OPINION OF MR ADVOCATE-GENERAL WARNER

My Lords,

On 14 July 1971 the Federal German Parliament enacted a new statute about wine, the Weingesetz of that date. Among the implementing regulations made under that statute the following day was a set of regulations concerning sparkling wine and brandy, the Schaumwein-Branntwein-Verordnung of 15 July 1971. I will, for convenience, refer to that statute and to those regulations, together, as ‘the 1971 legislation’.

So far as relevant to this case, that legislation provides:

(1)that, in the Federal Republic of Germany, sparkling wine, whether produced in Germany or elsewhere, is in general to be described as ‘Schaumwein’, but that, if it complies with certain prescribed standards of quality, it may be described as ‘Qualitatsschaumwein’;

(2)that Qualitätsschaumwein may be described as ‘Sekt’ if produced in a country on the whole of the territory of which German is an official language;

(3)that Sekt may be described as ‘Pradikatssekt’ (choice-sekt) if made from home-grown grapes to the extent of at least 60 %;

(4)that brandy, whether produced in Germany or elsewhere, is in general to be described as ‘Branntwein aus Wein’ (wine-spirit), but that, if it complies with certain prescribed standards, it may be described as ‘Qualitatsbranntwein aus Wein’; and

(5)that Qualitatsbranntwein aus Wein may be described as ‘Weinbrand’ if produced in a country on the whole of the territory of which German is an official language.

The question at issue in these proceedings, which are brought by the Commission against the Federal Republic under Article 169 of the EEC Treaty, is whether the Federal Republic is in breach of its obligations under that Treaty in thus reserving the descriptions ‘Sekt’ and ‘Weinbrand’ for products of countries where German is an official language, and the description ‘Pradikatssekt’ for a produkt made as to 60 % from home-grown grapes.

In essence the Commission's case is that ‘Sekt’ and ‘Weinbrand’ are, in German, generic terms, meaning respectively ‘sparkling wine’ and ‘brandy’; that both have more consumer appeal than the expressions ‘Qualitätsschaumwein’ and ‘Qualitätsbranntwein aus Wein’, which were newly coined by the 1971 legislation; and that accordingly to reserve the names ‘Sekt’ and ‘Weinbrand’ for the products of German-speaking countries places the products of other countries at a disadvantage. The 1971 legislation thus inhibits the sale in Germany of imported sparkling wines and brandies and so constitutes the imposition of measures having an effect equivalent to quantitative restrictions, contrary to Article 30 of the Treaty.

The Commission points in this connexion to its own Directive No 70/50/EEC of 22 December 1969, adopted under Article 33 (7) of the Treaty and addressed to the Member States.

The purpose of that Directive was to secure the abolition, as between Member States, of certain kinds of measures having an effect equivalent to quantitative restrictions on imports which were operative at the date of entry into force of the Treaty, and it would of course be quite inconsistent with the Directive for a Member State to introduce, after its date, a measure of a kind which, had it been operative at the time of the entry into force of the Treaty, the Directive would have required that State to abolish. The Directive, by Article 2, covers measures ‘which hinder imports which could otherwise take place, including measures which make importation more difficult … than the disposal of domestic production’ and ‘in particular, measures which favour domestic products or grant them a preference’.

By paragraph 3 (s) of that Article the measures referred to must be taken to include those which ‘confine names which are not indicative of origin or source to domestic products only’.

In the French text of the Directive, paragraph 3 (s) refers to ‘les mesures … qui . .. reservent aux seuls produits nationaux des denominations ne constituant pas des appellations d'origine ou des indications de provenance’. I mention this because it appears that, in French law, the expressions ‘appellations d'origine’ and ‘indications de provenance’ are terms of art, whereas there is nothing technical about the English phrase ‘names which are not indicative of origin or source’. The equivalent in the German text of paragraph 3 (s) is ‘Maßnahmen … die nur den inländischen Waren Bezeichnungen vorbehalten, die weder Ursprungsbezeichnungen noch Herkunftsangaben sind’.

In relation to ‘Sekt’ and ‘Pradikatssekt’ the Commission relies also, for good measure, on Article 12 (2) (b) of Council Regulation (EEC) No 816/70 of 28 April 1970, which prohibits the application of any quantitative restriction or measure having equivalent effect in trade in wine with third countries.

The Commission's case in regard to ‘Prädikatssekt’ is in essence that, by reserving this particularly attractive name for a product made as to at least 60 % from home-grown grapes, the 1971 legislation favours the home-grown raw material and inhibits imports of foreign raw material.

The cornerstone of the Federal Republic's defence is that, although ‘Sekt’ and ‘Weinbrand’ were at one time generic terms, their meaning has over the years evolved so that, by 1971, they had come to connote essentially German products, manufactured according to methods prescribed by German law. Those terms had thus become indirect indications of origin, which, by virtue of Article 2 (3) (s) of Directive No 70/50, it was permissible to confine to domestic products. Alternatively, their protection as indications of origin was permitted by Article 36 of the Treaty under the heads of ‘public policy’ and of ‘the protection of industrial and commercial property’. By the same token it was legitimate to confine the term ‘Pradikatssekt’ to a product made from at least 60 % of home-grown grapes as this would serve to enhance its ‘German flavour’.

The Federal Republic cakes two subsidiary points.

First it denies that ‘Sekt’ and ‘Weinbrand’ have greater consumer appeal than ‘Qualitatsschaumwein’ and ‘Qualitatsbranntwein aus Wein’.

Secondly it contends that the 1971 legislation has not in fact had the effect of inhibiting imports. In support of this contention it adduces statistics showing growing imports into Germany since 1971 of sparkling wine and of brandy, particularly from France and Italy; and it offers to produce such further evidence as the Court may think appropriate on this point. The Commission meets this contention in two ways.

First it says that the growth in German imports of sparkling wine and of brandy over the years in question is due to the removal of other restrictions on such imports: viz. quotas in the case of sparkling wine and a system of licensing by a State monopoly in the case of brandy. Secondly it says that, in any event, as a matter of law, the question whether a particular measure is one having equivalent effect to a quantitative restriction must be judged according to whether that measure is apt by its nature to have that effect, regardless of whether it can be shown to have had that effect in fact.

In my opinion, my Lords, the latter proposition is unquestionably correct. It accords with common sense, for, as the Commission points out, there are many situations in which it is necessary to decide, before a measure is enacted, whether it would, if enacted, have equivalent effect to a quantitative restriction, and also many situations in which, owing to the complexity of the factors affecting trade, the effect on it of a particular measure cannot factually be isolated from the effects of other factors. Perhaps more pertinently still, the proposition accords with what was expressly decided by the Court in Case 8/74 Procureur du Roi v Dassonville [1974] ECR 837, at p. 852:

‘All trading rules enacted by Member States which are capable of hindering, directly or indirectly, actually or potentially, intra-Community trade arc to be considered as measures having an effect equivalent to quantitative restrictions.’

I would therefore reject the Federal Republic's second subsidiary point.

I think it convenient to leave aside for the moment its first subsidiary point and to proceed to consider its main point.

On this the Federal Republic has adduced a fairly copious body of evidence and has offered a great deal more. The Commission in contrast, has been sparing of evidence, preferring to rely, for the most part, on arguments based on the undisputed facts. I propose to deal with these arguments before turning to the evidence.

The Commission's first argument is based on the fact that before 1971 the German legislation about wine permitted the use of the names ‘Sekt’ and ‘Weinbrand’ for foreign products as well as for German products. Indeed, in the case of brandy, not being cognac, but made in the same way as cognac, the description ‘Weinbrand’ was prescribed. That being so, the argument, as I understand it, runs, those names cannot have connoted in the minds of the German public purely German products.

I find this argument unimpressive. Popular usage of words is not tied to what is permitted by legislation. The meanings of words in a language evolve, sometimes slowly, sometimes rapidly, for all sorts of reasons, what is permitted by the law being but one, and seldom the most telling, factor. An apposite example occurs to me, drawn from the English language. The word ‘claret’, derived from the French ‘clairet’, originally meant, in English, a light red wine of any provenance. Through a long process of evolution, it has come to mean, exclusively, red Bordeaux. From the greatest Chateaux to the humblest ‘Bordeaux rouge’, the red wines of Bordeaux are known in English as clarets. For anyone today to sell in England as ‘claret’ a wine from elsewhere than Bordeaux would, I think, unquestionably lay him open to prosecution under the Trade Descriptions Acts.

The Commission's second argument is based on the discrepancy between the allegation of the Federal Republic that ‘Sekt’ and ‘Weinbrand’, in the minds of the German public, connote essentially German products and the fact that the 1971 legislation permits the use of those names in relation to the products of any country on the whole of the territory of which German is an official language, thus covering not only German products but also those of Austria, Switzerland and Liechtenstein. This, says the Commission, shows that ‘Sekt’ and ‘Weinbrand’ are really generic terms in the German language and not indications of origin; and it makes nonsense of the Federal Republic's allegation that ‘Sekt’ and ‘Weinbrand’ are characterized by a particular ‘German flavour’ derived from their German methods of manufacture. Moreover, adds the Commission, no-one has ever heard of an indication of origin so wide as to cover the products of a group of countries.

These are telling points, but I think that they are satisfactorily met by the explanation given on behalf of the Federal Republic. The purpose, it says, of the ‘official language clause’ was to meet a particular problem that arose in relation to Austria. In that country too there is legislation regulating the use of the names ‘Sekt’ and ‘Weinbrand’ and the methods of manufacture of the products entitled to those names, those methods being similar to, albeit not the same as, the German. German imports of Austrian ‘Sekt’ and ‘Weinbrand’ are small and mostly confined to frontier areas. In that situation, the Federal Parliament was faced with the choice of either forbidding the use on German territory of the terms ‘Sekt’ and ‘Weinbrand’ for the Austrian products bearing those names in their homemarket, which would have been regarded by Austria as an unfriendly act, or of making a concession in favour of those products, a concession of only limited commercial significance. The Parliament chose the latter course. In other words its decision was based on diplomatic rather than on oenological or philological grounds. As regards Switzerland and Liechtenstein, the whole matter may be dismissed as de minimis, if not irrelevant, as there are no imports of sparkling wine or of brandy from those countries into Germany. Indeed there is no reasor to think that either of them makes sparkling wine (in the sense of ‘vin mousseux’, which is the relevant sense in the present case, as distinct from ‘vin pétillant’) or brandy, in commercial quantities.

A third argument advanced by the Commission is based on paragraph (6) of Article 75 of the Weingesetz of 1971. Article 75 is the last of that statute and deals with its commencement. Paragraph (6) is the last of Article 75 and provides, in the widest terms, that any wine, fortified wine, sparkling wine, winebased beverage, brandy or compound not complying with the requirements of the statute or of the regulations made thereunder may nonetheless be sold after the commencement of the statute under a name previously permitted, if the product in question was then already in a labelled container and complied with the relevant provisions in force at the date of its being put into that container. This, obviously, was a transitional provision, of no lasting significance, designed to spare traders the need to re-label products already on sale or prepared for sale. Yet the Commission suggests that the existence of that provision demonstrates that the Federal Parliament was not really concerned when enacting the Weingesetz of 1971, with protecting the consumer from being misled. To my mind, in putting forward that argument, the Commission is asking the Court to shed its sense of proportion. The Commission also suggests that the existence of Article 75 (6) demonstrates that the Parliament knew, when it passed the Weingesetz of 1971, that the names ‘Sekt’ and ‘Weinbrand’ were currently being used to describe foreign sparkling wines and brandies imported into Germany. In my opinion, this is to read a lot into a provision in such general terms; but, in any event, the question in the present case is not so much whether foreign products were being imported into Germany under the names ‘Sekt’ and ‘Weinbrand’, but whether it was legitimate to stop any imports of them under those names.

Fourthly, the Commission argues that if there was in fact any danger that German consumers might be misled, or German producers subjected to unfair competition, by the indiscriminate use of the names ‘Sekt’ and ‘Weinbrand’ for both German and foreign products, that danger could have been guarded against by requiring each bottle of sparkling wine and of brandy sold in Germany to be labelled with the name of its country of origin, as indeed was required by the pre — 1971 German legislation. The labels would then have borne such descriptions as ‘Deutscher Sekt’, ‘Französischer Sekt’ and so on.

In my opinion, there is a fallacy underlying that argument. Either ‘Sekt’ (for example) currently connotes in the German language a German product, or it does not. If it does not, that is the end of the matter: the word cannot be turned into an indication of origin by legislation, for no legislation can give birth to an indirect indication of origin; only usage can do that. But if the word ‘Sekt’ does currently connote in the German language a German product, if, in other words, it is the equivalent of a label ‘Made in Germany’, its inherent meaning should not be contradicted by coupling with it an adjective indicating a different provenance. Certain English authorities are, I think, helpful in illustrating the principle applicable here.

In English law, as in other systems of law, there are two aspects to the protection of indications of origin, namely on the one hand the protection of consumers against misleading descriptions of goods and, on the other hand, the protection of producers who have acquired what has been described in argument in this case as ‘collective commercial property rights’ against infringement of those rights. But in English law the distinction between those two aspects is particularly marked. The former rests on the Trade Descriptions Acts (replacing the earlier Merchandise Marks Acts) which enable prosecutions to be brought against traders who market goods under false descriptions. The latter owes nothing to statute. It rests on a development by the Courts of the common law tort of passing-off. Under both heads, however, the central theme is the same: that people should not be misled. So it is not surprising to find authority under both heads illustrating the principle to which I have referred.

In <span class="italic">Holmes v Vipers Ltd.</span> [1914] 1 K.B. 57 the respondent was prosecuted under the Merchandise Marks Act then in force for selling a bottle bearing the label ‘Fine British Tarragona Wine’. In fact the bottle contained a mixture composed as to 85 % of a wine made in England from dried raisins and as to 15 % of Mistella, a heavy form of Tarragona wine made and used solely for blending purposes. It was argued for the respondent that the presence on the label of the adjective ‘British’ would prevent a purchaser from being misled into thinking that the contents of the bottle were Tarragona wine. The argument was unanimously rejected by the King's Bench Divisional Court. The label was held to be false because it suggested that the bottle contained a British species of the genus Tarragona wine. A man who wanted Tarragona wine, because he had tasted it before or because it had been recommended to him, but who did not know where Tarragona was, could be misled.

In <span class="italic">J. Bollinger and others v Costa Brava Wine Co. Ltd.</span> [1961] 1 W.L.R. 277, the leading French champagne houses brought a passing-off action in the Chancery Division to restrain the defendant from selling as ‘Spanish Champagne’ a sparkling wine made in Spain from grapes grown in Spain. The defendant contended that the addition of the word ‘Spanish’ to the description of its wine prevented it from being mistaken for real champagne. Danckwerts J. (as he then was) rejected the contention. Adopting the test laid down in <span class="italic">Holmes v Pipers Ltd.</span>, he held that a substantial portion of the public was likely to be misled by the description. The evidence having established that the word ‘champagne’ meant, in England, wine produced in the Champagne district of France, it was untruthful to describe Spanish wine by that name, even with the addition of the adjective ‘Spanish’.

We have it on the authority of Lord Hunter that the law of Scotland should be regarded as being, in this respect, the same as that of England: see <span class="italic">Argyllshire Weavers Ltd. and others v A. Macaulay (Tweeds) Ltd.</span> (the ‘Harris Tweed’ case) [1964] R.P.C. 477, at p. 569.

I do not, my Lords, overlook the decision in the ‘sherry’ case, <span class="italic">Vine Products Ltd. v Mackenzie & Co. Ltd.</span> [1969] R.P.C. 1. In that case Cross J. (as he then was) held that the word ‘sherry’, standing alone, meant, in England, a wine from the Jerez district of Spain (the word ‘sherry’ being an English corruption of the name of that town, which was called ‘Shereesh’ by its Moorish conquerors, then, later, Xeres, and finally Jerez). But he went on to hold that it was permissible for the plaintiffs to call their imitations of sherry by such names as ‘British sherry’, ‘English sherry’, ‘south African sherry’, ‘Cyprus sherry’, ‘Australian sherry’ and even ‘Empire sherry’. His decision, however, turned entirely on the fact that the producers of real sherry had, for decades, stood by and allowed the plaintiffs to use these names without objecting. Those producers were thus precluded, under the equitable doctrine of laches, from asserting the right that they would otherwise have had to restrain the misuse of the name.

I conclude that the Commission's fourth argument must fail.

Fifthly, the Commission argues that the entire territory of a state cannot be the subject of an indirect indication of origin. The Federal Republic counters this submission by referring to a judgment of the Landgericht of Dortmund of 15 June 1970 in which that Court held that ‘Korn’ and ‘Weizen’ (meaning, in German, respectively ‘corn’ and ‘wheat’) were indirect indications of origin for grain spirits made in Germany, because the majority of German consumers thought of them as such. The Commission criticizes the decision as being wrong in German law. Be that as it may, I can see no logical reason why an indirect indication of origin cannot relate to the whole territory of a state, when a direct one certainly can, e.g. Irish Whiskey, and ‘vin d'Algérie’.

Sixthly, the Commission argues that the provisions of the 1971 legislation to which it objects were of a kind unprecedented in German law. It appears that, in Germany, the protection of indications of origin is in the main secured by means of Article 3 of the general statute on unfair competition of 7 June 1909 (‘UWG’) as amended by further statutes of 21 July 1965 and 26 June 1969. That Article enables civil proceedings to be brought against anyone who, in the course of trade and with a view to obtaining a commercial advantage, misdescribes, among other things, the origin of goods. The proceedings can be brought by other traders, or by trade or consumer associations. The Commission contends that the correct course, in the present case, to establish that ‘Sekt’ and ‘Weinbrand’ were, according to popular usage, indicative of German products, would have been for proceedings to have been brought under that enactment against importers of foreign products described by those names. The Commission further contends that, on the authority of certain decisions of the Bundesgerichtshof, the proceedings would have failed. It concludes that the 1971 legislation was arbitrary and designed to give to German producers of sparkling wine and of brandy a competitive advantage which the previous German law would have denied to them.

I do not think, my Lords, that it is for this Court to form a view on these matters. The question for this Court is not whether the enactment of the 1971 legislation was consistent with German legal tradition, nor is it whether, in the absence of that legislation, a German Court could have come to this or that conclusion. The question for this Court is simply whether the 1971 legislation is compatible with Community law.

In order to answer that question, it is, if I may state the obvious, first necessary to consider what the relevant Community law is.

In so doing I propose to leave aside Article 12 (2) (b) of Regulation No 816/70, not because it is irrelevant, but because it throws no light on the case, one way or the other. If the Commission is right in saying that the 1971 legislation enacted measures having an effect equivalent to quantitative restrictions in breach of Article 30 of the Treaty, those measures must be condemned, and it adds nothing helpful to the decision of the case to say that, in relation to sparkling wines, they were in breach also of that Regulation. If, on the other hand, the measures in question are permissible under the provisions of the Treaty about trade between Member States, then <span class="italic">a fortiori</span> must they be permissible under the provisions of a Regulation about trade with third countries.

As I mentioned earlier, the Federal Republic puts its case in the alternative. It relies in the first place on Article 2 (3) (s) of Directive No 70/50, saying that this provision affords a binding interpretation of Article 30, excluding from the scope of that Article measures designed to protect indications of origin. In support of that view, the Federal Republic refers to Article 5 (2) of the Directive, which provides that the Directive ‘shall apply without prejudice to the application … of Article 36 … of the EEC Treaty’. Alternatively the Federal Republic relies on Article 36, contending that the protection of indications of origin by the legislation of Member States is permitted by that Article, in so far as such legislation is concerned to protect the consumer, under the head of ‘public policy’ and, in so far as it is concerned to protect producers, under the head of ‘the protection of industrial and commercial property’.

My Lords, I agree with the Commission that that approach is unnecessarily complicated and that it is misconceived. The basic provisions, after all, are those of the Treaty. The Directive can do no more than seek to implement them. It is a legal solecism to regard the Directive as interpretative of the Treaty. The provisions of Article 5 (2) of the Directive are readily understandable and explicable, if one looks at that instrument as a whole, without any need to attribute to the Commission the intention, by adopting them, to arrogate to itself the function of interpreting Article 30 by delegated legislation.

The key thus lies in Article 36, for there can be no doubt that national legislation confining in a Member State the use of a particular name, at all events if that name has consumer-appeal, to a domestic product, must <span class="italic">prima facie</span> infringe Article 30, as that Article has been interpreted by the Court, notably in the <span class="italic">Dassonville</span> case.

It is, I think, common ground between the Commission and the Federal Republic that the protection of indications of origin has the two aspects that I have mentioned: on the one hand the protection of the consumer and on the other the protection of producers. Nor do I doubt that the Federal Republic is right in saying that the first aspect is covered, in Article 36, by ‘public policy’ and the second by ‘the protection of industrial and commercial property’.

But Article 36 does not leave a free hand to Member States to enact any prohibitions or restrictions they please in the name of ‘public policy’ or of ‘the protection of industrial and commercial property’. In the first place, such prohibitions or restrictions must be ‘justified’ on those grounds. Secondly, the requirement must be observed that ‘such prohibitions or restrictions shall not constitute a means of arbitrary discrimination or a disguised restriction on trade between Member States’.

Reference was made, on behalf of the Federal Republic, to Article 222 of the Treaty, which provides, in general terms, that the Treaty ‘shall in no way prejudice the rules in Member States governing the system of property ownership’. It seems to me, with all respect, that this provision is but remotely in point.

The question, I think, is: what does Community law permit Member States to enact in the field of the protection of indications of origin?

In answering that question, one must of course have regard to the legal traditions of Member States, for it is against the background of those traditions that the Treaty, and with it the Directive, should be interpreted. I do not, for my part, think that there is here called for a meticulous examination of the law of each Member State such as is appropriate in some cases that come before the Court. But, having outlined to Your Lordships the relevant English and German law, I think that I should be presenting to Your Lordships a limping Opinion unless, on a subject such as this, I referred also to French law.

Happily, my task in this respect is made easy because the Federal Republic annexed to its Rejoinder a report by Professor Plaisant of the University of Le Mans.

There is no doubt that, in the field of the protection of indications of origin, French legislation has been a pioneer and that, as a result, French law is, in many respects, more advanced in that field than that of any other Member State. Professor Plaisant explains the two basic concepts of French law, the ‘appellation d'origine’ and the ‘indication de provenance’. The former is defined, by a French statute of 6 May 1919, re-enacted in 1966, as follows:

‘Constitue une appellation d'origine la denomination d'un pays, d'une region ou d'une localité servant à désigner un produit qui en est originaire et dont la qualité ou les caractères sont dus au milieu géographique, comprenant des facteurs naturels et des facteurs humains.’

There seems to be no such statutory definition, in French law, of an ‘indication de provenance’. Professor Plaisant's opinion is that, whereas there are four requirements for an ‘appellation d'origine’, namely:

a geographic description,

connoting a product from the place thus described,

having an adequate reputation,

and the merits or characteristics of the product being due exclusively or essentially to natural and human factors connected with that place,

only the first three of those requirements are called for in the case of an ‘indication de provenance’. Moreover Professor Plaisant makes it clear that, when he refers to a geographic description, he does not exclude a name which, whilst not in itself geographical has in the popular mind a geographical connotation — in other words an indirect indication of origin.

Professor Plaisant also refers to three international agreements.

Of these the first is the International Convention for the Protection of Industrial Property, signed in Paris in 1883, and revised at Brussels in 1900, Washington in 1911, The Hague in 1925, London in 1934, Lisbon in 1958 and Stockholm in 1967. This created what is sometimes called the ‘Paris Union’. All the Member States of the EEC are parties to it, though not all to the same extent. I do not, for my part, think that this matters, because I do not think, with all respect to Professor Plaisant, that a perusal of the Paris Convention, in any of its versions, throws any light on any question Your Lordships have to decide in the present case.

The second international agreement referred to by Professor Plaisant is the Arrangement of Madrid. This was entered into in 1891 and has been revised several times since, always in conjunction with revisions of the Paris Convention. Only five Member States are parties to it, namely the Federal Republic, France, Ireland, Italy and the United Kingdom. Again, it does not, to my mind, throw any light on any question at issue in this case.

The third international agreement referred to by Professor Plaisant is the Arrangement of Lisbon. This was entered into in 1958, in connexion with that year's revision of the Paris Convention. Only two Member States of the EEC were parties to it, France and Italy. This Arrangement, of which the French text only is authentic (although official translations of it exist in other languages) contains, in Article 2, a definition of ‘appellation d'origine’ which is narrower than that contained in French statute law. It seems that that definition is picked up in a recent Belgian statute.

In my opinion, the definition contained in the Lisbon Arrangement, and reproduced in the Belgian statute, affords no guidance in interpreting Article 36 of the EEC Treaty, for it excludes indirect indications of origin of any kind and also indications of origin that are mere ‘indications de provenance’. I cannot think that the authors of the Treaty intended, by Articles 30 to 36, to abolish all protection in Member States for these two important categories of trade descriptions.

I conclude that Member States are permitted by Article 36 of the Treaty and Article 2 (3) (s) of Directive No 70/50 to adopt any reasonable measures to protect indications of origin, whether direct or indirect, and whether they would be classed in French law as ‘appellations d'origine’ or as ‘indications de provenance’. (As to the requirement of reasonableness, see para. 6 of the Judgment of the Court in the <span class="italic">Dassonville</span> case).

What distinguishes an indirect from a direct indication of origin is of course that, in the case of the former, popular or, in some cases, trade usage attributes a geographical connotation to a word which is not inherently a geographical name. I have mentioned the example of ‘claret’ in English. Other examples were referred to in argument. In my opinion, the crucial question in the present case is whether ‘Sekt’ and ‘Weinbrand’ have come to connote, in the minds of the German public, German products, or whether, as the Commission asserts, they remain generic German terms for ‘sparkling wine’ and ‘brandy’. A slightly more complex question arises as regards ‘Prädikatssekt’, to which I shall come.

Before I turn to the evidence, I should mention one fact which is, I think, common ground. This is that, before the Treaty of Versailles came into force, ‘Champagner’ and ‘Kognac’ were much used in Germany as generic terms for ‘sparkling wine’ and ‘brandy’ respectively. The use of those terms for other than real champagne and real cognac was, however, forbidden as from 1923 pursuant to Articles 274 and 275 of that Treaty.

It seems that etymologically the word ‘Sekt’ is derived from the latin ‘<span class="italic">vinum siccatum</span>’, meaning simply ‘dry wine’. Paul's ‘Deutsches Worterbuch’ (sixth ed. 1968) cites the use of the word in 1830 by a famous German actor, Ludwig Devrient, in the role of Falstaff, to translate ‘Give me a cup of sack’ (Henry IV Part I, Act 2, Scene 4). Shakespeare's ‘sack’ was, of course, white wine imported from Spain or the Canaries. It was, in his day, considered dry, but it is difficult to tell whether it would have seemed so to a modern palate. At all events, dictionaries confirm that the English word ‘sack’ was derived from the French ‘sec’.

According to the Defence (p. 11), which is in this respect uncontradicted by the Commission, ‘Sekt’ was first introduced in Germany as a name for sparkling wine in about 1880; and, in 1908, the association of German sparkling wine manufacturers changed its name from ‘Verband Deutscher Schaumweinkellereien’ to ‘Verband Deutscher Sektkellereien’. Between 1923 and 1971, the legislation in force in Germany treated ‘Sekt’ and ‘Schaumwein’ as synonyms.

I turn to the evidence about current usage of the word ‘Sekt’.

The Commission relies firstly on a Judgment of the Bundesgerichtshof of 19 June 1970 (BGH NJW 1970, 2105) in which that Court repeatedly used the expression ‘Sekt’ in relation to French sparkling wine.

Secondly the Commission relies on the comments of the German association for the protection of industrial property and copyright (the ‘Deutsche Vereinigung fur gewerblichen Rechtsschutz und Urheberrecht’) on a bill introduced in the Bundestag in 1967 which was an early version of what eventually became the Weingesetz of 1971. In these comments, addressed to the President of the Bundestag on 12 September 1967, that association, on the basis of a report from its technical committee, forcefully objected to the proposal that the terms ‘Sekt’ and ‘Weinbrand’ should be reserved for German products and for the products of countries where German was an official language (see pp. 16 and 17 of the Application).

Thirdly the Commission relies on an article by Dr Walter Brogsitter of Munich on the same bill, in which he expressed the view that the proposal in question was unjustifiable (p. 17 of the Application).

The Commission also refers to some proceedings at present pending in the Bundesverfassungsgericht, in which two French manufacturers and four German importers of sparkling wines challenge the compatibility of the 1971 legislation about ‘Sekt’ with the Federal Constitution. The Commission submits that the fact that those manufacturers and importers have taken the trouble to bring those proceedings evinces, among other things, their concern over the reservation of that name for the products of German-speaking countries.

The Federal Republic relies to some extent on the provisions of certain bilateral treaties that it has negotiated with other countries, in particular a treaty with France dated 8 March 1960, a treaty with Italy of 2 June 1961, a treaty with Switzerland of 7 March 1967 and a treaty with Spain of 11 September 1970. In each of these treaties the other contracting state agreed to protect on its territory the descriptions ‘deutscher Sekt’ and ‘deutscher Weinbrand’. Professor Plaisant gave it as his opinion that the consequence was that ‘Sekt’ and ‘Weinbrand’ were protected in France as names for German products, because of their similarity to 'the descriptions ‘deutscher Sekt’ and ‘deutscher Weinbrand’. At the hearing he added that he had never heard of an indication of origin that was protected abroad but not at home. The implication, as I understood it, was that, as ‘Sekt’ and ‘Weinbrand’ were protected in France as names for German products, they must be so in Germany also.

My Lords, I confess that I am unimpressed by these points. Nothing is more natural than that the use, to describe a product, of its generic name in the language of one country should be regarded by people in other countries as an indication of its origin from that country. An example given in the course of argument was ‘Slivowitz’. I do not doubt that to most of us this connotes, as was submitted, a product of Yugoslavia. But it does not follow that that is its meaning to a Yugoslav, for, I understand, in Serbo-Croat ‘Slivowitz’ merely means a spirit distilled from plums. Another example is ‘Grappa’. To any Englishman who has heard of it that name connotes an Italian product, but I understand it to be open to question whether that is its connotation in Italian. Of course ‘Sekt’ and ‘Weinbrand’ sound like German products to those who are not German. Indeed, were it not for the existence of Austrian ‘Sekt’ and ‘Weinbrand’, I would have agreed with the Commission that it told against the Federal Republic's contention that it did not seek or obtain from the states with which it was entering into those bilateral treaties the protection of those names without the adjective ‘deutscher’. At all events, the question at issue in this case is not what ‘Sekt’ and ‘Weinbrand’ may mean to Frenchmen, Italians, Swiss, or Spaniards, but what those words mean to Germans. And it does not seem to me, with all respect to Professor Plaisant, that the principles of French law to which he referred, can have the slightest bearing on that question.

Of more pertinence, I think, is the reference made on behalf of the Federal Republic to the terminology of the Community legislation on the common organization of the market in wine. Throughout the German texts of Council Regulation (EEC) No 816/70 and of Council Regulation (EEC) No 2893/74 the word used for ‘sparkling wine’ is ‘Schaumwein’, whilst in Council Regulation (EEC) No 2894/74 one finds ‘Qualitatsschaumwein’ used for ‘quality sparkling wine’. The word ‘Sekt’ is not, so far as I have been able to find, used in that legislation. This is consistent, of course, with the view that Schaumwein is, in German, the generic term for sparkling wine.

At the hearing, there were put in evidence on behalf of the Federal Republic a number of labels found on bottles of French, Italian and Spanish sparkling wines sold in Germany before the 1971 legislation came into force. The significant point about these labels is that they did not describe the contents of those bottles as ‘Sekt’, but as ‘Französischer Schaumwein’, 'Italienischer “Schaumwein” or “Spanischer Schaumwein”. Most of the French wines in question were champagnes from well-known houses, and their labels also bore, in each case, the description “Champagne” and the name of the house whose wine it was. One could think that, with such famous names available to them, the producers and importers of those wines were indifferent whether they described them as “Französischer Sekt” or “Französischer Schaumwein”. But two of the French wines in question were humble “vins mousseux”. There was one brand of Asti Spumante, and there were two brands of Spanish sparkling wines. These labels were, as I understand it, put in primarily as evidence in support of what I have called the Federal Republic's first subsidiary point i.e. its allegation that “Qualitätsschwaumwein” does not have less consumer appeal than “Sekt” — but the inference can be drawn from them that the producers and importers concerned regarded “Sekt” as connoting a German, or perhaps Austrian, wine and did not want their wines to be thought of as such.

The Commission countered this evidence with an offer to prove, first, that many French producers of “vins mousseux” had, since 1946, used the description “Französischer Sekt” for their exports to Germany and had complained to the Commission, on 16 December 1969, of the proposed new German legislation and, secondly, that other French producers of “vins mousseux”, as well as the well-known Letzeburgish firm of Bernard Massard, used the description “Sekt” on the labels for their exports to Germany. More concretely, the Commission offered to put in evidence labels in its possession tending to establish those facts. On behalf of the Federal Republic it was rejoined that, in fact, Messrs Bernard Massard supplied the German market from their cellars in Trier, so that they were treated under the 1971 legislation as German producers, entitled to use the description “Sekt” for their products.

My Lords, as I shall mention in a moment, I think that the evidence at present before the Court on the meaning of the word “Sekt” in current German usage is inconclusive. For that reason, I shall propose to Your Lordships at least one measure of inquiry on that question. If Your Lordships should agree that that measure ought to be undertaken, there will be no harm in the Commission being given leave to put in evidence, at the same time, such labels as it has, to prove the facts that it alleges. On the other hand, I doubt whether the case would be advanced by evidence of complaints made to the Commission in 1969 by French producers of “vins mousseux” or by a detailed analysis of Messrs Bernard Massard's business.

At the hearing, my Lord the Judge Rapporteur asked Counsel for the Federal Republic to what extent the descriptions “Sekt” and “Weinbrand” were applied commercially, before the 1971 legislation came into force, to products imported into the Federal Republic. The answer was that no statistics as to this existed. But, in a supplementary written answer dated 11 December 1974, the Federal Republic emphasized that, in the case both of sparkling wine and of brandy, imports accounted for a very small proportion of national consumption. It gave as an example the figures for 1966, in which year, of nearly 120 million bottles of sparkling wine sold in the Federal Republic, imports accounted for only just over 6 million.

Quite rightly, in my opinion, the Federal Republic contends that the crucial evidence, in this case, on the question of current usage in Germany, is that afforded by public opinion surveys. I do not say that such evidence must always be pre-eminent in cases about indirect indications of origin. Cases will occur where evidence of trade usage will be more pertinent. But what distinguishes this case from other possible cases is that it raises the question whether particular German words do or do not connote German products in the German language. As to that, it seems to me that the opinion of the generality of German consumers must be the definitive guide.

Annex 1 to the Defence is a report by the “Institut für Demoskopie Allensbach” on a public opinion survey that it carried out in 1966 at the instance of the Verband Deutscher Sektkellereien. The object of the survey is stated in that report to have been to establish whether “Sekt” meant to the generality of German consumers a German product. The survey was based on a representative sample of the adult population of the Federal Republic and of West Berlin. About 3000 people aged over 16 were interviewed. They were first asked whether they had ever heard of the word “Sekt”. Only 1 % had not. The remaining 99 % were asked whether, when they read or heard about “Sekt”, the term connoted to them a German product or a foreign product. 76 % answered that it connoted to them a German product, 11 % that it connoted a foreign product and 12 % did not know. Lastly they were asked whether they sometimes drank or bought “Sekt”. To this question 48 % of the sample answered in the affirmative. Of those 48 %, 83 % were among those who thought that “Sekt” connoted a German product, 10 % among those to whom it connoted a foreign product, and 7 % among those who did not know.

Those are, at first sight, impressive figures. The Commission, however, whilst not challenging either the integrity or the competence of the Institut für Demoskopie Allensbach, criticizes the survey on two grounds.

First the Commission says that the sample should not have included people as young as 16. I do not, for my part, think that this is a serious criticism, particularly in view of the fact that the views of those who “sometimes drank or bought Sekt” were separately collated. But I think that the Commission's second criticism is valid. That criticism is that it was tendentious to ask whether “Sekt” connoted a German or a foreign product, when the true question was whether it connoted a German product or a product that might come either from Germany or from elsewhere.

Conscious, perhaps, of the validity of this criticism, the Federal Republic envisaged in its Defence (at pp. 24 and 54) that a fresh public opinion survey might be undertaken as a measure of inquiry to be ordered by the Court in these proceedings and on lines to be laid down by the Court. The Commission, in its Reply (at pp. 15 and 28), suggested that, if such a fresh public opinion survey were to be undertaken, it should also be designed to ascertain whether or not the name “Sekt” had greater consumer appeal than “Qualitätsschaumwein”. The Federal Republic, in its Rejoinder (at p. 30), welcomed this suggestion. It added the further suggestion that the Court should call for an expert's report on the manner in which the questions to be put in the survey should be formulated. The Federal Republic raised at the same time a question that its Counsel stressed repeatedly at the hearing, viz. the question of the burden of proof, its point being that that burden is, in these proceedings, on the Commission.

My Lords, it is of course true that, in proceedings under Article 169, the burden of proof is on the Commission. But it seems to me that, even in such proceedings that burden may shift according to the course taken by the pleadings; and I am not sure that where, as here, a Member State alleges that what was once a generic term in its language has become an indirect indication of origin, the burden does not shift to that State to make the allegation good. At all events I should be loath to decide between the Commission and a Member State a question of such lasting importance as the present one, in a largely uncharted field of Community law, according to the burden of proof.

On the other hand I do not think that the evidence at present before the Court is sufficient to enable the Court to decide the question. I am therefore of the opinion that the Court should, pursuant to Article 60 of the Rules of Procedure, order that a fresh public opinion survey be conducted in Germany to ascertain (i) whether or not, to the German public, the term “Sekt” has a greater appeal than the term “Qualitatsschaumwein” and (ii) whether, to that public, the term “Sekt” connotes a German product or a product that may come either from Germany or from elsewhere. I do not think that it is necessary for the Court first to receive an expert's report on the way in which the questions to be put in that survey should be formulated. It seems to me that parties of the standing and responsibility of the Commission and of the Federal Republic ought to be able to agree on the formulation of those questions and that it should be enough for the Court to provide in effect, pursuant to Article 45 (3) nf the Rules of Procedure, that in the event of any disagreement between them the matter shall be determined by the Judge Rapporteur, after hearing the Advocate-General.

In my opinion no further evidence is necessary to enable the Court to reach a decision as to “Prädikatssekt” or as to “Weinbrand”.

As I have already mentioned, the Federal Republic's contention with regard to “Prädikatssekt” is essentially that for the 1971 legislation to confine that description to a sparkling wine made as to at least 60 % from home-grown grapes is justified because this serves to enhance the “German flavour” of the product.

My Lords, even assuming, in the Federal Republic's favour, that “Sekt” connotes a German product, I do not think that that contention as to “Pradikatssekt” can be accepted. The Federal Republic asserts over and over again in its pleadings that what distinguishes “Sekt” from foreign sparkling wines is the method of manufacture. Nor is there any suggestion that “Sekt”, to be recognized as such by the German public, must be made from German grapes to any extent. Indeed there is no allegation that the description “Pradikatssekt” has any meaning to the generality of German consumers other than that of “choice Sekt”.

I might, conceivably, have come to a different conclusion if the 1971 legislation had required “Pradikatssekt” to be made from grapes of a particular variety, e.g. Riesling, or perhaps even from a selection of particularly esteemed varieties, or perhaps again from grapes grown in one or more particularly favoured districts. But that legislation does none of these things. It permits any grapes to be used to complete the 60 % German quota, whatever the variety, and wherever they may have been grown, provided that they have been grown on German soil. This means that they may be of very differing varieties and may have been grown anywhere from the Moselle to Franconia, from the Rheingau to Baden. To suggest that such an indiscriminate mixture produces a distinctive wine overtaxes my credulity.

In saying this I am not of course rejecting the point made on behalf of the Federal Republic that, in general, sparkling wines of quality are made from wines which would rank low, as table wines, if kept still. I am saying only that I can detect neither a philological nor an oenological basis for its reservation of the name “Pradikatssekt” for a wine made as to 60 % from home-grown grapes.

I would therefore hold that, in reserving the name “Pradikatssekt” for a wine produced as to a specified proportion from home-grown grapes, the Federal Republic is in breach of its obligations under the Treaty.

I turn, lastly, to “Weinbrand”.

Here the evidence adduced on behalf of the Commission does not include the phraseology of any Judgment of the Bundesgerichtshof, but it does include the comments of the German association for the protection of industrial property and copyright and those of Doctor Brogsitter that I have mentioned in connexion with “Sekt”.

It also includes a reference to two sets of proceedings pending in the Bundesverfassungsgericht, one set brought by three German importers of foreign spirits and the other brought by two Italian brandy distillers and their German subsidiaries. Here again, the relevance of the existence of those proceedings is, of course, that it evinces the concern of those who brought them over the reservation of the name “Weinbrand” for brandy from German-speaking countries.

In addition the Commission refers to the nomenclature adopted by the German Federal Statistical Office (to classify, I infer, imports into Germany): this includes an item “Cognac, Armagnac und andere Weinbrande”, which seems unquestionably to mean “Cognac, Armagnac and other brandies”.

Perhaps, the best point put forward on behalf of the Commission is that, whereas there was always a German word to rival “Sekt” as a generic term for sparkling wine, i.e. “Schaumwein”, there was none to rival “Weinbrand” for brandy, the phrase “Branntwein aus Wein” being, as I understand is common ground, an innovation introduced by the authors of the 1971 legislation. “Branntwein” simpliciter means “spirit”.

This point was however undermined by the Commission itself, when it asserted that, until recently, imports of brandy into Germany were licensed by a State monopoly, which confined them almost entirely to cognac and armagnac. In this connection, the statistics supplied by the Federal Republic in its supplementary answer of 11 December 1974 are interesting. It appears from these that total German production of brandy in 1966 amounted to 943618 hi whilst imports amounted to only 59516 hl of which 42881 were accounted for by cognac and armagnac. One can readily believe, in those circumstances, that German consumers were induced to think of Weinbrand as essentially a domestic product. The Federal Republic alleges that, in spite of the Treaty of Versailles, to the common man, the generic term in Germany for brandy remains “Kognac”. Indeed the Federal Republic goes so far as to ask the Court to order a public opinion survey to establish this fact. The Commission argues that the Court should follow a decision of the Bundesgerichtshof (BGHZ 30357/365), where it was held that, when a monopoly operates in a market, no generic term used by the monopolist can, while the monopoly subsists, become an indication of origin. I do not for my part think that the Court should regard itself as bound by that decision. It seems to me that the essential question for this Court, in a case such as the present, is not why a particular popular usage came into existence, but simply whether it exists; for, in my opinion, the public is entitled to be protected against misleading descriptions, no matter why, as a matter of history, those descriptions came to have their particular connotations.

According to the Federal Republic (Defence p. 11) “Weinbrand” is a very old German word, of which the earliest recorded form is to be found in a reference, in a document of 1319, to one “Diederich dictus Winbrant”, who was a brandy distiller. It appears that the word was deliberately introduced, early in the 20th century, by one Hugo Ansbach as a commercial description of brandies distilled in Germany. Paul's “Deutsches Worterbuch” attributes its introduction into general use to the prohibition of the use of the word “Kognac” for German products under the Treaty of Versailles.

Annexed to the Defence (as Annexes 2,3 and 4) are the reports of no fewer than three public opinion surveys conducted in Germany as to the meaning of the word “Weinbrand”.

The first of those surveys was carried out in 1966 by the Institut für Demoskopie Allensbach at the request of the German association of brandy distillers (the “Verband der Weinbrennereien”). It was parallel in all respects to the survey carried out by that institute, at the same time, about “Sekt” and it is open to the same criticism, in that the crucial question asked was whether the person interviewed, if offered a glass of Weinbrand, would think, of it as a German product or as a foreign product. Of the persons interviewed, 2 % had never heard of Weinbrand, 75 % said that they thought of it as a German product, 5 % that they thought of it as a foreign product and 18 % did not know. Of those who said that they did sometimes drink or buy Weinbrand (47 % of the Sample), 85 % said that they thought of it as a German product, 5 % that they thought of it as a foreign product and 10 % did not know.

The other two surveys were carried out in 1973 by a different agency, the G.F.K. of Nuremberg. These surveys were basically similar, but the crucial question, in each of them, was put differently. In the first, each interviewee was asked: ‘When you read the description “Weinbrand” on a bottle of spirits, from what country do you think that the product comes?’ In the second, he or she was asked: ‘When you read the description “Weinbrand” on a bottle of spirits, do you think of it as a German product or do you think that it could as well be a foreign product?’

My Lords, it seems to me that, whilst the formulation of the question in the first of those surveys is open to criticism, in that it suggests that there is only one country from which Weinbrand can come, there can be no serious criticism of its formulation in the second. The results of these two surveys were as follows. In the first, 80 % of the sample answered ‘Germany’, 20 % answered with the names of other countries or areas, 3 % answered ‘from various countries’ and 4 % did not know. Seemingly, the reason why the total exceeds 100 % is that some interviewees gave several answers. This appears to have happened in the second survey also, but to a lesser extent. Here 87 % of the sample thought that the bottle would contain a German product, 13 % that it could as well contain a foreign product and 1 % did not know.

A perfectionist could, I think, criticize some aspects even of the last survey. But, to my mind, the only reasonable conclusion that can be drawn from the evidence afforded by these surveys, against the background of the other evidence adduced by the parties, is that the Federal Republic has made good its allegation that, for the vast majority of German people, ‘Weinbrand’ connotes a German product.

If that is right, the Federal Republic's defence succeeds as regards ‘Weinbrand’, and there is no need to consider the subsidiary question whether that name has greater consumer appeal than ‘Qualitatsbranntwein aus Wein’.

In the result, I am of the opinion that Your Lordships should —

(1)declare that, in reserving the name ‘Prädikatssekt’ for sparkling wines produced as to a specified proportion from home-grown grapes, the Federal Republic of Germany is in breach of its obligations under the EEC Treaty;

(2)declare that, in reserving the name ‘Weinbrand’ for brandies produced in any country on the whole of the territory of which German is an official language, the Federal Republic is not in breach of those obligations;

(3)order that there shall, as a measure of inquiry in these proceedings, be carried out in Germany a public opinion survey on the questions (i) whether or not, to the German public, the name ‘Sekt’ has a greater appeal than the name ‘Qualitatsschaumwein’ and (ii) whether, to that public, the name ‘Sekt’ connotes a German product or a product that may come either from Germany or from elsewhere;

(4)assign the conduct of that measure of inquiry to the Judge Rapporteur and fix a time limit (which I suggest might be 4 weeks) for the parties to submit to the Judge Rapporteur agreed proposals for the conduct of the survey;

(5)order that the Commission have leave to put in evidence such labels as are in its possession tending to prove the facts that it alleges about imports into Germany of sparkling wines as ‘Sekt’; and

(6)reserve costs.