In June the IPKat commented [writes his distinguished and practical friend Richard Ashmead] on a recent reference to the Court of Justice of the European Union (CJEU) which arose from a UK trade mark application for IP TRANSLATOR (here), seeing it as “a much-needed and very welcome reference to the CJEU of some questions which could greatly benefit from a preliminary ruling ...”. The questions in need of a ruling are about the “class heading dispute”, a term which may suggest no more than an argument over classification. It is not.
1. The list of goods as registered defines the boundary between what does, and therefore also what does not, infringe a registered trade mark. It tells you whether a third party's goods can infringe because they are identical or, on the evidence, similar, to those for which a mark is registered.
2. To consider if there is infringement, one reads the list of goods for which the mark is registered and check whether third party goods are listed there. If they are, they infringe as being identical. If not you move on to possible similarity and the heavier burden of proof which that involves.
3. Class heading registrations, with list(s) which are the same as Nice Classification Class Headings, are commonplace in the EU. Our dispute centres on how you read these. The two sides of the dispute require you to apply either
a. the "means-what-it-says" approach: construe the words of the class heading list essentially according to their own dictionary significance, or
b. the “class-heading-covers-all” approach: construe the class XX heading list as if it reads “All goods in class XX”.
4. All classes include extra goods not in the class heading wording, specifically or by genus/species significance.
5. Under "means-what-it-says", these extra goods cannot infringe as identical goods, but may infringe as similar goods if the evidence demonstrates that there exists a likelihood of confusion.
6. Under “class-heading-covers-all”, these extra goods are deemed to be within the class heading wording as registered and so infringe as identical goods. There can in these circumstances be no similar goods within class XX, only identical. As examples, under “class-heading-covers-all”:
a. stands for musical instruments in class 15 infringe a registration for “Musical instruments” (the class 15 heading) as identical
b. repair outfits for inner tubes in class 12 infringe a registration for “Vehicles; apparatus for locomotion by land, air or water” (the class 12 heading) as identical, and
c. sleeping bags for camping in class 20 infringe a registration for “Furniture, mirrors, picture frames; goods (not included in other classes) of wood, cork, reed, cane, wicker, horn, bone, ivory, whalebone, shell, amber, mother-of-pearl, meerschaum and substitutes for all these materials, or of plastics” (the class 20 heading) as identical.
These examples do not infringe as existing goods under "means-what-it-says", although they may each infringe as similar goods, with supporting evidence.
7. The protection afforded to a registered trade mark is absolute in the case of identity between the marks and goods, with no defence on evidence relating to the goods.
8. Similarity of goods is arguable, subject to evidence of a likelihood of confusion on the part of the public. Stands for musical instruments may on the evidence be similar to musical instruments themselves but, under “class-heading-covers-all”, the defence argument disappears.
9. Current indications are that the IPOs of AT, BX, CH, CZ, DE, DK, ES, FR, IE, LV, NO, PL, SE, SI, SK and UK apply "means-what-it-says", while the IPOs of BG, EE, FI, HU, IT, LT and RO (along, most importantly, with OHIM) apply “class-heading-covers-all” (thank you, MARQUES, for this data).
10. As a final sting in the “class-heading-covers-all” approach, OHIM’s version of it as explained it its 2003 Presidential Communication No. 4/03 not only extends the scope of a class heading registration to goods not within its wording, but does the same to any registration using a “general indication” from a class heading, giving those words a goods scope beyond their dictionary meanings. As an example under “class-heading-covers-all” a registration list reading simply “data processing equipment and computers” (which appears in the class 9 heading) is infringed by computer software as identical goods, but not under "means-what-it-says" unless the evidence supports similarity.Both approaches cannot be right, and their coexistence across the EU brings uncertainty as to precisely what it is that class heading registrations (and others) cover. This uncertainty is also likely to produce conflicting decisions, giving equivalent EU national registrations different scopes.
The CJEU reference in IP TRANSLATOR (here) was made in May 2010 by Geoffrey Hobbs QC, sitting as an Appointed Person. It converges its questions on OHIM’s explanation of “class-heading-covers-all” in Presidential Communication No. 4/03, asking whether it is:
“....necessary or permissible for such use of the general words of the Class Headings of the said International Classification of Goods and Services to be interpreted in accordance with Communication No. 4/03 of the President of the Office ......”In essence this asks the CJEU whether "means-what-it-says" or “class-heading-covers-all” is correct in EU law.
The IPKat commented earlier that he saw justification for both positions, despite preferring the "means-what-it-says" practice. He continued to support a 'one-way-fits-all-offices’ solution, whichever it is. On that latter point there is widespread support. The UK IPO, in the IP TRANSLATOR appeal hearing, commented that the dispute puts it in a difficult situation which can only be resolved by a reference, and OHIM has said publicly that it would welcome an explicit ruling from the CJEU to resolve this difference between different EU jurisdictions. I agree since the existence of the dispute at IPO level, and possibly above, adds an extra level of uncertainty on the scope of a registration in trade mark clearances searches, infringement matters, oppositions and cancellation actions.
I too prefer the "means-what-it-says" practice, but more to the point I believe that legal analysis makes “class-heading-covers-all” untenable. My views on the legalities can be found in the CIPA Journal of April 2009, Volume 38, Number 4 pp 257-260. I have yet to see a reasoned legal analysis supporting the contrary view, but should welcome one, not least for balance in consideration of the further stages of the CJEU reference process.
The keys to resolution of the dispute are, I suspect, in two terms, (i) the prospect of a registered list of goods being construed to provide “invisible protection” for the many items in
every class which are not naturally described by any of the terms used in the class
heading, and (ii) “directly related”, from OHIM Presidential Communication 4/03 as regards goods. On “directly-related” 4/03 says (the support for point 10 above):
“"computer software" is classified in cl. 9 because of the presence of the general indication "data processing equipment and computers" in that class, because computer software is seen as directly related to the corresponding apparatus or equipment. Therefore, when a specification contains "data processing equipment and computers" as a general indication, this will be considered as embracing also computer software.” [Emphasis added]We know about identical and similar goods, and the increased burden of proof when similarity is argued, but the oft-applied effect of 4/03, eg in CTM oppositions, is that all of these extra goods in any class are directly related to those in their class heading wording, and so are identical. Can then goods be similar and directly related and, if so, does the likelihood of confusion proviso apply or not?
The idea of “invisible protection” comes from a UK IPO opposition decision in case O-026-06 concerning COVENT GARDEN, asking in essence whether a registered list of goods should, through the hidden code of corresponding to (part of) an IC class heading, have a greater goods scope than a registration not so corresponding?
All contributions to this debate, whether they are in agreement with this piece or not, are welcome. Please post a comment below and/or email Richard Ashmead.