair specification by identifying and defining not the particular examples of goods for which there had been genuine use, but the particular categories of goods they should realistically be taken to exemplify. For that purpose the terminology of the resulting specification should accord with the perceptions of the average consumer of the goods or services concerned. It is for the tribunal to decide what is a fair specification, not the parties. Furthermore, the tribunal is obliged to ensure that the resulting specification is clear and precise: see BDO at [49].

The hearing officer considered that “cleaning, polishing, scouring and abrasive preparations” would be understood by the average consumer as referring to household and industrial preparations. If so, then the conclusion he should have drawn was that Aveda had not demonstrated any use of its mark in relation to such goods. On that basis, Aveda’s use in relation exfoliants, body polishers and dermabrasives would be use in relation to some other category or sub-category of goods.

The Oxford English Dictionary defines “exfoliant” as meaning “a cosmetic product designed to remove dead cells from the surface of the skin”. It does not have a definition for “dermabrasive”, but it defines “dermabrasion” as “the removal of superficial layers of the skin with a rapidly revolving abrasive tool”. My understanding is that a dermabrasive in the present context is similar to an exfoliant, but perhaps a little more intensive. Equally, I understand a body polisher to be similar, but perhaps a little less intensive.

I see no reason to differ from the hearing officer as to the meaning of “cleaning, polishing, scouring and abrasive preparations”. In my judgment the average consumer would not regard exfoliants, body polishers and dermabrasives as exemplifying a sub-category of “cleaning, polishing, scouring and abrasive preparations for the body”. Rather, he or she would regard them as examples of “skin care products, namely … cleansing lotions” and/or “body cleansing lotions”.

It remains necessary to consider whether “cleaning, polishing, scouring and abrasive preparations” in Dabur’s Class 3 specification are similar to any goods in respect of which Aveda proved use or was not required to prove use. The closest goods are “skin care products, namely … cleansing lotions” and/or “body cleansing lotions”, as exemplified by exfoliants, body polishers and dermabrasives. Applying the criteria laid down in Case C-39/97 Canon KK v Metro-Goldwyn-Mayer Inc [1998] ECR I-5507 at [23] and British Sugar plc v James Robertson & Sons Ltd [1996] RPC 281 at 296-297, I consider that these are not similar. Their uses, physical nature and trade channels are all different, and they are neither competitive nor complementary. An overlap in users is not enough.

It follows that I have reached essentially the same conclusion as the hearing officer, but by a different route.

Aveda’s trade mark attorney argued that the hearing officer’s reasoning at [32] was flawed because it relied on his reasoning at [28]. In my judgment the hearing officer asked himself the wrong question here: he should have asked himself whether dentrifices were similar to “skin care products, namely … cleansing lotions” and/or “body cleansing lotions”, as exemplified by exfoliants, body polishers and dermabrasives. In my view they are reasonably similar. Their uses, users, physical nature and trade channels are all similar, and they could be regarded as complementary.

As Aveda’s trade mark attorney pointed out, the hearing officer was in error in treating the relevant term of Dabur’s specification as being “preparations for destroying fungicides”: the correct term is simply “fungicides”. He also pointed out that there are medicated skin creams which are used to treat fungal infections. On that basis, he argued that the hearing officer’s reasoning at [34] was inconsistent with what he had said in [33]. I agree with this. It follows that I consider that the hearing officer should have concluded that there was similarity in this respect.

It follows from my previous conclusions that it is necessary for me to re-assess the likelihood of confusion not only in relation to those goods which the hearing officer found to be highly or reasonably similar, but also in relation to dentrifices and fungicides. Taking all of the relevant factors into account, and applying the interdependency principle, I conclude that there is a likelihood of confusion in relation to all of these goods.

The appeal succeeds except in relation to cleaning, polishing, scouring and abrasive preparations in Class 3. Accordingly Dabur’s mark may proceed to registration only in respect of the following goods:

“Class 3

Bleaching preparations and other substances for laundry use; cleaning, polishing, scouring and abrasive preparations.

Class 5

Veterinary preparations; sanitary preparations for medical purposes; disinfectants; food for babies; plasters; materials for dressings; material for stopping teeth, dental wax, preparations for destroying vermin; herbicides.”