O2 Holdings Ltd & Anor v Hutchison 3G Ltd

The trial judge, Lewison J, opened his judgment [2006] EWHC 534 (Ch); [2006] RPC 699 saying “this is a case about bubbles.” And so it is. But it is also about a lot more. How aggressively does EU law permit comparative advertising to go? Particularly can the advertiser not only use his rival’s main name or trade mark but also, honestly, his subsidiary registered trade marks including marks embodying his trade imagery (what the Americans call trade dress)? The question deep down involves a decision based upon the philosophy of how competitive the law allows European industry to be.

For that reason I have come to the firm conclusion that an EU-wide answer is needed. That can only be provided by the European Court of Justice, for both sides agreed that the position was not so straightforward as to be acte clair. Mr Hobbs QC for the defendants, “H3G,” indicated that his client would prefer it if we just decided the case, leaving the question of a reference to the House of Lords. But the House (even when considering a Petition for leave to appeal) is, pursuant to Art. 234 of the EU Treaty bound to refer a question of EU law if it is not acte clair. Each side sees a question of principle at stake so there is no reasonable prospect of settlement. So if we decided matters for ourselves, whichever way we went, and whether or not we granted leave to appeal to the House of Lords, a reference in the end would be inevitable. So it is better to refer now rather than allow the delay and expense which would otherwise be involved.

At the outset of these proceedings the claimants, “O2”, raised a large number of complaints. By trial these had been whittled down (by a mixture of self-culling and the exercise of the court’s case management powers by the Vice-Chancellor as he then was) to allegations of infringement of four registered trade marks. All four of these consist of static pictures of bubbles. The specifications of goods and services of all the marks include telecommunications apparatus and telecommunications services. The marks were divided into pairs; one pair was alleged to be infringed pursuant to Art.5(1)(b) of the Trade Marks Directive (implemented by s.10(2) of the Trade Marks Act 1994) and the other pair pursuant to Art. 5(2) (implemented by s.10(3)). Lewison J had to decide a great many issues, many of which are not pursued on appeal. These included the validity of the registrations (now accepted) and the issue of infringement of the pair said to infringe by virtue of Art. 5(2) (which he held did not infringe). There is no appeal from the latter decision. So that leaves just the issues of infringement of the pair of marks said to infringe by virtue of Art.5(1) and whether, if so, there is a defence.

I reproduce the marks here:

The parties were agreed that the effect of some cryptic headings and entries in the register (which I need not set out) meant no more than that Technical was registered in both blue and monochrome and Fizz just in blue – just as the pictures are on the Register. There was no suggestion that the rights in them were limited to the colours of registration.

In the market place O2 used bubble imagery in a host of ways. The evidence clearly established that images of bubbles in water (particularly against a graduated blue background) in the context of mobile phones had become associated with O2 and none other. O₂ is the chemical symbol for the oxygen molecule and the idea is to link the name O2 with the logo O₂ and the imagery of the life-giving gas. The Judge describes the use in more detail at [8]-[13] and the resulting success in implanting in the public mind the image of bubbles as denoting O2’s mobile phone business at [29]-[33]. His conclusions are not challenged.

H3G, also a mobile phone service provider, was a late entrant to the UK market, starting its service in March 2003. It calls its service “3”. By then four providers, Vodafone, Orange, O2 and T-mobile, were well established and 3 did not find the going easy. In March 2004 3 launched a pay-as-you-go service called “Threepay” Again the going was not easy. So it was decided to advertise against the big brands comparatively. The intention was to do so “in a robust but fair way.”

After checking with TV advertising voluntary regulators and their own solicitors, H3G started their comparative advertising campaign using not only TV but other advertisements too. We have seen the TV advertisements upon which the case is based. O2 and their rivals were the subject of comparisons, each main player being the subject of an individual comparative advertisement. The general scheme of the advertisements is to begin by using the name and something of the imagery of the rival, then to use the Threepay and 3 imagery with a message that it was cheaper in a specific way. The imagery used for the comparison against Orange was a spinning orange square, that for Vodafone flying abstract shapes resembling commas, that for T-Mobile a pink triangle and that for O2, bubbles.

The judge described the “anti-O2” advertisement as follows but with the accurate caution that it is difficult to describe in words:

Note that, for the purposes of the proceedings, it was accepted that the price comparison was true and, moreover, that as whole, the advertisement was not misleading in any way. In particular it did not suggest any form of trade connection between O2 and 3. The average member of the public would see the use of O2 and the bubbles as a reference to O2 and its imagery and realise that this was an advertisement from a trade rival, 3, saying it was cheaper.

So far as the points in issue before us are concerned, the Judge held that:

O2, whose case was argued by Mr Richard Arnold QC and Mr Mark Vanhegan, appeal the finding that there was a defence. H3G, whose case was argued by Mr Geoffrey Hobbs QC and Miss Emma Himsworth, contend that the Judge was wrong to find that the advertisement fell within Art.5(1)(b) at all, as well as resisting O2’s appeal.

Mr Hobbs advanced two reasons as to why the advertisement did not fall with Art.5(1)(b) and several reasons as to why there was a defence. The logical way to approach the individual points thus raised on this appeal is thus not in the order they were argued but as follows. I set them out in the form of the propositions advanced by Mr Hobbs.

Where a registered mark is used only descriptively, Mr Hobbs submits that it does not fall within Art5.1(b) at all. To understand the argument it is necessary to set forth not only Art.5(1) but also the “descriptiveness” defence, Art.6.1(b):

“Art. 5 Rights conferred by a trade mark

1.

The registered trade mark shall confer on the proprietor exclusive rights therein. The proprietor shall be entitled to prevent all third parties not having his consent from using in the course of trade:

(a)

any sign which is identical with the trade mark in relation to goods or services which are identical with those for which the trade mark is registered;

(b)

any sign where, because of its identity with, or similarity to, the trade mark and the identity or similarity of the goods or services covered by the trade mark and the sign, there exists a likelihood of confusion on the part of the public, which includes the likelihood of association between the sign and the trade mark.

Art.6 Limitation of the effects of a trade mark

“1.

The trade mark shall not entitle the proprietor to prohibit a third party from using, in the course of trade,

(a)

his own name and address;

(b)

indications concerning the kind, quality, quantity, intended purpose, value, geographical origin, the time of production of goods or of rendering of the service, or other characteristics of goods or services;

(c)

the trade mark where it is necessary to indicate the intended purpose of a product or service, in particular as accessories or spare parts;

provided he uses them in accordance with honest practices in industrial or commercial matters.”

Art.5.1(b), apart from requiring a consideration of the similarity of the goods or services (there is, of course, identity in this case), requires the similarity of the sign used by defendant to that of the registered mark to be considered. Taking both considerations in mind, one has to form a “global assessment of the likelihood of confusion.”

The “global assessment” test is now very firmly established in the Court’s case-law. It was first articulated in SABEL v Puma Case C-251/95[1997] ECR I-6191 para. 22 and has been repeated in many other cases since, down to L’Oréal v OHIM Case C-235/05 where the Court described the test as “settled,” para. 34. The person one has to consider is:

“an average consumer who is reasonably well-informed and reasonably observant and circumspect.”

This notional person was first described outside the context of trade marks in Gut Springenheide Case C-210/96 at [31] (about non-misleading marketing of eggs) and carried over into trade marks by Lloyd Schuhfabrik Meyer & Co GmbH v Klijsen Handel BV Case C-342/97 [1999] ECR I-3819 para. 26.

Mr Hobbs emphasises that the test requires, as is said by the court, “all the circumstances of the case” to be taken into account. So, he submits, in an infringement action one must consider exactly how the mark is being used – the context of use is crucial.

He goes on to say that if one finds that in the actual context of use the defendant is using the mark complained of “purely descriptively” then he is not to be regarded as using the “sign” at all for the purposes of Art 5.1(b): a pure description is just not a “sign”. So one therefore does not get as far as considering whether a limitation under Art.6 may apply.

Mr Hobbs’ principal authority for this submission is Hölterhoff v Freiesleben Case C-2/00 [2002] ECR I-4187. The plaintiff had registered marks “Spirit Sun” and “Context Cut” for what are essentially uncut diamonds (“diamonds for further processing as jewellery”). The facts as summarised by the Court were set out at [8] and [10], namely that the marks were only used orally in the course of negotiation and “solely to describe the qualities, and more specifically the type of cut.”

The referring court asked the following question:

“Does an infringement of a trade mark in the sense contemplated in Article 5(1)(a) and (b) of Directive 89/104/EEC occur where the defendant reveals the origin of goods which he has produced himself and uses the sign in respect of which the plaintiff enjoys protection solely to denote the particular characteristics of the goods he is offering for sale so that there can be no question of the trade mark used being perceived in trade as a sign indicative of the firm of origin?”

The ECJ answered a rather narrower question, sufficient for disposal of the case in hand. It said:

“[16] … it is sufficient to state that, in a situation such as that described by the national court, the use of the trade mark does not infringe any of the interests which Article 5(1) is intended to protect. Those interests are not affected by a situation to which:

-

the third party refers to the trade mark in the course of commercial negotiations with a potential customer, who is a professional jeweller,

-

the reference is made for purely descriptive purposes, namely in order to reveal the characteristics of the product offered for sale to the potential customer, who is familiar with the characteristics of the products covered by the trade mark concerned,

-

the reference to the trade mark cannot be interpreted by the potential customer as indicating the origin of the product.

[17] In those circumstances, without its being necessary, in the present case, to discuss further what constitutes the use of a trade mark within the meaning of Article 5(1)(a) and (b) of the directive is to be interpreted as meaning that the proprietor of a trade mark cannot rely on his exclusive right where a third party, in the course of commercial negotiations, reveals the origin of goods which he has produced himself and uses the sign in question solely to denote the particular characteristics of the goods he is offering for sale so that there can be no question of the trade mark used being perceived as a sign indicative of the undertaking of origin.”

Mr Hobbs’ submission thus was that “in order to be within the grasp or scope of Art.5(1) it is necessary for the defendant to be using the ‘offending’ sign for the purpose of distinguishing the goods or services in which he is trading, his goods or services.”

He supported the argument by reference to Anheuser-Busch, Case C-252/02, [2004] ECR I-10989 at [58]-[85]. He submitted that the thrust of this case was that Art. 5.1 did not come into play at all (was not “engaged” in the language of human rights law) where the use complained of in no way jeopardised the essential function of a trade mark. In the present case (and indeed in most comparative advertising cases) the whole point of the use complained of is to use the registered mark in a context where it solely refers to the mark owner’s own goods or services. If you don’t do that you can’t make the comparison.

Finally Mr Hobbs pointed out that the present point may in effect already be the subject of the pending reference in Adam Opel v Autotec, Case C-48/05. There the Opel car company is suing on its registered trade mark (registered for toys) a toy car company for using, on its replica Opel cars, the well-known Opel logo. The toy-maker says he is not using the trade mark for toys at all. The first question referred is whether or not the use can be regarded as use of the trade mark for toys and hence within Art.5.1. Advocate-General Ruiz-Jarabo Colomer suggests the answer is no. His reason is that the average consumer would not perceive the use to indicate a trade origin of the toys themselves, rather as merely a reproduction of what is on the real cars (Opinion 7.3.2006, [23]-[46]). The use does not affect any of the aspects that Art 5.1 aims to protect.

Mr Arnold submitted that the Art. 5.1 test had no threshold of the sort suggested. It called only for a straight comparison between the sign complained of and the registered mark. He first pointed out that that must be the operative test at the application stage under Art.4.1 – where the mark applied for is devoid of any context of use and is to be compared with an earlier trade mark. Puma was, for instance, such a case and was indeed the case where the “global” test was first articulated.

He added that it was because of this test that a defendant could infringe even if he was just dealing in the mark owner’s own goods, as for instance in the case of parallel imports unless protected by Art.7. He took us to Silhouette Case C-355/06 [1998] ECR I-4799 at [25], Davidoff Joined Cases C-414/99, C/415/99 and C-416/99 [2001] ECR I-8691 at [39], Peak Holding Case C-16/03 [2004] ECR I-11313 at [30], BMW Case C-16/03 [1999] ECR I-905 at [31]-[42], [45] and the opinion of Sharpston A-G in Boehringer Ingelheim Case C/415/99 (6 April 2006) at [10]-[20].

Moreover, submitted Mr Arnold, in a comparative advertisement the advertiser is actually using the rival’s mark in two ways: first to refer to the rival’s goods or services and then as a way of describing the advertiser’s own goods. When a man says “my film is half the price of Kodak and just as good” he is using Kodak in both of those ways. Here, submits Mr Arnold, the bubble imagery is used to refer to O2 and also by way of contrast to Threepay’s prices. So, unlike the Adam Opel case there is some sort of use in relation to the defendant’s goods.

As for Hölterhoff, Mr Arnold submitted that it was an unsatisfactory case. The facts were not certain and in particular it was not clear whether the registered marks were in fact recognised generic descriptions in the trade for a kind of cut. Mr Arnold submitted that this case was distinguishable in that the bubbles are not used purely descriptively because they are used referentially too. There was, suggested Mr Arnold, a claim by the defendant to that effect recorded at [13] in Advocate-General Jacobs’ opinion.

Mr Arnold also relied on Arsenal where in the context of a use which at the point of sale made it plain that the goods were “unofficial” there was nonetheless infringement, Case C-206/1 [2002] ECR I-20273, [2003] RPC 696. [I interpolate to say that it seems obvious that trade marks have significance beyond the point of sale: think of picking up a cup to look underneath to see who made it. The potter’s mark may indeed be the very first and oldest kind of trade mark,]

Mr Hobbs answered that by saying that in all the cases cited the defendant was in some way actually dealing in the mark owner’s goods, as for instance in the case of a parallel importer. That was not the case here, nor was it in Hölterhoff.

I am clear that the position is not clear. Nor do I think that the decision in Adam Opel will necessarily cover the context of the present case. I would therefore refer the following question to the Court:

“1 “Where a defendant in the course of trade uses a sign in a context purely for the purpose of comparing the merits (including price) of his goods or services with those of the trade mark owner and in such a way that it cannot be suggested that the essential function of the trade mark to guarantee the trade mark as an indication of origin is in any way jeopardised, can his use fall within either (a) or (b) of Art.5.1 of Directive 89/104?

In case it would be of assistance to the court, I offer my view. I think the answer is “No.”. I say this principally because it accords with the key function of trade marks – as repeatedly identified by the Court. The defendant is in no way using the mark to indicate the trade origin of his goods or services, not even that he is dealing in those of the trade mark owner. There is simply no reasonable need for trade mark law to cover this kind of use. If the comparative advertising is unfair, derogatory or otherwise fails to comply with all the conditions of the CAD, there are independent mechanisms for dealing with it. That is what the CAD requires. They come into play whether or not the “victim” of the advertising has a registered mark. There is no need for the law of registered trade marks to get involved with any of this.

I also see no good reason to say that the consideration (for either Art.5.1(a) or (b)) is limited to a comparison between the defendant’s sign with the registered mark. Indeed the very Article clearly calls for an examination of the context of the use – you have to consider how the defendant is using the sign complained of to answer the basic question of whether he is using it “in the course of trade.” So it is particularly artificial then to go on to try to isolate the sign of which complaint is made devoid from the context of its use. Nor am I impressed by the fact that the same test applies at the registration stage. For then any notional fair use of the mark applied for must be considered, whether it has in fact been used at all.

I go little into the case law because this question is fundamental and can best be answered by an application of first principles. There is indeed a real danger if one does go into a welter of cases that one loses sight of such principles. European trade mark law is in danger of becoming a forest of case law where everyone loses their way.

However if one does consider the case-law then I see no real distinction from Hölterhoff. For I do not read the paragraph [13] referred to by Mr Arnold as saying that the name itself was old or purely descriptive, only that the style of cut was. If the words registered as trade marks were indeed old and purely descriptive it is difficult to see why the registrations were not invalid. Enquiries made by the parties suggested that a defence of invalidity can be raised in German trade mark infringement proceedings – even if not directly in the infringement court at least in the Federal Patent Court. Either way an argument as to invalidity would surely have been raised. Besides, even if the words were a generic description, that is no answer – for on Mr Arnold’s argument the only point that could come in is by way of defence under Art. 6.1.

I turn to the other, related, way Mr Hobbs argues that O2’s use does not fall within Art.5.1(b). He submits that if one has regard to the whole context of use there is clearly no risk of confusion or association. On the contrary, Threepay and the associated animated “3” are used in contradistinction both to the name O2 and the bubbles. If one applies the well settled “global appreciation” test to the use in context, the average reasonably well-informed consumer will not be misled.

The difference between this argument and the former is that it is confined to Art. 5(1(b) and its requirement for a likelihood of confusion. Otherwise it is essentially the same. The question I propose covers Art 5.1, both limbs and therefore covers this case too. For the same reasons I would decide the answer “no.”

If the ECJ answers the foregoing question as I suggest, there will be no need to go further. But if the answer is “yes” so that the H3G’s use falls within Art.5.1 then the question of whether it has a defence arises. The first way Mr Hobbs puts his case is to say that there is a defence under Art.6.1 independently of the CAD.

That required the relationship, if any, between Art.6.1 and the CAD to be considered. As a matter of history, the first relevant provision was the 1984 Misleading Advertising Directive. This contained no provision about comparative advertising. In 1989 came the TMD, part-harmonising national trade mark laws, followed by the TM Regulation 40/94 creating the Community TM. Neither of these explicitly dealt with comparative advertising. Only in 1997, with the CAD, was there explicit Community harmonisation specifically directed at comparative advertising.

Different national commercial cultures in the past took different views as to the allowability of comparative advertising. Some, such as Germany, took the view that practically all forms of comparative advertising, even if true, were unfair competition – the notion was that the newcomer should not “piggyback” on the established reputation of another. Others took the view that, provided it was true, comparative advertising was all right – legitimate competition. It did enable the newcomer to “piggyback” a known brand, but the interests of consumers, competition, and a free market allowed this. Such is the basic UK common law position. There was a statutory qualification introduced in 1938 for “Part A” registered marks, see Bismag v Amblins (1940) 57 RPC 209, but the provision concerned, s.4 of the 1938 Trade Marks Act, has of course now gone.

I here set out the key recitals and operative provisions of the CAD. The operative provisions were inserted by amendment into the 1984 Directive as Art.3a. The recitals are not so inserted, but nonetheless must obviously be taken into account to construe it.

Recitals of the CAD

“(10)

Whereas the international conventions on copyright as well as the national provisions on contractual and non-contractual liability shall apply when the results of comparative tests carried out by third parties are referred to or reproduced in comparative advertising;

(11)

Whereas the conditions of comparative advertising should be cumulative and respected in their entirety; whereas, in accordance with the Treaty, the choice of forms and methods for the implementation of these conditions shall be left to the Member States, insofar as those forms and methods are not already determined by this Directive;

(12)

Whereas these conditions should include, in particular, consideration of the provisions resulting from Council Regulation (EEC) No 2081/92 of 14 July 1992 on the protection of geographical indications and designations of origin for agricultural products and foodstuffs (7), and in particular Article 13 thereof, and of the other Community provisions adopted in the agricultural sphere;

(13)

Whereas Article 5 of First Council Directive 89/104/EEC of 21 December 1988 to approximate the laws of the Member States relating to trade marks (8) confers exclusive rights on the proprietor of a registered trade mark, including the right to prevent all third parties from using, in the course of trade, any sign which is identical with, or similar to, the trade mark in relation to identical goods or services or even, where appropriate, other goods;

(14)

Whereas it may, however, be indispensable, in order to make comparative advertising effective, to identify the goods or services of a competitor, making reference to a trade mark or trade name of which the latter is the proprietor;

(15)

Whereas such use of another's trade mark, trade name or other distinguishing marks does not breach this exclusive right in cases where it complies with the conditions laid down by this Directive, the intended target being solely to distinguish between them and thus to highlight differences objectively.”

Operative provisions of the CAD, Art 3a of the 1984 Directive

“1.

Comparative advertising shall, as far as the comparison is concerned, be permitted when the following conditions are met:

(a) it is not misleading according to Articles 2 (2), 3 and 7 (1);

(b) it compares goods or services meeting the same needs or intended for the same purpose;

(c)

it objectively compares one or more material, relevant, verifiable and representative features of those goods and services, which may include price;

(d)

it does not create confusion in the market place between the advertiser and a competitor or between the advertiser's trade marks, trade names, other distinguishing marks, goods or services and those of a competitor;

(e)

it does not discredit or denigrate the trade marks, trade names, other distinguishing marks, goods, services, activities, or circumstances of a competitor;

(f)

for products with designation of origin, it relates in each case to products with the same designation;

(g)

it does not take unfair advantage of the reputation of a trade mark, trade name or other distinguishing marks of a competitor or of the designation of origin of competing products;

(h) it does not present goods or services as imitations or replicas of goods or services bearing a protected trade mark or trade name.”

There is also a definition of “comparative advertising” inserted as Art. 2a:

“'2a 'comparative advertising' means any advertising which explicitly or by implication identifies a competitor or goods or services offered by a competitor;”

And the unamended original 1984 Directive contains a definition of “Advertising”:

“’Advertising’ means the making of a representation in any form in connection with a trade, business, craft or profession in order to promote the supply of goods or services, including immovable property, rights and obligations”

It is worth importing these definitions into the new Art. 3a so as to see more clearly the width of what it covers:

“The making of a representation in any form in connection with a trade, business, craft or profession in order to promote the supply of goods or services, including immoveable property, rights and obligations which explicitly or by implication identifies a competitor or goods or services offered by a competitor shall as far as the comparison is concerned, be permitted when the following conditions are met.”

By writing out the opening words of Art. 3a in this way one can see clearly how even an implied reference to a competitor is brought in.

Now there are some oddities about the CAD. First and foremost is the fact that there is no explicit provision which says that compliance with the Art.3a conditions is a defence to a trade mark infringement suit or indeed to any other intellectual property or quasi-intellectual property suit. A whole range of intellectual property rights may be invoked to try to prevent comparative advertising. Registered trade marks are but one of these. Apart from them copyright or design rights (registered or unregistered) could all come into play, as also PDO’s (Protected Designations of Origin) and general unfair competition rules. Copyright could be of great significance, for many trade marks are also the subject of copyright – indeed those in suit here are photographs and so will have an associated copyright.

One has the uneasy feeling that those responsible for this legislation had but a hazy idea of how various kinds of IP rights could apply to comparative advertising. One’s confidence is not increased by reading the travaux preparatoires. The Explanatory Memorandum to the Draft of 1991 COM(91)147 final – SYN of 21^st June 1991 does not consider the different kinds of IP rights which may be invoked to try to stop comparative advertising It even has a passage saying:

“4.5

The general authorisation of comparative advertising requires some explanation of its relation to patent rights, especially the trade mark law.”

This suggests no more than a populist understanding of IP. Expressions used by journalists such as “to copyright a trade mark” come to mind.

Nor are the recitals particularly clear about the relationship of the CAD to IP rights. Recital 10 refers to copyright, but only in the context of a third party’s copyright in reports of comparative tests. Doubtless the sort of test carried out by consumer organisations is what the framers had in mind here – they were concerned that the copyright rights of third parties should not be affected. What the Directive does not deal with is any copyright belonging to the rival of the comparative advertiser.

Recital 12 refers to PDOs, but only in the context of their general enforcement. It is not clear whether the intention is to exclude them from the CAD – a problem for another day.

Recitals [13]-[15] are about trade marks. The parties are in dispute about their effect, and particularly whether or not they import a requirement of “indispensability’ into the explicit conditions of Art.3a.

No other IP or quasi-IP rights are referred to at all in the recitals. One naturally asks whether or not enforcement of rights in these are excluded where a comparative advertisement complies with the Art. 3a conditions.

Some other features of the CAD must be observed. Firstly it does not require Member States to create anything in the nature of an IP right – an independent right of the “target” of comparative advertising to sue or prosecute where the advertising is non-3a compliant. If the target has an independent right to sue (e.g. on a trade mark or copyright) the position is simply that a non-complying defendant cannot invoke the CAD by way of defence.

Next, Art 3a sets out conditions under which comparative advertising is permitted. It does not actually say that comparative advertising which does not comply with the conditions is not permitted. But this seems implicit, particularly from the fact that Recital 23 says that a purpose of the CAD is to lay down “uniform general principles” – a purpose which would not be achieved if different Member States allowed comparative advertising which did not comply with Art. 3a.

The fact there is provision about ensuring compliance confirms this view. Art. 5 of the 1984 Directive as amended requires that “Member States ensure that adequate and effective means exist … for compliance with the provisions on comparative advertising.” In the UK this requirement has been achieved by a Statutory Instrument made under s.2 of the European Communities Act, 1972. It is SI 1998/915 as amended. There is no need to spell out the machinery here.

The upshot of this discussion is that because of the lack of clarity of the legislation it will in due course be up to the ECJ to decide whether compliance with the Art. 3a conditions affords a complete defence to any private right action based on any IP or quasi-IP right and aimed at preventing comparative advertising.

I do not think, however, such a wide question is called for in this case. This is because, as it seems to me, Art.6.1 of the TMD covers the position. The reasoning is in two parts. First, Art.6.1(b) must surely extend to any case of comparative use. Its language is:

“indications concerning the kind, quality, quantity, intended purpose, value, geographical origin, the time of production of goods or of rendering of the service, or other characteristics of goods or services;”

I cannot think of any comparison which would not fall within one or more of these indications.

Next, such indications are only allowed as a defence where the defendant “uses them in accordance with honest practices in industrial or commercial matters.” A man whose comparative advertisement is not Art.3a compliant cannot be regarded as so acting. True he may not be actually dishonest in the sense of misleading people, but he would be breaking the law in some way, for there is necessarily an enforcement mechanism as required by Art.5. In that sense his use would not be honest. It means that even a non-misleading but disparaging use would now be caught.

I regard this reasoning as so strong as to be acte clair and see no need to refer a question about it. I therefore reject Mr Hobbs’ third point set out above but accept his fourth point. Mr Arnold also accepts that if a comparative advertisement complies with the CAD, there is a defence to a trade mark infringement action. So there is no need to refer a question about that either.

I would only add that, insofar as part of my first instance decision in BA v Ryanair [2001] FSR 541 at [26]-[28] is at variance with my present conclusion, I reject it. By way of mitigation I would point out that the argument was different from that here, based as it was around the “home-grown” s.10(6) of the UK Trade Marks Act. The Judge held that this provision adds nothing to the CAD. Neither side relied upon it before us. It is a pointless provision (and could not apply to a Community TM). It should be repealed as an unnecessary distraction in an already complicated branch of the law.

Where the parties were in dispute was over the meaning of the Art.3a conditions. Mr Arnold submitted that the Art.3a conditions included a requirement of necessity. He then went on to submit that on the facts there was no need to use the bubble imagery at all to make the price comparison. Moreover, even if there was, there was no need to use a mark confusingly similar to the registered mark – a variant of it. Mr Hobbs disputed all of that. He submitted that the Art 3a conditions did not include any requirement of necessity – and if they did, the use made was indeed “necessary” because it made the advertisement more effective.

I turn to consider Mr Arnold’s argument. It is centred on Recital 14 to the CAD. That says,“it may be indispensable, in order to make comparative advertising effective, to identify the goods or services of a competitor, making reference to a trade mark or trade name of which the latter is the proprietor.” That recital, it is submitted, tells you how to construe the eight conditions of Art.3a. Of these Mr Arnold relied on (c), (d) and (g).

The strongest of these is clearly (g), and I say no more of the others, not least because I do not really see how they could come into play at all, and certainly not beyond anything in (g). Mr Arnold says that unless the use of the trade mark relied upon is “indispensable” to the comparative advertising, then its use must be regarded as taking “unfair advantage” of the reputation of the trade mark. An “advantage” is unfair unless it is “indispensable.”

So, says Mr Arnold, here it is obviously essential to use the name O2 to advertise comparatively against it. But no more is essential (“indispensable”), so the bubbles are taking an unfair advantage.

He reinforces his argument by coming at it from the point of view of Art.6.1 of the TMD. The cases, he submits, show that where a defendant has referred to the registered mark, he has no defence under Art.6.1 if the reference is not necessary. Putting it the other way round, an unnecessary reference will be one which is not “in accordance with honest practices etc.” This he developed, saying that the Court had repeatedly said that the proviso to Art.6.1 “constitutes in substance the expression of a duty to act fairly in relation to the legitimate interests of the trade mark proprietor.” He referred us to BMW at [61], Gerolsteiner Brunnen Case C-100/02 [2004] ECR I-691 at [24], Anheuser-Busch Case C-245/02 [2004] ECR I-10989 at [82] and Gillette v LA Case C-228/03 [2005] ECR I-2338 at [41]. One is not so acting if one makes an unnecessary reference was his point.

In reply Mr Hobbs submitted that Mr Arnold was in effect creating a ninth condition in Art.3a, one of indispensability. That is simply not there. Moreover, he submitted, such a contention was entirely inconsistent with the principle of construction of the CAD laid down by the court in Toshiba, Case C-112/99 [2001] ECR I-7945, and repeated in Pippig, Case C-44/01 [2003] ECR I-3095. That principle is that the “the conditions required of comparative advertising must be interpreted in the sense most favourable to it” (Toshiba [37], Pippig [52]).

He elaborated on Pippig by reference to its facts. It was an unfair competition claim in Austria. Pippig was a well-known firm of specialist opticians. The defendant, Harltauer, was a major distribution chain selling a range of different products, including spectacles. Hartlauer put out a leaflet and a TV advertisement in which it made price comparisons with Pippig. The TV advertisement included a picture of one of its shops and its company logo. The defence was the CAD. The referring court asked a number of questions, one of which was:

“Is Art 3a(1)(e) of the directive to be interpreted as meaning that the information on the identification of the competitor must be restricted to the extent absolutely necessary and it is therefore not permitted if, in addition to the competitor’s name, it company logo (if it exists) and its shop are shown.”

In its judgment the Court noted particularly that it was Pippig’s contention:

“that it was not indispensable for the advertiser to appear ‘triumphantly’ before the shop of the competitor whose products were being compared”

So, submitted Mr Hobbs, the very point as to “indispensability” was alive in the case. And it was in that context that the Court ruled [84]:

“Secondly Art 3a(1) of Directive 84/450 does not prevent comparative advertising, in addition to citing the competitor’s name, from reproducing its logo and a picture of its shop front, if that advertising complies with the conditions for lawfulness laid down by Community law.”

Mr Hobbs submitted that the Court could not have said that if there was an overall requirement of “necessity” or “indispensability.”

As to the suggestion that Art. 6.1 had its own inbuilt requirement of necessity, Mr Hobbs pointed out that that was indeed inbuilt so far as the spare parts etc. exception of Art.6.1(c) is concerned – it uses the word. But by contrast Art.6.1(b) did not use that word. There was no reason to suppose that it contained such a requirement by implication. And the closing proviso is derived from Art 10bis of the Paris Convention. It cannot in any way be tied in to the CAD.

Mr Arnold’s answer on Pippig was to say that it was confined to an interpretation of condition 3a(1)(e), was not a trade mark infringement case and was not concerned with condition 3a(g). The ruling left open the question of whether or not there was compliance with the latter.

I cannot be sure which of these rival arguments is right. It seems to me therefore that a reference is sensible. As to the question, the following may be appropriate:

“2.

“Where a defendant uses, in a comparative advertisement, the registered trade mark of another, in order to comply with Art. 3a of Directive 84/450 as amended, must that use be “indispensable” and if so what are the criteria by which indispensability is to be judged?”

Again I will give my own view. The answer is “no”. I say this for a variety of reasons. The first is short and simple: the principle of construction laid down in Toshiba and Pippig is inconsistent with the answer “yes.”

Next I cannot think that the Court in Pippig was confining itself to Art. 3a.1(e). To have said that using the shop-front is all right as far as that paragraph is concerned but not to have said that it is in trouble under another, namely Art.3a.1(g), would be to have given a wholly false impression as to the law.

Thirdly I do not accept Mr Arnold’s attempt to tie in the CAD with the closing words of Art. 6.1 of the TMD. They are derived from a wholly different international source. There is no reason why they should have corresponding contents.

Fourthly I cannot think of rational legal rules for the indispensability test proposed. After all no-one has to advertise comparatively at all. So in one sense if Mr Arnold is right, all comparative advertising does not comply with the CAD conditions. That is absurd. Recognising that, Mr Arnold accepts that the use of O2’s main trade mark is “indispensable”. But why is that so? Answer, because it makes the comparison effective. Why then should the use of a subsidiary mark such as the bubbles, aimed at making the advertising even more effective, be regarded otherwise? And is this not setting the court up to judge not legal questions but the sort of thing which is uniquely within the experience of advertisers and their advisors?

Fifthly I think the reliance on Recital 14 is misplaced. The Recital is saying that the freedom to advertise comparatively must even extend to the use of another’s trade mark or trade name. That is why, after the reference to trade mark rights in Recital 14, it opens with the words “It may, however ..” So it is simply not the basis for a restrictive interpretation of Art.3a(1)(g).

Sixthly the expanded (by incorporation of definitions) form of the opening words of Art. 3a makes it clear that provided the conditions are satisfied by “any representation ….which explicitly or by implication identifies a competitor”. These are wide words, hardly consistent with an overriding requirement of necessity.

I turn to the last point relied upon by Mr Arnold, that even if the CAD does permit the use of a “subsidiary” trade mark, it does not go so far as to permit use of a confusingly similar mark. His point is essentially founded on his earlier point that there is a requirement of indispensability. If the answer to question 2 is “no,” as I suggest, I think this third point will fall too. But if some of my suggested reasons are wrong then the point may be alive. So I would ask an additional question:

“3.

In particular, if there is a requirement of indispensability, does that requirement preclude any use of a sign so similar to the registered trade as to be confusingly similar to it?”

Again I will offer my view. I cannot see any reason why a requirement of necessity should include within it a requirement that the mark can only be used in a comparative advertisement exactly as registered. Such a requirement could lead to absurd results. Suppose, for instance, that the well-known Ford logo is registered as a trade mark (I expect it is). And suppose a defendant truly said “my cars are as good as FORD but cheaper.” He would not be using the mark exactly as registered, yet the plain word is obviously confusingly similar to the registered mark, the logo. To say that he could only make the comparison by using the registered logo and not the word itself would as a practical matter preclude any comparative advertising. That cannot be the intention of the CAD and is an absurd conclusion. Another absurdity is this: suppose the plain word “Ford” is also registered (as I expect it is). Then, if there were the suggested requirement, the defendant would have a defence to an infringement suit based on that mark, but not one based on the logo. That is absurd. So I would answer the question “no”.

In truth what has happened here is that H3G use the registered mark “O2” to make their comparative advertising effective and the bubbles, which are confusingly similar to the registered marks in suit, to make it even more effective. To construe the CAD as not permitting that would not be to interpret it in the “sense most favourable” to comparative advertising.

For the above reasons I would refer three questions to the ECJ. I do not regard those I have suggested as set in stone and they should be open to post-judgment discussion. On the other hand, I do not accept the draft questions proposed by Mr Arnold. To my mind they are insufficiently focussed.

I agree.

I also agree.