61989C0010

++++

My Lords,

I - The background to the case

1 This case comes before the Court by way of a reference for a preliminary ruling from the Bundesgerichtshof ( German Federal Supreme Court ). It is concerned primarily with the relationship between the principle of the free movement of goods laid down in Articles 30 to 34 of the EEC Treaty and the exception to that principle laid down in Article 36 thereof with regard to restrictions "justified on grounds of ... the protection of industrial and commercial property ". The present case constitutes a sequel to Case 192/73 Van Zuylen v HAG [1974] ECR 731 . Inevitably, that case and the present one will become known as HAG I and HAG II respectively . These are convenient epithets and I shall use them myself .

2 The plaintiff in the main proceedings, HAG GF AG ( hereafter "HAG Bremen "), is a German company based in Bremen . It has been in existence since 1906 and its main activities, arising from the invention of the first process for decaffeinating coffee, have long been the production and distribution of such coffee . In 1907 it had the trade mark "HAG" registered in its name in Germany . The following year the same mark was registered in its name in Belgium and Luxembourg . In 1927 it set up a subsidiary company in Belgium, trading as "Café HAG SA", which was wholly owned and controlled by it . In 1935 it transferred the Belgian and Luxembourg trade marks to the subsidiary . In 1944 the entire assets of the subsidiary, including the trade marks for Belgium and Luxembourg, were sequestrated as enemy property . The company was sold en bloc to the Van Oevelen family . In 1971 the trade marks, which had at some stage been converted into Benelux marks, were assigned to Van Zuylen Frères, a firm based in Liège .

3 When in 1972 HAG Bremen began exporting coffee to Luxembourg under the mark "Kaffee HAG", Van Zuylen Frères commenced infringement proceedings before a Luxembourg court . Those proceedings led to the preliminary ruling in HAG I, in which the Court held that :

"To prohibit the marketing in one Member State of a product legally bearing a trade mark in another Member State for the sole reason that an identical trade mark, having the same origin, exists in the first State, is incompatible with the provisions for the free movement of goods within the common market ."

4 The implications of that ruling seem clear . It was drafted in such wide terms as to give the impression that, if Van Zuylen could not rely on their Benelux trade mark to prevent HAG Bremen from selling coffee under that mark in Luxembourg ( and indeed Belgium ), neither could HAG Bremen rely on their German trade mark to prevent Van Zuylen from supplying the German market under the same mark . Van Zuylen did not, however, attempt to do that . For the next decade HAG Bremen continued to enjoy undisturbed use of the HAG trade mark in Germany .

5 In 1979 the firm Van Zuylen Frères was purchased by a Swiss company now called Jacobs Suchard AG, which is the market leader in coffee products in Germany . According to HAG Bremen' s observations, Jacobs Suchard AG disposed of the bulk of Van Zuylen' s coffee business, retaining only the shell of the firm and the HAG trade marks . The firm was transformed into a wholly owned subsidiary of Jacobs Suchard AG trading under the name SA CNL-SUCAL NV ( hereafter "HAG Belgium ").

6 In 1985 HAG Belgium began to supply decaffeinated coffee under the HAG trade mark to the German market . HAG Bremen, which maintains that "Kaffee HAG" has acquired the status of a famous brand in Germany and that its product is, by virtue of a new manufacturing process, superior in quality to the coffee supplied by HAG Belgium, applied to the competent German court for an injunction restraining HAG Belgium from infringing its trade mark . HAG Bremen succeeded before the Landgericht Hamburg and, on appeal, before the Hanseatisches Oberlandesgericht . HAG Belgium appealed to the Bundesgerichtshof, which referred the following questions to the Court for a preliminary ruling, under the third paragraph of Article 177 of the EEC Treaty :

"( 1 ) Is it compatible with the provisions on the free movement of goods ( Articles 30 and 36 of the EEC Treaty ) - having regard also to Article 222 - that an undertaking established in Member State A should, by virtue of its national rights in trade names and trade marks, oppose the importation of similar goods of an undertaking established in Member State B if, in State B, those goods have legally received a mark which :

( a ) may be confused with the trade name and trade mark reserved in State A to the undertaking established there, and

( b ) had originally existed in State B - albeit registered later than a mark protected in State A - for the benefit of the undertaking established in State A and had been transferred by that undertaking to a subsidiary undertaking set up in State B and forming part of the same concern, and

( c ) was, as a consequence of the expropriation in State B of that subsidiary, transferred as an asset of the sequestrated subsidiary ( together with that undertaking as a whole ) to a third party which, in turn, assigned the mark to the legal precursor of the undertaking which now exports the goods bearing that mark to State A?

( 2 ) Should the answer to the first Question be negative :

Would the answer to the above question be different if the mark protected in State A has become a 'famous' brand name in that State and it is probable that, as a result of the exceptional prominence which it enjoys, if the same mark is used by a third-party undertaking, the task of informing the consumer as to the commercial origin of the goods could not be accomplished without adverse repercussions on the free movement of goods?

( 3 ) Alternatively, again if the first question is answered in the negative :

Does the same answer hold good even if consumers in State A associate the mark protected in that State not only with a certain commercial origin but also with certain perceptions as to the characteristics, in particular the quality, of the marked goods and if the goods imported from State B under the same mark do not meet those expectations?

( 4 ) If the first, second and third questions are all answered in the negative :

Would the answer be different if the separate conditions set out in the second and third questions were cumulative and were all satisfied?"

Question 1

II - The fundamental issues raised by the question

7 If Question 1 is approached in the light of the Court' s ruling in HAG I, it may be said to raise two points of considerable importance . First, the case squarely raises the issue whether the doctrine of "common origin", as it is generally known, which the Court laid down in HAG I, is correct; and although the situations to which the doctrine applies are likely to be rare, so that the implications of maintaining or abandoning the doctrine are of relatively limited scope, yet the possibility of directly reversing the previous case-law is a matter of fundamental concern . Secondly, the question has implications going beyond the doctrine of common origin, which are liable to affect far wider classes of trade mark . For understandable reasons, some of those fundamental issues are not fully addressed by the parties . Having been the successful party in HAG I, HAG Bremen does not wish to see the judgment overruled; it merely wishes to curtail its scope . HAG Belgium, on the other hand, contends that the previous judgment should be confirmed and the doctrine applied in the present case . As a result, neither party addresses the fundamental issues that will determine whether the principle laid down in HAG I is valid . Nor have the issues been fully explored by the other participants in the proceedings . Before attempting to resolve those issues, I shall first examine the relevant Treaty provisions and then consider the main principles that can be deduced from the Court' s case-law in the field of intellectual property .

III - The relevant Treaty provisions and Community legislation

8 Article 30 of the Treaty provides that :

"Quantitative restrictions on imports and all measures having equivalent effect shall, without prejudice to the following provisions, be prohibited between Member States ."

Article 36 of the Treaty provides in pertinent part that :

"The provisions of Articles 30 to 34 shall not preclude prohibitions or restrictions on imports, exports or goods in transit justified on grounds of ... the protection of industrial and commercial property . Such prohibitions or restrictions shall not, however, constitute a means of arbitrary discrimination or a disguised restriction on trade between Member States ."

Article 222 of the Treaty provides that :

"This Treaty shall in no way prejudice the rules in Member States governing the system of property ownership ."

Finally, Article 85(1 ) of the Treaty - which is not directly relevant to the present case but which still needs to be borne in mind because it contains one of the essential yardsticks for judging whether a particular course of conduct is acceptable under Community law - provides that :

"The following shall be prohibited as incompatible with the common market : all agreements between undertakings, decisions by associations of undertakings and concerted practices which may affect trade between Member States and which have as their object or effect the prevention, restriction or distortion of competition within the common market ."

9 Articles 30 and 36 articulate a conflict between two competing interests . On the one hand, Article 30, together with the succeeding articles, lays down the fundamental principle of the free movement of goods . On the other hand, Article 36 safeguards, amongst other things, intellectual property rights, which, owing to their territorial nature, inevitably create obstacles to the free movement of goods . Article 36 itself goes some of the way towards explaining how that conflict is to be resolved . It is clear from the wording of the article that not all restrictions on trade created by intellectual property rights are excluded from the prohibition laid down by Article 30 . In order to be excluded from the prohibition, a restriction must, in the first place, be "justified" within the meaning of the first sentence of Article 36 . Secondly, it must not constitute a "means of arbitrary discrimination or a disguised restriction on trade between Member States" within the meaning of the second sentence of Article 36 .

10 In keeping with its nature as a traité-cadre, the EEC Treaty does not purport to lay down an exhaustive code of rules governing the status of intellectual property rights in Community law . It merely provides a skeleton . The task of putting flesh on the bones falls to the Community legislature and to the Court of Justice . In the field of trade mark law the legislature has not been as active as it might have been, having undertaken only two major initiatives, one of which - the proposed Council Regulation on the Comunity trade mark ( Official Journal 1984 C 230, p . 1 ) - has not yet come to fruition . The one measure that has so far been adopted is Council Directive 89/104/EEC of 21 December 1988 approximating the legislation of Member States on trade marks ( Official Journal 1989 L 40, p . 1 ), hereafter "the trade mark directive ". The relevance of that directive to the present case is a matter that I shall deal with later .

IV - The principles established by the Court' s case-law

11 In view of the modest scale of legislative activity in relation to trade marks and to intellectual property in general, the task of reconciling the competing interests enshrined in Articles 30 and 36 of the Treaty has fallen mainly to the Court . It has worked out three fundamental principles, which have played a central part in the entire field of intellectual property, and all of which have their origin in Case 78/70 Deutsche Grammophon v Metro [1971] ECR 487 .

( i ) While the Treaty does not affect the existence of intellectual property rights, there are none the less circumstances in which the exercise of such rights may be restricted by the prohibitions laid down in the Treaty ( see, for example, Deutsche Grammophon, paragraph 11 ).

( ii ) Article 36 permits exceptions to the free movement of goods only to the extent to which such exceptions are necessary for the purpose of safeguarding the rights that constitute the specific subject-matter of the type of intellectual property in question ( Deutsche Grammophon, loc . cit ). Perhaps the main advantage of this formula, apart from the fact that it narrows the scope of the exceptions permitted by Article 36, is that it allows subtle distinctions to be made depending on the type of intellectual property in issue .

( iii ) The exclusive right conferred on the owner of intellectual property is exhausted in relation to the products in question when he puts them into circulation anywhere within the common market . Spelt out more fully, "the proprietor of an industrial or commercial property right protected by the legislation of a Member State may not rely on that legislation in order to oppose the importation of a product which has lawfully been marketed in another Member State by, or with the consent of, the proprietor of the right himself or a person legally or economically dependent on him" ( see, for example, Case 144/81 Keurkoop v Nancy Kean Gifts [1982] ECR 2853, at p . 2873, one of many cases confirming a principle first developed in the Deutsche Grammophon case ).

12 In addition, the Court has developed the principle that the rights conferred under national law by a trade mark ( or presumably by any other form of intellectual property ) cannot be exercised in such a way as to frustrate the competition rules of the Treaty ( Joined Cases 56 and 58/64 Consten and Grundig v Commission [1966] ECR 299, at p . 346; Case 35/83 BAT v Commission [1985] ECR 363, at p . 385 ). The exercise of such rights must not result from agreements or concerted practices that have as their object or effect the isolation or partitioning of the common market, contrary to the terms of Article 85 of the Treaty ( see Case 51/75 EMI Records v CBS United Kingdom [1976] ECR 811 ). In particular, the proprietor of a trade mark may not use it to erect "impenetrable frontiers between the Member States" by assigning the mark to different persons in different Member States ( Case 40/70 Sirena v Eda [1971] ECR 69, at p . 83, paragraph 10 ).

13 It is against that background that there falls to be considered the doctrine of common origin, under which, where similar or identical trade marks that have a common origin are owned by different persons in different Member States, the proprietor of one of the marks cannot rely on it to prevent the importation of goods lawfully marketed under the other mark by its proprietor in another Member State . The doctrine of common origin was laid down by the Court in HAG I and confirmed in Case 119/75 Terrapin v Terranova [1976] ECR 1039 . To a large extent, the outcome of HAG II will depend on whether that doctrine is to be recognized as a legitimate child of Community law .

14 If the four principles enunciated above and the doctrine of common origin are set against the provisions of Articles 30, 36, 85 and 222 of the Treaty, one cannot help being struck by a certain discrepancy . Whereas the four principles may reasonably be deduced from those Treaty provisions, it is much less easy to find therein an obvious basis for the doctrine of common origin . The dichotomy between existence and exercise embodied in the first principle follows from the wording of Articles 36 and, perhaps, 222 . It matters little in this context whether, as is sometimes suggested, Article 222 provides a property guarantee akin to those found in many national constitutions, with the result that Community law cannot threaten the existence of intellectual property rights . That is in any case confirmed by Article 36, which expressly safeguards such rights, and it is unnecessary, in my view, to consider Article 222 independently . But it is equally clear from the limited nature of the derogation accorded by Article 36 that there are circumstances in which the prohibition laid down by Article 30 will none the less apply to the exercise of the right . The concept of specific subject-matter embodied in the second principle is an essential concomitant of the existence/exercise dichotomy, because it makes it possible to determine, in relation to each type of intellectual property, the circumstances in which the exercise of the right will be permissible under Community law . The third principle, namely the principle of the exhaustion of rights, is also firmly anchored in Articles 30 and 36 . Without it traders could, as the Court observed in Deutsche Grammophon, isolate national markets and indulge in practices that would be "repugnant to the essential purpose of the Treaty ". A measure safeguarding such practices would clearly not be "justified" within the meaning of the first sentence of Article 36 . As for the fourth principle mentioned above, it is a straightforward application of Article 85 .

15 What then of the doctrine of common origin? It is much less easy to find a justification for that principle in the Treaty . One can search in vain for a basis in the Treaty for the proposition that the proprietor of a trade mark should not be allowed to prevent the importation of goods produced by the proprietor of a parallel trade mark in another Member State simply because the two marks have a common origin . Without wishing to prejudge the issue at this stage, I must point out that a principle of Community law for which there is no obvious basis in the Treaty is of somewhat dubious pedigree .

V - The nature and function of trade marks

16 Before going any further into the question what, if any, justification can be found for the doctrine of common origin laid down in HAG I, I must first make one preliminary observation about the approach of the Court in the earlier cases to the nature and function of trade marks . With the benefit of hindsight, one can see there were in the previous case-law signs of an unduly negative attitude to the value of trade marks . Thus Advocate General Dutheillet de Lamothe observed in Case 40/70 Sirena [1971] ECR 69, at p . 88 ):

"Both from the economic and from the human point of view the interests protected by patent legislation merit greater respect than those protected by trade marks .

... From the human point of view, the debt which society owes to the 'inventor' of the name 'Prep Good Morning' [a brand of shaving cream] is certainly not of the same nature, to say the least, as that which humanity owes to the discoverer of penicillin ."

The Court echoed those remarks in the judgment ( paragraph 7 ):

"The exercise of a trade mark right is particularly apt to lead to a partitioning of markets, and thus to impair the free movement of goods between States which is essential to the common market . Moreover, a trade mark right is distinguishable in this context from other rights of industrial and commercial property, inasmuch as the interests protected by the latter are usually more important, and merit a higher degree of protection, than the interests protected by an ordinary trade mark ."

17 It is noteworthy that this conception of the relative merits of trade marks and other forms of intellectual property was based on an invidious comparison between a rather trivial trade mark and one of the most important discoveries in the history of medicine . Different comparisons might have produced different results, more favourable to trade marks . The truth is that, at least in economic terms, and perhaps also "from the human point of view", trade marks are no less important, and no less deserving of protection, than any other form of intellectual property . They are, in the words of one author, "nothing more nor less than the fundament of most market-place competition" ( W . R . Cornish, Intellectual property : patents, copyright, trade marks and allied rights, 2nd edition, 1989, p . 393 ).

18 Like patents, trade marks find their justification in a harmonious dovetailing between public and private interests . Whereas patents reward the creativity of the inventor and thus stimulate scientific progress, trade marks reward the manufacturer who consistently produces high-quality goods and they thus stimulate economic progress . Without trade mark protection there would be little incentive for manufacturers to develop new products or to maintain the quality of existing ones . Trade marks are able to achieve that effect because they act as a guarantee, to the consumer, that all goods bearing a particular mark have been produced by, or under the control of, the same manufacturer and are therefore likely to be of similar quality . The guarantee of quality offered by a trade mark is not of course absolute, for the manufacturer is at liberty to vary the quality; however, he does so at his own risk and he - not his competitors - will suffer the consequences if he allows the quality to decline . Thus, although trade marks do not provide any form of legal guarantee of quality - the absence of which may have misled some to underestimate their significance - they do in economic terms provide such a guarantee, which is acted upon daily by consumers .

19 A trade mark can only fulfil that role if it is exclusive . Once the proprietor is forced to share the mark with a competitor, he loses control over the goodwill associated with the mark . The reputation of his own goods will be harmed if the competitor sells inferior goods . From the consumer' s point of view, equally undesirable consequences will ensue, because the clarity of the signal transmitted by the trade mark will be impaired . The consumer will be confused and misled .

20 I should add that the Court, shortly after HAG I, modified its attitude and recognized the twin functions of trade marks as defined above - namely to protect the proprietor' s goodwill and to save the consumer from confusion and deception : see Case 16/74 Centrafarm v Winthrop [1974] ECR 1183 . I will turn in due course to the later case-law . However, the earlier, more negative, approach to trade marks may well help to explain the decision in HAG I itself .

VI - The failure to justify the doctrine of common origin in HAG I

21 Since there is so little authority in the Treaty for the doctrine of common origin and nothing in the previous case-law to suggest the existence of such a doctrine, one might perhaps have expected to find in the Court' s judgment in HAG I a detailed, convincing statement of the reasons that led it to give birth to this new principle of Community law . But that is not the case . The reasoning is condensed into 10 short paragraphs ( paragraphs 6 to 15 ):

"6 . as a result of the provisions in the Treaty relating to the free movement of goods and in particular of Article 30, quantitative restrictions on imports and all measures having equivalent effect are prohibited between Member States;

7 . by Article 36 these provisions shall nevertheless not preclude prohibitions or restrictions on imports justified on grounds of the protection of industrial or commercial property;

8 . nevertheless, it is shown by this very article, in particular its second sentence, as well as by the context, that whilst the Treaty does not affect the existence of rights recognized by the legislation of a Member State in matters of industrial and commercial property, yet the exercise of these rights may nevertheless, depending on the circumstances, be affected by the prohibitions in the Treaty;

9 . inasmuch as it provides an exception to one of the fundamental principles of the common market, Article 36 in fact only admits derogations from the free movement of goods to the extent that such derogations are justified for the purpose of safeguarding rights that constitute the specific subject-matter of this property;

10 . thus the application of the legislation relating to the protection of trade marks at any rate protects the legitimate holder of the trade mark against infringement on the part of persons who lack any legal title;

11 . the exercise of a trade mark right tends to contribute to the partitioning off of the markets and thus to affect the free movement of goods between Member States, all the more so since - unlike other rights of industrial and commercial property - it is not subject to limitations in point of time;

12 . accordingly, one cannot allow the holder of a trade mark to rely upon the exclusiveness of a trade mark right - which may be the consequence of the territorial limitation of national legislations - with a view to prohibiting the marketing in a Member State of goods legally produced in another Member State under an identical trade mark having the same origin;

13 . such a prohibition, which would legitimize the isolation of national markets, would collide with one of the essential objects of the Treaty, which is to unite national markets in a single market;

14 . whilst in such a market the indication of origin of a product covered by a trade mark is useful, information to consumers on this point may be ensured by means other than such as would affect the free movement of goods;

15 . accordingly, to prohibit the marketing in a Member State of a product legally bearing a trade mark in another Member State, for the sole reason that an identical trade mark having the same origin exists in the first State, is incompatible with the provisions providing for free movement of goods within the common market ."

22 I must confess that I do not find this reasoning at all convincing . It is, with respect, flawed in a number of ways .

( i ) First, the emphasis is placed at the outset on the second sentence of Article 36, which was not relevant, since it could not seriously be argued that Van Zuylen' s use of its trade mark amounted to "a means of arbitrary discrimination or a disguised restriction on trade between Member States ".

( ii ) Second, the reasoning is deficient because, having stated in paragraph 9 that Article 36 only admits derogations that are justified for the purpose of safeguarding the rights that constitute the specific subject-matter of the type of intellectual property in question, it fails to define the specific subject-matter of a trade mark . In fact, it was not until several months later, in its judgment in Case 16/74 Centrafarm v Winthrop, already cited, that the Court first defined the specific subject-matter of trade mark rights .

( iii ) Third, the statement in paragraph 11 that "the exercise of a trade mark right tends to contribute to the partitioning off of the markets" does not assist the argument because the statement applies equally to any intellectual property right limited to the territory of a Member State . As for the point that trade mark rights are not subject to limitations in point of time, it is true that in that respect those rights are potentially more permanent in their effects . Against that, however, must be set the fact that in another respect trade marks are less detrimental to the free movement of goods and competition than certain other forms of intellectual property, such as patents, copyright and industrial designs . Whereas the latter entitle the proprietor of the right to exclude a competitor' s goods from the market altogether, a trade mark merely entitles its owner to exclude goods bearing that mark; a competitor has unrestricted access to the market, provided he uses a different mark .

( iv ) Fourth, the reasoning is defective because it states in paragraph 12 a conclusion that simply does not follow from the premises . What was presumably meant by the previous paragraph was that the divided ownership of a trade mark - that is to say, its belonging to different persons in different Member States - tends to partition off the markets . But that nefarious consequence ensues in any case from divided ownership ( or indeed from the coexistence of separate but similar marks ), regardless of whether the trade marks had a common origin . Why then did the Court attach such importance to that element? In fact, the principal defect of the judgment in HAG I is that the Court nowhere explained why the mere fact that the trade marks were of common origin should be relevant, in the absence of any market-sharing agreement . It should be noted in this regard that the Court held in paragraph 5 of the judgment that Article 85 did not apply, since there was "no legal, financial, technical or economic link" between the undertakings .

VII - The attempt to justify the doctrine of common origin in Terrapin v Terranova

23 It was not until two years later, in Case 119/75 Terrapin v Terranova, already cited, that the Court, provoked perhaps by the chorus of criticism that greeted its judgment in HAG I, attempted to explain ex post facto why it attached such importance to the common origin of the trade marks . That case concerned a German and a British trade mark, of independent origin, which the German courts found to be confusingly similar . The Court stated, in paragraph 6 of the judgment, that :

"... the proprietor of an industrial or commercial property right protected by the law of a Member State cannot rely on that law to prevent the importation of a product which has lawfully been marketed in another Member State by the proprietor himself or with his consent . It is the same when the right relied on is the result of the subdivision, either by voluntary act or as a result of public constraint, of a trade mark right which originally belonged to one and the same proprietor . In these cases the basic function of the trade mark to guarantee to consumers that the product has the same origin is already undermined by the subdivision of the original right ".

24 That is a valiant attempt to legitimize the doctrine of common origin, but the logic on which it is based is, I think, fallacious . It is true that the essential function of a trade mark is "to guarantee to consumers that the product has the same origin ". But the word "origin" in this context does not refer to the historical origin of the trade mark; it refers to the commercial origin of the goods . The consumer is not, I think, interested in the genealogy of trade marks; he is interested in knowing who made the goods that he purchases . The function of a trade mark is to signify to the consumer that all goods sold under that mark have been produced by, or under the control of, the same person and will, in all probability, be of uniform quality . That basic function of the HAG mark has never been undermined in Germany, where it has, since its inception, been in the hands of one company . Nor had it been undermined in Belgium and Luxembourg until the Court' s judgment in HAG I . Admittedly, the mark underwent a change of ownership in 1944 in Belgium and Luxembourg . That may or may not have led to a change in quality, which may or may not have been detected by consumers in Belgium and Luxembourg . But that is of no consequence, for the owner of a mark is in any event at liberty to modify the quality of his goods . What matters is that throughout its history ( until 1974, that is ) the mark had, in each territory, been in the exclusive ownership of a single person who had the power either to build up the goodwill associated with it by maintaining the quality of the product or to destroy that goodwill by allowing the quality to deteriorate . Once the owner of the mark is deprived of his exclusive right to its use, he loses the power to influence the goodwill associated with it and he loses the incentive to produce high-quality goods . Looking at matters from the consumer' s point of view, the result of all this is thoroughly unsatisfactory because the trade mark no longer acts as a guarantee of origin . At best he is confused; at worst he is misled . In the circumstances, it is difficult not to conclude that the essential function of the mark is compromised, its specific subject-matter is affected and - most seriously of all - its very existence is jeopardized . But none of those consequences ensued from the fragmentation of the HAG trade mark in 1944; they ensued from the Court' s judgment in HAG I .

25 It might be objected that the above analysis postulates the continued existence of separate markets delimited according to national frontiers and that the question whether a trade mark continues to perform its function as a guarantee of origin must be examined, not with reference to the situation existing in separate national markets, but from a Community-wide viewpoint . One author points out that millions of German tourists spend their holidays in Belgium and that many Belgians travel to Germany ( H . Johannes, "Zum Kaffee-Hag-Urteil des Gerichtshofes der Europaeischen Gemeinschaften", GRUR Int . 1975, p . 111 ). If only Belgian HAG is sold in Belgium and German HAG in Germany, will not these transnational consumers be confused and misled as to the origin of the goods? Superficially, that is an attractive argument . It cannot, however, salvage the doctrine of common origin for two reasons :

( i ) first, the transnational consumer will in any event be confused and misled, even if both types of HAG are available in all the countries concerned . There is in fact no way in which such people can be spared confusion so long as the mark continues in divided ownership in the various countries that they visit ( unless of course one accepts that the products can be differentiated by the use of additional distinguishing matter, a subject that I shall deal with shortly ). I can see no merit in the proposition that, because a minority of transnational consumers are confused and misled as to the origin of certain goods, we must, as a matter of Community law, require the domestic consumers of the entire Community to be similarly confused and misled .

( ii ) secondly, the confusion suffered by transnational consumers in such cases does not depend on whether the two marks have a common origin . The German consumer who goes to Belgium and buys Café HAG believing it to be of the same commercial origin as the coffee that he uses at home is misled in exactly the same degree as the German consumer who goes to the United Kingdom and associates Terrapin' s products with the Terranova products that he is familiar with at home . The fact that in one case the two marks are of common origin, while in the other case they are of independent origin, is irrelevant .

VIII - The conclusion that there is no rational basis for the doctrine of common origin

26 The unpalatable but inescapable conclusion that emerges from the above analysis is that the doctrine of common origin is not a legitimate creature of Community law . There is no clear basis for it in the Treaty and no explanation of its necessity was put forward in HAG I . The attempt to legitimize it ex post facto in Terrapin v Terranova failed for the reasons that I have explained . While the problems caused by the divided ownership of identical or confusingly similar trade marks should not be underestimated, there is, so far as I can see, no rational basis for making the solution to such problems depend on whether the marks have a common origin . Moreover, any fears that the abandonment of the doctrine would open the way for attempts to partition the market by assigning trade marks to different persons in different Member States are illusory . Such attempts could always be defeated either by recourse to Article 85 or by application of the principle of the exhaustion of rights . In fact, the four principles that I described in paragraphs 11 and 12 above ( namely, ( i ) the existence/exercise dichotomy, ( ii ) the confinement of protection to the specific subject-matter of the right in question, ( iii ) the principle of the exhaustion of rights and ( iv ) the applicability of Article 85 to market-sharing assignments of trade marks ) constitute a complete system that allows the requirements of a unified market to be balanced against the interests of the owners of intellectual property and those of the consumer . There was no lacuna needing to be filled by the doctrine of common origin .

IX - The difficulty of reconciling the doctrine of common origin with subsequent developments in the case-law

27 As regards subsequent developments in the case-law, it must be remembered that HAG I was decided at a time when the Court' s case-law on intellectual property was in its infancy . There had been only a handful of cases in that field and some of the basic principles had not been fully worked out . Most of the cases had been dealt with on the basis of the rules on competition, the exception being the Deutsche Grammophon case, which could be disposed of relatively easily on the basis of the exhaustion principle . It is unfortunate that the Court had to resolve such a difficult case as HAG I at a time when it had had little chance to define the relationship between the free movement of goods and the protection of intellectual property rights . That factor alone weakens the value of HAG I as a precedent .

28 There have been two main developments since HAG I . First, there was the articulation of the specific subject-matter of the trade mark right . Secondly, there was the further refinement of the exhaustion principle first laid down in Deutsche Grammophon and the fundamental importance attached to consent on the part of the owner of intellectual property rights .

29 As regards the first development, I have already mentioned that the Court recognized the twin functions of trade marks : protecting the proprietor' s goodwill and saving the consumer from confusion and deception . In the terminology adopted by the Court, the former is described as the specific subject-matter of the right and the latter as its essential function . The specific subject-matter was first defined in Case 16/74 Centrafarm v Winthrop [1974] ECR 1183, paragraph 8 . The case-law was further developed in later judgments : see Case 102/77 Hoffmann Laroche v Centrafarm [1978] ECR 1139, at p . 1164; Case 3/78 Centrafarm v American Home Products Corporation [1978] ECR 1823, at p . 1840; Case 1/81 Pfizer v Eurim-Pharm [1981] ECR 2913, at p . 2925 et seq . In the American Home Products case the Court stated ( at paragraphs 11 to 14 ):

"In relation to trade marks, the specific subject-matter is in particular the guarantee to the proprietor of the trade mark that he has the exclusive right to use that trade mark for the purpose of putting a product into circulation for the first time and [is] therefore his protection against competitors wishing to take advantage of the status and reputation of the mark by selling products illegally bearing that trade mark .

In order to establish in exceptional circumstances the precise scope of that exclusive right granted to the proprietor of the mark regard must be had to the essential function of the trade mark, which is to guarantee the identity of the origin of the trade-marked product to the consumer or ultimate user .

This guarantee of origin means that only the proprietor may confer an identity upon the product by affixing the mark .

The guarantee of origin would in fact be jeopardized if it were permissible for a third party to affix the mark to the product, even to an original product ."

30 As regards the exhaustion principle, the Court held in Case 187/80 Merck v Stephar [1981] ECR 2063, that the holder of a national patent in one Member State who markets the patented product in another Member State where it is not patentable cannot rely on his patent in order to prevent parallel imports . The fact that he had not enjoyed the patent-holder' s ordinary privilege of marketing his product under monopoly conditions and thus had to accept a lower profit was considered irrelevant . All that mattered was his consent . The overriding importance of consent was further emphasized, again in relation to patents, in Case 19/84 Pharmon v Hoechst [1985] ECR 2281, in which the Court held that a patent proprietor may prevent the importation of a product which has been manufactured in another Member State under a compulsory licence granted in respect of a parallel patent held by the same proprietor, irrespective of whether he has accepted or refused royalties payable under the compulsory licence .

31 HAG Bremen relies heavily on the Pharmon judgment, arguing that the victim of an expropriation may be equated with the patent holder against whom a compulsory licence is granted . There is much force in that argument . In both cases a coercive act of the public authorities deprives the property owner of the power to determine, of his own volition, what use he wishes to make of his property . I have already indicated that I consider trade marks to be no less worthy of protection than patents . I find it very difficult to understand why Community law should afford less protection to a trade mark holder who is the victim of an expropriation than it does to a patent holder whose patent is the subject of a compulsory licence, especially when the expropriation occurs without compensation and a royalty is paid under the compulsory licence . Moreover, it might be argued that the patent-holder who obtains a parallel patent in a country where provision is made for the granting of compulsory licences does at least accept the risk that such a licence may be granted, whereas it can hardly be said that a trade mark owner who registers his mark in a foreign land accepts the risk that the mark may one day be expropriated .

32 The conclusion which I draw from this survey of subsequent developments in the case-law is that the doctrine of common origin is indeed difficult, if not impossible, to reconcile with those developments .

X - The general problem of trade mark conflicts in Community law

33 I have expressed the view that the problems caused by the divided ownership of identical or confusingly similar trade marks cannot be resolved on the basis of a distinction as to whether the marks are of common or independent origin . In other words, the situation that arose in HAG I should not be treated differently from the situation that arose in Terrapin v Terranova . In my view, the former case was wrongly decided and the latter case ( subject to certain reservations ) was correctly decided . There are, however, proponents of the opposite view . Even before Terrapin v Terranova was decided there were authors who expressed the hope that the integrationist approach adopted in HAG I would be extended to cases in which the marks were not of common origin ( see, for example, H . Johannes, "Anwendung der Prinzipien des Kaffee-Hag-Urteils auf nichtursprungsgleiche Warenzeichen und Freizeichen", RIW/AWD 1976 p . 10 et seq .; and Roettger, "Kollision von identischen oder verwechslungsfaehigen Warenzeichen und Firmennamen innerhalb der Europaeischen Gemeinschaft", RIW/AWD 1976 p . 354 et seq .).

34 That view does at least have the virtue of emphasising what is at stake . Although there may not be many cases of identical trade marks being owned by different persons in different Member States ( as in HAG ), there are very many cases in which a trade mark protected in one Member State is found to be confusingly similar to a trade mark owned by someone else in another Member State ( as in Terrapin v Terranova ). Trade mark law does not generally distinguish between these two types of case; the owner of the trade mark may rely on it to prevent other persons from purveying goods under an identical or confusingly similar mark . That much is confirmed by Article 5(1 ) of the trade mark directive, which provides as follows :

"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 ."

35 Consequently, if the Court' s ruling in HAG II follows the approach adopted in Terrapin v Terranova, it will affect not only the limited number of cases in which identical marks clash but also the far more numerous cases in which similar marks are held to be confusing . It has been suggested that the number of confusingly similar marks within the Community amounts to several hundred thousand ( apparently this figure was put forward by the German Government in its observations in Terrapin v Terranova : F . K . Beier, "Trade mark conflicts in the common market : Can they be solved by means of distinguishing additions?", IIC 1978, p . 221 ). Even if that estimate is too high, it is obvious that trade mark conflicts can constitute a considerable hindrance to intra-Community trade .

36 To make matters worse, the concept of confusingly similar marks must inevitably vary from one Member State to another and this can lead to a certain lack of reciprocity and to distortions of trade . For example, in Terrapin v Terranova the German courts found that there was a risk of confusion between the two marks . It is questionable whether an English court would take the same view . This could have the unfortunate consequence that the British manufacturer would be prevented from trading in Germany under his usual mark, while the German manufacturer would have unrestricted access to the British market . In fact, it appears that the German courts take a particularly broad view of the concept of confusingly similar marks, as is well illustrated by the facts of the proceeding before the Commission reported as Tanabe Seiyaku Company v Bayer AG CMLR [1979], 2, p . 80 . In one notorious case ( BPatG, 28.3.1973, GRUR 1975, p . 74 ) the Bundespatentgericht held that the mark "LUCKY WHIP" was liable to be confused with the mark "Schoeller-Nucki", a decision that seems to postulate a body of consumers afflicted with an acute form of dyslexia . It is against the background of that kind of national case-law that the Court must consider whether to confirm and extend the approach adopted in Terrapin v Terranova .

XI - The arguments in favour of the coexistence of conflicting marks : examples from national law

37 Those who defend the approach adopted in HAG I and maintain that it should be extended to cases in which the marks are not of common origin argue that identical or confusingly similar trade marks can coexist on the same market if they are distinguished by means of additional markings . They also cite examples of coexistence taken from national law, such as the doctrine of honest concurrent user in English law or the German Law of 1959 concerning the Integration of the Saarland in connection with Industrial Property ( Gesetz ueber die Eingliederung des Saarlandes auf dem Gebiete des gewerblichen Rechtsschutzes, BGBl . 1959 I, p . 388 ). These examples taken from national law are worthy of attention and I shall consider them before going on to examine the question whether it is possible to distinguish identical or similar marks by means of additional markings .

38 The common-law doctrine of honest concurrent user was developed in the nineteenth century . Trade had hitherto been largely local and it sometimes happened, quite by chance, that identical or similar trade marks were adopted by two or more traders in different parts of the country . There was no risk of confusion because the marks, although applied to similar products, were not in use on the same geographical market . However, if two traders with confusingly similar marks expanded beyond their own localities the marks could come into conflict with each other . To meet that situation the English courts developed the doctrine of honest concurrent user, under which each of the traders concerned was entitled to continue using his mark in such circumstances ( see General Electric Co . v The General Electric Co . Ltd All E.R . 1972 2, p . 507, at p . 519, Lord Diplock ). The doctrine is now embodied in Section 12(2 ) of the Trade Marks Act 1938, under which the competent authorities are empowered to permit the registration of identical or confusingly similar marks in case of honest concurrent user, subject to such conditions and limitations as they think fit .

39 The German Law on the Integration of the Saarland provided for the extension to the whole of Germany of trade mark rights previously recognized in the Saarland, and vice versa . In the event of conflicts between identical or confusingly similar trade marks provision was made for one or both of the marks to be supplemented by additional distinguishing matter . Disputes over the type of additional distinguishing matter that was needed were to be resolved by an arbitration board attached to the German Patent Office .

40 Attractive though these precedents from national law may be, I do not think that either of them is suitable for resolving trade mark conflicts in Community law . As regards the doctrine of honest concurrent user, one must be careful not to overestimate its significance . In its modern form it simply amounts to a discretionary power, conferred on the Registrar of Trade Marks and on the competent courts, to permit the registration of identical or confusingly similar marks subject to certain conditions . The conditions imposed frequently involve territorial limitations ( see Kerly' s Law of Trade Marks and Trade Names, 12th Edition 1986, by T . A . Blanco White and R . Jacob, p . 159 ), which might not be considered acceptable in Community law . Moreover, even if concurrent registrations are permitted, each owner of the mark will still be able to succeed in an action for "passing-off" if he can show that in his part of the country the goodwill attaching to the mark belongs to him and that the other owner' s goods would be mistaken for his ( see Cornish, op . cit ., p . 452 ). The result is that if the doctrine of honest concurrent user were applied to the present case HAG Bremen and HAG Belgium might still be able to exclude one another from their respective territories .

41 As regards the German Law on the Integration of the Saarland, it must be borne in mind, in the first place, that the scale of the economic interests involved is hardly comparable . Secondly, much depends on the extent to which it is possible to differentiate between identical or confusingly similar marks by means of additional distinguishing matter . I shall deal with that question in the following paragraphs .

XII - The use of additional distinguishing matter

42 The possibility of differentiating between conflicting trade marks by means of additional markings is alluded to directly in the second question referred by the Bundesgerichtshof . It is, however, of general importance and could affect the answer to the first question referred . If the Court were considering replying to that question in the negative, i.e . to the effect that HAG Bremen cannot oppose the imports in question, then serious consideration would have to be given to this aspect of the case . Two questions arise : First, is it possible to differentiate effectively between conflicting marks by adding further distinguishing matter ( or perhaps by the use of different colours, as in the Persil case mentioned in the pleadings ), so as to dispel the confusion created in the mind of the consumer by identical or similar marks? Secondly, is it in practice possible to do so in a manner that interferes with the free movement of goods less than would the requirement to use a completely different mark? In HAG I the Court assumed ( in paragraph 14 of the judgment ) that both those questions should receive an affirmative answer, but it did not attempt to explain why that should be so . The subject is by no means as simple as the Court seemed to imagine and has generated a considerable amount of literature ( see, for example, F . K . Beier, "Trade mark conflicts in the common market : Can they be solved by means of distinguishing additions?", IIC 1978, p . 221 ).

43 With regard to the first of the two questions that I have formulated, everything must of course depend on the facts . Where the conflicting marks are identical, as in the present case, the initial impression conveyed to the consumer that the goods have the same commercial origin is so strong that I doubt whether it could be dispelled by any amount of additional markings or by the use of different colours . I question whether any consumer seeing blue and green packets of Persil side by side on a supermarket shelf would think for one moment that they were not produced by, or under the control of, the same firm . As regards marks that are confusingly similar but not identical, the problem may not be quite so insuperable . For instance, it would, I think, be possible to overcome any confusion between Terrapin and Terranova' s products by means of additional printed matter, such as a statement to the effect that there is no connection between the two firms . On the other hand, any consumer who is so inattentive as to entertain confusion between "LUCKY WHIP" and "Schoeller-Nucki" is not likely to be enlightened by any amount of additional information .

44 As to the second of the questions that I formulated above, it is plain that reliance on a trade mark does not constitute an absolute barrier to imports; one suggestion, made at the hearing by counsel for the United Kingdom, was that the owner of a mark that conflicts with a mark owned by someone else in another Member State is merely required to obliterate that mark with a sticker bearing a different mark . It would, it is argued, be illogical to maintain that such a requirement is contrary to the provisions on the free movement of goods but that a requirement for him to apply a sticker to his wares disclaiming any connection with those of the other trader would be compatible with those rules, since both measures would be equally burdensome and would interfere with the free movement of goods to the same degree . There is much force in that argument . It is not, however, quite as devastating as it may appear . There is in fact a considerable difference between a sticker, placed adjacent to the trade mark, disclaiming any connection with another trader' s products, and a sticker that obliterates the trade mark and replaces it with a different one . Whereas the former might be entirely acceptable to consumers, if only for its frankness, the latter would, I think, tend to excite suspicion that there was something wrong with the goods . Moreover, the consumer might feel deceived if he were to remove the sticker after purchasing the goods and discover that the goods appeared to have a different origin from that which he had assumed . Rather than risk damaging his goodwill in this way, the manufacturer might prefer to put up the goods in completely different packaging, which would be more costly than the simple expedient of adding a sticker .

45 The conclusion that follows from the above considerations is that there are circumstances in which it might be practical to distinguish between conflicting trade marks by means of additional markings, but that such circumstances constitute the exception rather than the rule . I doubt whether that method would ever be effective in the case of identical trade marks used for identical products . Above all, it must be stressed that it is not a panacea to all the problems posed by trade mark conflicts, as the Court seemed to imply in HAG I .

XIII - The conclusion : the trade mark owner may rely on his right against the owner of a parallel right in another Member State

46 In view of the above considerations, I am convinced that the owner of a trade mark must be allowed to exclude from his territory goods on which an identical trade mark has been placed by another, unrelated person who is the owner of the mark in another Member State . The same goes for confusingly similar marks, except perhaps in cases where it would be practical to differentiate between them by means of additional distinguishing matter . Such a conclusion is justified from the point of view both of the trade mark owner and of the consumer . From the owner' s point of view, the specific subject-matter of the right - namely, his exclusive right to use the mark in the territory concerned and thus to protect himself against unfair competition - would be affected and the existence of the right would be threatened if he were obliged to tolerate the use of the mark by a competitor . From the consumer' s point of view, the essential function of the trade mark, which is to prevent him from being confused and misled as to the origin of the goods that he purchases, would be undermined . It would be highly undesirable for Community law to promote the coexistence, on the same market, of identical or confusingly similar trade marks .

47 The conclusion reached above confirms my belief that the approach followed in Terrapin v Terranova ( subject to my reservations about the genuineness of the risk of confusion and the possibility of eliminating it by means of additional information ) was correct and that HAG I was wrongly decided . I could not, however, recommend that view unreservedly without first satisfying myself that there is some means of preventing the worst excesses that may ensue from the divergent interpretation given in national law to the concept of confusingly similar marks . But such means are certainly available .

48 In my view, an unduly broad view of the concept of confusingly similar marks - exemplified in an extreme form in the "LUCKY WHIP" decision - would run counter to Article 30 of the Treaty and would not be "justified" under the first sentence of Article 36 . Moreover an excessively wide approach is prohibited by the second sentence of Article 36 . Reliance on a trade mark in order to exclude goods manufactured in another Member State where the risk of confusion between the two marks is minimal would amount, if allowed by national courts, to a disguised restriction on trade between Member States . If the rights conferred by the trade mark were enforced in a discriminatory manner, that would amount to arbitrary discrimination . That was implied by the Court in paragraph 4 of the judgment in Terrapin v Terranova .

49 Furthermore, in that judgment the Court did not rule out the possibility that it might legitimately be called upon to rule on the issue of similarity and the risk of confusion, at least as regards the implications thereof in Community law . That possibility has been strengthened by the trade mark directive because the concept of confusingly similar marks is now a concept of Community law ( see Articles 4(1)(b ) and 5(1)(b ) of the directive ). Although the directive cannot produce direct effect as against an individual ( Case 152/84 Marshall v Southampton and South-West Hampshire Area Health Authority [1986] ECR 723 ), the national courts will, after the expiry of the period for its implementation, be required to interpret national law, in particular national provisions implementing the directive, in the light of the directive' s wording and purpose ( Case 14/83 Von Colson and Kamann v Land Nordrhein-Westfalen [1984] ECR 1891 ). They will be empowered, or obliged, to request preliminary rulings and the Court will, by ensuring a uniform - and perhaps restrictive - interpretation of the concept of confusingly similar trade marks, be able to eliminate the abuses and discrepancies that I have alluded to above .

50 I am therefore satisfied that the Court may safely confirm the general approach adopted in Terrapin v Terranova and extend it to cases involving trade marks that are of common origin .

XIV - The trade mark directive

51 I must deal at some length with the possible implications of the trade mark directive already referred to, because a great deal of weight was put on it in argument . The German Government contends that the present case should be resolved on the basis of its provisions . It recognizes that the directive cannot have direct effect as between individuals and that in any case the period for its implementation has not yet expired, but maintains that account should none the less be taken of the directive, because it contains a definitive statement of the legislature' s opinion as to what limitations of the free movement of goods are justified for the protection of industrial and commercial property . By deliberately omitting to incorporate the doctrine of common origin into the directive, the Council has tacitly indicated that there is no place for such a doctrine in Community law . Moreover, the provisions of German law relied on by HAG Bremen fully accord with the directive and cannot therefore be considered incompatible with Community law .

52 From the Court' s point of view, that might appear an attractive solution because it would avoid the need to admit that HAG I was wrongly decided . It would be possible to say that HAG I was correctly decided but that the legal basis for the decision has been destroyed by subsequent legislation . However, before embracing that solution, the Court must be satisfied that the directive does indeed have the effect for which the German Government contends . I am not convinced that it does .

53 In the first place, one must be wary of reading too much into the silence of the legislature . Silence is by nature ambiguous and can be interpreted either as approbation or condemnation, depending on the interpreter' s subjective point of view . With equal facility one could infer from the failure to mention the doctrine of common origin, in the trade mark directive, either an intention to confirm it or an intention to abrogate it . We would be justified in making the latter assumption only if it were clear that the directive purported to codify the Court' s case-law on the relationship between the free movement of goods and the protection of trade mark rights and that it sought to deal with that subject exhaustively . That is not however the case . It is true that in Article 7 the directive has adopted the Court' s case-law on the exhaustion of rights, including part of the case-law on repackaging . But even that subject has not been dealt with exhaustively by the directive . For proof of that, one need only compare the remarkably vague provisions of Article 7(2 ) with the exceptionally detailed rules laid down by the Court in Case 102/77 Hoffmann-La Roche v Centrafarm [1978] ECR 1139, at p . 1165 et seq . Moreover, there is another important aspect of the Court' s case-law that is completely ignored by the directive . In the American Home Products case the Court held that, if a manufacturer uses different marks in different Member States for the purpose of artificially partitioning the market, he may lose the right to prevent unauthorized use of the marks by third parties, since he would be exercising his right in such a way as to create a disguised restriction on trade . I should be reluctant to infer from the Council' s silence on that point that it intended to confer on trade mark owners the power to create such disguised restrictions on trade . But that would be the logical consequence of the German Government' s argument; for we cannot construe the Council' s silence on that point as implicitly confirming the rule laid down by the Court, and yet draw exactly the opposite conclusion from its silence in relation to the doctrine of common origin .

54 There is moreover the further difficulty whether the Council could, by legislation, abrogate a doctrine purportedly based on the Treaty . The German Government attempts to surmount that difficulty by arguing that the principles worked out by the Court governing the relationship between Articles 30 and 36 cease to be relevant once the substantive law of the Member States has been harmonized . Pending harmonization those principles perform an "Ersatzfunktion"; after harmonization they become redundant because they are replaced by the provisions of the harmonizing directive . National legislation cannot be contrary to Article 30 if it is consistent with the directive and it cannot be saved by Article 36 if it is inconsistent with the directive .

55 There may be some truth in that argument as regards a directive that harmonizes national rules on, for example, the composition of animal feedingstuff, as in the Tedeschi case cited by the German Government ( Case 5/77 [1977] ECR 1555 ), because in such a case the source of the impediment to free movement - namely, the discrepancies in national legislation - is removed by the directive . But the argument breaks down when it is applied to the trade mark directive . Most of the conflicts between intellectual property rights and the free movement of goods, including conflicts caused by the divided ownership of a trade mark, are due not to discrepancies in national law but solely to the territoriality of national law . The directive has done nothing to limit that territoriality and so has done nothing to solve the problems caused by it . Hence, national laws that allow a trade mark owner to oppose imports from another Member State continue to fall foul of Article 30 and must continue to look for their salvation in Article 36 . It is therefore illusory to pretend that the Court' s case-law on Article 36 has been rendered redundant by the directive . Furthermore, it would in any event be erroneous to imagine that all discrepancies in the laws of Member States have been removed by the directive . Indeed the directive ( as its title, "First Directive", indicates ) is merely the first stage in the harmonization of national laws .

56 In parentheses, I would add that the problems will not be solved even after the adoption of the proposed regulation creating a single Community mark . The existing national marks will continue to coexist with the Community mark and, as the Commission recognized at the hearing, where there is a divided mark there will be no possibility of obtaining a Community mark .

57 Returning to the directive, I conclude from the above analysis that the directive is not directly relevant to the present case . Still less can it be contended, on the basis of the directive, that HAG I was correctly decided at the time, but that the decision has since been deprived of its validity by the directive . Tempting though such a solution may appear, it cannot, I think, be pretended that the doctrine of common origin was born a legitimate child of Article 36, but has since been orphaned by an act of the legislature .

XV - The possibility of distinguishing HAG I and HAG II

58 The view that HAG II can be distinguished from HAG I is actively canvassed by HAG Bremen . That is hardly surprising . Having regained access to the Belgian and Luxembourg markets under the HAG trade mark, as a result of the previous litigation, it does not wish to surrender that benefit by calling in question the validity of the decision in HAG I . Instead it seeks to show that that decision should have been based on different grounds that would not have the effect of forcing it to share the HAG trade mark with HAG Belgium in Germany and in the rest of the Community . The principal legal arguments it advances are as follows :

( i ) in the first place, it emphasizes the importance of consent in the Court' s case-law on the exhaustion of rights . Having been compulsorily deprived of the HAG trade mark in Belgium and Luxembourg, it cannot be said to have consented to the subdivision of the mark or to its use by third parties . HAG Belgium, on the other hand, derives its rights to the mark from the Van Oevelen family and from the Belgian Government, and so cannot be in a stronger position legally than they were . They, however, consented to the subdivision of the mark and acquired their rights in it with full knowledge that outside Belgium and Luxembourg it was owned by a third party;

( ii ) secondly, it contends that to allow HAG Belgium to use the HAG mark in Germany would be tantamount to giving extraterritorial effect to the expropriation that took place in 1944, thus violating an established principle of international law;

( iii ) thirdly, it maintains that, although the trade mark' s essential function of indicating the origin of goods was impaired in Belgium and Luxembourg, as a result of the expropriation, it has never been impaired in Germany, where it has throughout remained in the same ownership .

59 In so far as those arguments tend to suggest that the doctrine of common origin should be modified so as to apply in HAG I but not in HAG II, because of material differences in the facts of the two cases, they need not be considered further, since on the view I take the doctrine as applied in HAG I will be abandoned . That is true, in particular, of the third argument mentioned above . That argument can certainly be supported on the basis of the doctrine of common origin as explained and modified in Terrapin v Terranova but, as I have sought to show above, even that modified doctrine is not, on analysis, defensible .

60 The arguments can also be understood as leading to a different conclusion, namely that, if the doctrine of common origin is abandoned, HAG Bremen would still be entitled to use its mark in Belgium and Luxembourg on the basis of some other legal principle . That may well be so, but it would not be appropriate to express a view on that question in these proceedings, where it does not arise : the fate of HAG Bremen in Belgium and Luxembourg is plainly not a matter to be resolved by the German courts in these proceedings . I must nevertheless consider the arguments, in case they should be regarded as relevant .

61 If the argument relating to consent were used in order to justify the decision in HAG I, it would involve changing the entire basis of the decision . It would amount to saying that the decision in HAG I should not have been based on the spurious doctrine of common origin; instead it should have been based either on the principle laid down in relation to Article 85 in Case 40/70 Sirena v Eda, already cited, or on a rather unusual application of the exhaustion-of-rights principle . Both of those solutions are fraught with difficulty .

62 The Sirena case resembled HAG inasmuch as it concerned a trade mark that was subdivided long before the entry into force of the EEC Treaty . In that case, however, the subdivision was effected by a contractual assignment rather than by a coercive act of the public authorities . The case was dealt with purely on the basis of Article 85 . The Court held that the exercise of a trade mark right "might fall within the ambit of the prohibitions contained in the Treaty each time it manifests itself as the subject, the means or the result of a restrictive practice" and that "if the restrictive practices arose before the Treaty entered into force, it is both necessary and sufficient that they continue to produce their effects after that date ". In HAG I the Court held that Article 85 was not applicable in the absence of any link - legal, financial, technical or economic - between the holders of the mark . It might, however, have been possible to argue that, although HAG Bremen did not surrender the mark by agreement, Van Oevelen and their successors in title acquired the mark in that way . It would not be unreasonable to say that those parties should be treated in the same way as a contractual assignee of the mark and that as such they might be caught by the rule laid down in the Sirena case . The problem is that that rule was modified, in one important respect, in Case 51/75 EMI Records v CBS United Kingdom, already cited . That case also concerned a trade mark that was subdivided, by means of contractual assignments, long before the Treaty came into force . The Court ruled that :

"An agreement is only regarded as continuing to produce its effects if from the behaviour of the persons concerned there may be inferred the existence of elements of concerted practice and of coordination peculiar to the agreement and producing the same result as that envisaged by the agreement .

This is not so when the said effects do not exceed those flowing from the mere exercise of the national trade mark rights ."

If we apply that test to the facts of HAG I, it is still difficult to avoid the conclusion that that case was wrongly decided . There is manifestly no concerted practice between HAG Belgium and HAG Bremen and the effects of the putative "assignment" do not exceed those flowing from the mere exercise of the national trade mark rights .

63 The alternative solution, consisting in the application of the exhaustion-of-rights principle, is equally problematical . In the first place, it would involve a slightly unusual application of that principle . If HAG Bremen had voluntarily assigned the Belgian and Luxembourg marks to Van Oevelen, it would be easy to say that HAG Bremen had consented to the use of the mark by Van Oevelen in another Member State and had therefore exhausted its rights . HAG Bremen would not therefore be able to rely on its German trade mark in order to prevent imports of Van Oevelen' s products into Germany . But would the same principle apply in reverse? Logically it should, even though the assignee of the subdivided mark could hardly be said to have exhausted his right; it would be more accurate to say that he acquired a right that was already exhausted . But once again the matter is complicated by the fact that the subdivision of the mark took place in 1944, 14 years before the entry into force of the Treaty . The essence of the exhaustion theory is that the owner of an intellectual property right in a Member State exhausts that right throughout the Community by consenting to the marketing of the product in question in another Member State . Can the owner of an intellectual property right be said to have exhausted his right throughout the Community by an act that he committed long before the Community came into being? That is an important question of principle which has not been resolved by the case-law of the Court . Such an important question should not, in my view, be resolved by means of an obiter dictum . For that reason I do not think that it would be wise to attempt to uphold the decision in HAG I on the ground that Van Oevelen and their successors consented to the subdivision of the right .

64 As regards the argument that, by allowing HAG Belgium to use the HAG mark in Germany, the Court would be giving extraterritorial effect to the expropriation that took place in 1944, I cannot see that it adds very much to HAG Bremen' s already overwhelming case or that it justifies distinguishing between HAG I and HAG II . I say so because Community law cannot in any case have the effect of expropriating someone without compensation, in particular by depriving him of his intellectual property rights . For the reasons I have given, I believe that the decision in HAG I did affect the existence of Van Zuylen' s rights and that a similar decision in HAG II would affect the existence of HAG Bremen' s rights . The essential point is that the goodwill associated with the HAG mark is HAG Bremen' s property in Germany, whereas in Belgium and Luxembourg it is HAG Belgium' s property . The goodwill in Belgium and Luxembourg ceased to be HAG Bremen' s property when it was expropriated in 1944 . It is not the business of Community law to undo the effect of that expropriation in Belgium and Luxembourg any more than it is the business of Community law to extend the territorial effect of the expropriation into Germany .

65 As regards the argument based on the trade mark' s essential function of indicating the origin of goods, I have already sought to demonstrate that, once that function is properly understood, it should be clear that it was not impaired, before 1974, in Belgium and Luxembourg any more than in Germany . Hence, that argument cannot justify distinguishing between HAG I and HAG II .

66 I conclude that, while there may be a case for distinguishing between HAG I and HAG II, the arguments for so doing are not convincing . The real difference between the two cases is perhaps that in HAG II the injustice capable of being wrought by the doctrine of common origin is more obvious than it was in HAG I . But it is only a question of degree . It would, I think, be healthier to recognize that HAG I was wrongly decided, rather than to compound that error by inventing a spurious distinction between the two cases .

XVI - The issue of reversing the previous case-law

67 If, as I consider it must be, Question 1 is answered in the affirmative, then the Court should in my view make it clear, in the interests of legal certainty, that it is abandoning the doctrine of common origin laid down in HAG I . The Court has consistently recognized its power to depart from previous decisions, as for example by making it clear that national courts may refer again questions on which the Court has already ruled : see Joined Cases 28, 29 and 30/62 Da Costa en Schaake [1963] ECR 31 where the Court accepted that a "materially identical question" could be referred again, and Case 283/81 Cilfit v Ministry of Health [1982] ECR 3415, paragraph 15; see also Case 28/67 Molkerei-Zentrale [1968] ECR 143, at pp . 152 to 155 where the Court expressly reconsidered a previous ruling . That the Court should in an appropriate case expressly overrule an earlier decision is I think an inescapable duty, even if the Court has never before expressly done so . In the present case the arguments for expressly abandoning the doctrine of common origin are exceptionally strong; moreover, the validity of that doctrine is already, as I have suggested, in doubt as a result of the intervening case-law . To answer Question 1 in the affirmative without abandoning the doctrine, or to seek to rationalize such an answer on some other ground, would be a recipe for confusion .

XVII - The remaining questions

68 Questions 2, 3 and 4 only arise if Question 1 is answered in the negative . Since in my view Question 1 must be answered in the affirmative, it is not therefore necessary to consider the remaining questions . In case, however, they should be thought relevant, I shall briefly comment on them .

69 Question 2 asks in substance whether the answer to Question 1 would be different if the mark relied on were so well known that, in the event of its being used by more than one undertaking in the same territory, it would be impossible to inform the consumer as to the commercial origin of the goods without adverse repercussions on the free movement of goods . Question 3 asks whether the answer would still be the same even though consumers associate the mark relied on not only with a certain commercial origin but also with certain perceptions as to the quality of the goods . For the sake of completeness, Question 4 asks whether the conditions set out in Questions 2 and 3 might together change the answer to Question 1, even though neither of them would by itself have such an effect .

70 The two additional factors referred to in Questions 2 and 3 are of course important - so important, in fact, that I have been unable to deal with Question 1 without already touching upon them . I need add very little to what I have already said .

71 As regards Question 2, I do not think that it would be wise to create two categories of trade mark - well-known ones and little-known ones . Admittedly, the trade mark directive recognizes that marks enjoying a certain reputation are entitled to a higher standard of protection in some respects . However, I cannot see that the provisions in question ( Articles 4(3 ) and ( 4 ) and 5(2 ) ) have any bearing on the present case . As regards the task of informing the consumer about the commercial origin of the marked goods, I have already expressed the view that, in the case of identical marks for identical products, the risk of confusion is so great that no amount of additional distinguishing matter could prevent it . That must be true not only of marks that are household names but also of marks that are relatively obscure; for any mark may be well known to the limited circle of consumers, whether great or small, who purchase or contemplate purchasing the product in question . The confusion suffered by consumers familiar with a mark does not increase or decrease in proportion to their number .

72 As regards the tendency of a particular trade mark to convey to consumers certain perceptions as to the quality of the marked goods, I have already pointed out that that tendency is bound up with the essential function of trade marks in general . It is sometimes said that the essential function of the trade mark is to act as a guarantee of origin but not as a guarantee of quality . That is true in the limited sense that the manufacturer is not under an obligation to ensure that all goods sold under a particular mark are of the same quality . But, as I have suggested, the relevance of the trade mark' s function as a guarantee of origin lies none the less in the fact that the trade mark conveys to the consumer certain perceptions as to the quality of the marked goods . The consumer is not interested in the commercial origin of goods out of idle curiosity; his interest is based on the assumption that goods of the same origin will be of the same quality . That is how trade mark protection achieves its fundamental justification of rewarding the manufacturer who consistently produces high-quality goods . To answer Question 1 in the negative would be to ignore this aspect of the trade mark' s essential function . It is not therefore possible to answer Question 1 in the negative but then say that the answer might be different if the factor referred to in Question 3 were present .

XVIII - The reply to the questions referred

73 As is often the case in proceedings under Article 177, one of the most difficult tasks is the drafting of the reply to the questions referred by the national court . One of the criticisms that might be levelled against the ruling in HAG I is that it was considerably wider than necessary . Mindful of the need to avoid a repetition of that mistake, I shall propose a reply to the national court' s first question that will enable it to give judgment in the case before it, but will not prejudice the issue of HAG Belgium' s rights against HAG Bremen in Belgium and Luxembourg . That issue raises some difficult questions of law that have not been fully debated in the present proceedings . I therefore propose that the questions referred by the Bundesgerichtshof should be answered as follows :

"Articles 30 to 36 of the EEC Treaty do not prevent an undertaking from relying on a trade mark of which it is the proprietor in a Member State in order to oppose imports from another Member State of similar goods bearing an identical or confusingly similar trade mark which was originally owned by the same undertaking but was subsequently acquired by an entirely unrelated undertaking without the consent of the first undertaking ."

(*) Original language : English .