“So Precisely What Will You Use Your Trade Mark for?” Bad Faith and Clarity in Trade Mark Specifications

Abstract

The requirements for the specification setting out the list of goods and services covered by a trade mark registration are rarely considered in the literature or by the courts. The reference in C-371/18 Sky v. Skykick will enable to Court of Justice to consider two critical issues connected with the precision of the specification: whether clarity of the specification is a substantive requirement, and whether it is necessary to intend to use the mark on the date of filing. It is argued here that the Court of Justice should find that clarity is a substantive requirement for the registration of a trade mark and confirm that a lack of intention to use a mark is a type of bad faith which can lead to (partial) revocation of the trade mark.

1 Introduction

Often people talk of how trade marks work, what they do and why. But here we are looking at what happens when trade marks go wrong and the system does not work properly. Trade marks are different from other intellectual property rights. Whilst patents, copyright, designs and even plant varieties require a degree of creativity – be it originality, inventiveness or similar – trade marks require nothing more than selection. This is not to suggest that the creation of brand identity cannot be creative – and expensive – but in contrast to other rights it is not required. Both the mundane and the magnificent can be protected by trade mark law. We are not looking at even the creative bit of trade marks, but at its most functional that is the specification of goods and services. It is the least glamorous part of a trade mark application which has been considered little by academics or, really, even by the courts. Yet its role is vital and central to the trade mark property.

This is because a trade mark must be registered in relation to specified goods and services to determine its scope of protection.¹ It mandates when a trade mark is infringed – defining and placing a boundary around the exclusive rights. Indeed, the specification can determine when a person commits a criminal offence and when she does not.² Yet the drafting of a trade mark specification has traditionally expected little from the applicant leaving it to the public to work out what is infringing or not infringing, that is what is illegal or lawful. Leaving a difficult question as to how precise the specification must be.

2 The Need for the Specification

2.1 Broad and Vague Claiming

An applicant for a trade mark must specify the goods and services in respect of which she is applying. As there is no requirement to have actually used the mark at the time of application³ the applicant can apply for the mark to cover any and every good and service. Indeed, as developing a trade mark can be an expensive process with market research, clearance searches and so forth there is an incentive to broadly claim (maybe, “covetously” claim) to protect as many different goods and services as possible.

2.1.1 What Is Broad?

The reasons why a trade mark specification might cover more goods or services than the applicant uses or imminently intends to use are at least threefold. First, brand extension. The applicant might have plans, whether extensive, tentative or even speculative, to expand the range of goods or services sold under the mark. Secondly, the drafter accidentally, or through absence of specific instructions from the client, claims more than is used or intended to be used. Thirdly, there is the situation where the applicant includes goods or services which are similar to those where there is use (or intention to use) to broaden the scope of protection. For instance, there is an intention to use a mark for wine, but not spirits, yet the specification claims both,⁴ so that the trade mark is protected in relation to goods similar to wine and similar to spirits. What we are considering is which, if any, of those three types of broad claiming should attract censure and sanction.

There are two ways in which an applicant can claim broadly: (a) using broad or unclear language so that a few words cover a multitude of goods (taking account of the inclusiveness principle,⁵ so, for instance, “diary products” is identical to “milk” and “milk” is identical to “diary products”); (b) including goods in the specification which the applicant had no real intention of using. The first category at least as it applies to broad language obviously overlaps with the second. Say for example, an applicant intends to use a new trade mark for a pear cider (perry) but the specification says “alcoholic beverages”. If there is no intention to expand the product range beyond perry this would be using broad language to covers products which the applicant did not intend to use. The pertinent issue is whether this is acceptable and, if not, what should be required?

This question could be expressed in another way, how much time and effort does the applicant (and her trade mark attorney) need to invest in working out the current and future use of the mark. In patent law, the issue is clear, the patent attorney has to work out exactly what the invention is so to be able to describe it fully. Daniel Alexander QC, sitting as the Appointed Person,⁶ in HTC Corp v. One Max Ltd⁷ took a very different view in relation to trade marks:

specifications are drafted by trade mark attorneys who should not have to enquire in meticulous detail about an applicant’s precise business plans before putting forward a specification that bears a reasonable relationship to the business the applicant is already in or has some prospect of entering in the future.

Within six months of HTC in Sky plc v. Skykick UK⁸ Arnold J asked the Court of Justice whether, in effect, there should be such a duty by asking questions surrounding two related issues.⁹ First, whether it is a substantive requirement that a trade mark specification is sufficiently clear and precise (which will be called clarity)¹⁰ and where it falls short does this lead to an application being rejected or declared invalid. Secondly, whether the long-standing rule in the United Kingdom¹¹ that an applicant must have a bona fide intention to use the mark is compatible with EU law. Before addressing these questions, we will consider the costs associated with vague or broad claiming and concerns with “cluttering” the register.

2.2 The Cost of Vagueness

The view taken of overly broad specifications, whether through vagueness or claiming beyond an intention to use, determines when in a development cycle a trade mark application should be made: is it during the development of a business idea? Or is it after the applicant has determined what the business is or will be? The broader and vaguer the specification can be, the earlier it can be filed in the development cycle; whereas the more precise it has to be, the later it will have to be filed. The cost, or potential, for brand development must also be factored in, and with it the potential for expansion. The matter is so open because trade marks, in contrast to other intellectual property rights, enable the applicant to make a “free” choice as to the goods or services covered by the mark; absent earlier rights, the only restriction is one of cost.¹² Whereas a patent will only be granted over the inventor’s own invention, copyright only protects the author’s work and design rights cover only the designer’s design. While there are issues with the scope of protection for each of these other rights being unclear,¹³ it is always bound and tied to a thing created by the inventor, author or designer, creating a natural maximum boundary that can be claimed.

A broad or vague specification presents real and practical costs, and conversely, greater precision reduces that cost. In purely economic terms allowing marks to be registered which are not being used creates rent seeking,¹⁴ that is using the trade mark system to increase the applicant’s wealth without creating wealth for others. On the basis that, the trade mark system – when used – reduces search costs for consumers¹⁵ and when a mark is not used there is no reduction in search costs. While this provides a more formal model in terms of search costs, there are also costs faced by users of the system itself.

A sensible trade mark applicant, and more broadly, brand developer, will undertake a trade mark clearance search before an application is filed (or investing in a brand). This involves looking at the trade marks currently on the register¹⁶ and while the representation of the mark being similar is often the primary concern, a similar mark is only a problem where the goods or services covered by the earlier mark are sufficiently similar to the new product. The more uncertain the specification of the earlier mark, the more steps that need to be taken with the application either through complex disclaimers, carve outs or, ultimately, moving on to select a new mark. Greater precision in the specification, either from its wording or requirements of intent to use, reduce these clearance costs.

Conversely, the more precise the specification, the greater the costs associated with drafting the application both in time and, usually, in terms of professional fees. Over precision (or no current intention to use) can also lead to missing out. This could lead to additional applications being made for these missed goods or services and the attendant costs; or the mark could be taken by another trader requiring the original owner to bring potentially expensive opposition proceedings; or, if the other goods are not similar enough and the mark has been taken, it might require the development of totally new brand altogether. Ultimately, a stricter system places more costs on the applicant/proprietor whereas a less strict system places greater cost on other traders and, more distantly, on consumers. Accordingly, the level of precision required depends on a policy decision as to the allocation of these cost.

2.3 A Cluttered Register

Once registered, there is no institutional control of the register to remove unused marks, rather the approach is “confidently left to competitors”¹⁷ to clear out unused marks. So the register must contain a substantial number of unused marks.¹⁸ Indeed, the previous practice at OHIM (now EUIPO) actively encouraged overly broad claiming in that an applicant, upon paying the basic fee, could claim goods or services in three classes without additional cost (buy one – get two free). While the practice has changed so the basic fee covers only one class, there is a still the possibility of over claiming within a single class for no additional cost (our spirits and wine example for instance). This generous approach leads to unused marks being left “cluttering” up the register which, in principle, precludes others from using the “deadwood”.¹⁹ There is now empirical evidence this actually happens as it has been shown, for instance, that specifications for EU trade marks are generally broader than for their United States sibling registration.²⁰

Whether this “clutter” hampers third parties is not universally agreed.²¹ There have, however, been some firm statements from the Court of Justice on the matter. In C-149/11 Leno Marken,²² Advocate-General Sharpston stated in relation to the requirement of use:

It aims to ensure that the register does not contain marks that obstruct, rather than improve, competition in the market place because they limit the range of signs which can be registered as trade marks by others, serve no commercial purpose and do not actually help to distinguish between goods or services in the relevant market and associate them with the proprietor of the mark.

The Court itself adopted this reasoning²³ and went on to suggest that marks remaining on the register which are unused “risks restricting the free movement of goods and services” – one of the gravest of all sins. These comments refer to the purpose of revocation after five years of non-use, but the sentiments expressed apply to the requirement of use throughout the life cycle of a mark. Consequently, concerns about the cluttering of the register further lend support to a need for greater precision.

3 Two Ways to Precision

As already mentioned, the reference in Skykick²⁴ asks questions about two different ways a trade mark applicant can be required to trim the fat from her application. In this section, the discussion turns to the strength of the legal arguments behind the each of those propositions, first whether there is a requirement of clarity and, secondly, whether an applicant must have an intention to use the mark.

3.1 A Requirement of Clarity

In 2003 OHIM issued its infamous statement of practice²⁵ which read “It constitutes a proper specification of goods and services in a CTM application if the general indications or the whole class headings provided for in the Nice Classification are used. … The Office does not object to the use of any of the general indications and class headings as being too vague or indefinite”.²⁶ Critically, it went further and stated that “the use of a general indication found in a class heading will embrace all the individual goods or services falling under that general indication”.²⁷ This facilitated a practice whereby a person could use the class heading for Class 33 “Alcoholic beverages (except beer)” and this would be treated as covering wine, cider, gin and whisky and so forth. This is the case, notwithstanding, the fact that wine and spirits, and spirits and cider are treated as dissimilar goods.²⁸

The appropriateness of this practice was considered in a test case C-307/10 Chartered Institute of Patent Attorneys (CIPA) v. Registrar of Trade Marks (IP TRANSLATOR)²⁹ when the Appointed Person in the United Kingdom referred a question to the Court of Justice.³⁰ It held that the words in a specification would only have their natural meaning and could not be read to incorporate every good or service in the class.³¹ Critical for our discussion is the answer to the other questions regarding the existence of a requirement of clarity to which the discussion turns in a moment. Subsequently, the EUIPO, in conjunction with the national trade mark registries set out five terms which lack clarity: machines, machine tools, repair installation services, treatment of materials and personal and social services rendered by others to meet the needs of the individual.³² Two further examples which have been flagged by the High Court are computer software³³ and financial services.³⁴ Whether these words, or any other words, lack clarity will not be considered here, rather we will examine what happens when a term is imprecise or lacks clarity.³⁵

3.2 Is Lack of Clarity a Ground of Invalidity?

The argument that a lack of clarity is a ground for invalidity was originally considered, and rejected, by Arnold J in Stichting BDO v. BDO Unibank Inc.³⁶ Subsequently, in Total Ltd v. YouView TV³⁷ Sales J set out why the matter was arguable. Most recently, as already mentioned, Arnold J concluded in Skykick that to resolve the issue it required a reference to the Court of Justice. The starting point for considering whether clarity is a substantive requirement is Art. 4 of the Directive (2015/2436/EU)³⁸:

Article 4

Absolute grounds for refusal or invalidity

1. The following shall not be registered or, if registered, shall be liable to be declared invalid:

(a):

signs which cannot constitute a trade mark;

The requirements of a trade mark are set out in Art. 3³⁹:

Article 3

Signs of which a trade mark may consist

A trade mark may consist of any signs, in particular words, including personal names, or designs, letters, numerals, colours, the shape of goods or of the packaging of goods, or sounds, provided that such signs are capable of:

(a):

distinguishing the goods or services of one undertaking from those of other undertakings; and

(b):

being represented on the register in a manner which enables the competent authorities and the public to determine the clear and precise subject matter of the protection afforded to its proprietor

The key question, therefore, is whether the references to declarations of invalidity which apply to “signs” in Art. 4 are meant to be restricted to the representation requirement in Art. 3 only, or whether it means the representation requirement as it applies to particular goods and services. The first time particularity of services was considered by the Court of Justice, and so the starting point, is C-418/02 Praktiker Bau-Heimwerkermärkte⁴⁰ where the court held that the nature and content of the (goods and) services eligible for protection are substantive conditions for obtaining trade marks⁴¹ so as to ensure there is an identical meaning for each good or service in every Member State.⁴² In simple terms, this suggests the scope of protection conferred by a particular word used in a specification is a substantive matter. The next, and pivotal stage, requires us to return to IP TRANSLATOR.

In IP TRANSLATOR⁴³ the Court of Justice pointed out that the clarity and precision of the specification of goods and services is significant in respect of numerous aspects of trade mark law,⁴⁴ in particular the absolute and relative grounds of refusal depend on the goods and services.⁴⁵ It is unarguable, for instance, that whether a sign is descriptive, deceptive or generic is dependent on the goods to which it is applied. For example, the mark APPLE is descriptive for apples, deceptive for pears but distinctive for computer equipment.

The Court opined, therefore, that the more specific the description of goods and services in the specification, the easier it is to apply the law.⁴⁶ This limited bureaucratic reasoning was dwarfed by its central reasoning: the goods and services for which the protection of the trade mark is sought must be identified with sufficient clarity and precision to know the extent of the protection.⁴⁷ The key part of the judgment is as follows:

… the entry of the mark in a public register has the aim of making it accessible to the competent authorities and to the public, particularly to economic operators … . On the one hand, the competent authorities must know with clarity and precision the nature of the signs of which a mark consists in order to be able to fulfil their obligations in relation to the prior examination of applications for registration and the publication and maintenance of an appropriate and precise register of trade marks. … On the other hand, economic operators must be able to acquaint themselves, with clarity and precision, with registrations or for registration made by their actual or potential competitors, and thus to obtain relevant information about the rights of third parties. Accordingly, [the Trade Mark Directive] requires the goods and services for which the protection of the trade mark is sought to be identified by the applicant with sufficient clarity and precision to enable the competent authorities and economic operators, on that basis alone, to determine the extent of the protection sought.⁴⁸

This final part of this extract, the rule as it were, was incorporated in the Recast Trade Marks Directive and Regulation as part of the recent package of reforms.⁴⁹ In YouView, Sales J highlighted a number of points raised in IP TRANSLATOR which suggest that clarity is a substantive requirement. First, the Court of Justice highlighted that the provision of a specification is a “substantive condition”.⁵⁰ Secondly, the natural place to begin the assessment of clarity is in the definition of a trade mark in the Directive.⁵¹ Finally, he highlighted the analogies drawn by the Court of Justice with C-273/00 Sieckmann⁵² where it was held that the representation of the sign must be identified with sufficient clarity and precision, and the failure to provide such a representation leads to invalidity. Accordingly, clarity, he implies, has to be a substantive requirement.

The counter-argument, originally put forward by Arnold J in Stichting, is predicated on the grounds of invalidity being those expressly set out in the EUTM Regulation⁵³ (or Directive). The issue is whether the definition of a trade mark in Art. 3 of the Directive and a construction of the words “goods and services” is a reference to any goods and services or the goods and services in respect of which the registration was sought.⁵⁴ So, using the same example as above, as the sign APPLE can function as a trade mark for computer equipment it does not matter that it cannot do so for fruit – even if the application relates to fruit. This interpretation of the Directive, Arnold J suggested, meant that clarity need only be considered during examination and, later, if the specification were amended.⁵⁵

In any event, once Sales J had held that the matter was not acte clair in YouView even though it was not determinative in that case, when the question became determinative in Skykick, Arnold J decided to make a reference to the Court of Justice to resolve the question of whether the requirements of IP TRANSLATOR apply during examination or afterwards as well as a ground of invalidity.⁵⁶

3.3 Temporally Speaking

Whether the rule in IP TRANSLATOR applies to a particular application has now been addressed by the Court of Justice twice. It could be argued that these cases provide a complete answer to the first issue in SkyKick, but as will be explained both of these cases are based on a misconception concerning legal certainty and so, it is suggested, it would be undesirable for the Court to take the easy way out.

In the first case, C-577/14 Brandconcern (LAMBRETTA),⁵⁷ it was held that part of the ruling in IP TRANSLATOR (that relating to class headings) did not apply to trade marks already registered, rather it only applied to pending trade mark applications⁵⁸ and, more recently, in a second case, C-501/15 Cactus,⁵⁹ it was confirmed that the rules in Prakktiker Bau and IP TRANSLATOR only concern applications and not registered trade marks.⁶⁰ There are two ways these cases could be viewed. One way would be that the clarity requirements (including use of class headings) only ever apply to applications, and the other is that those rules only ever apply to applications pending at, or filed after, the date of the respective judgments. It is necessary therefore to unpick those decisions a little.

In LAMBRETTA the Court adopted the views of Advocate-General Campos Sánchez-Bordona.⁶¹ He opined that IP TRANSLATOR was intended to put an obligation on the Trade Mark Office to ensure that applications comply with requirements of clarity and precision.⁶² Extraordinarily, he saw the obligation as only indirectly applying to the applicant. He went on to say that the Office did not have a role after registration⁶³ and, in particular, it was not possible to amend the specification thereafter.⁶⁴ This led him to conclude that IP TRANSLATOR did not apply to existing registered trade marks.

While it is possible to amend the specification of goods and services before grant, this only tells half the story, as it is only possible to restrict the goods and services covered by the application (narrow the specification).⁶⁵ This can be done by a re-write amendment as well as by simple deletion (e.g. “diary products” could re-written so as to be restricted to “milk” or the words could be simply struck out). Yet where the description of a particular good is lacking precision, it is not possible for an applicant to remedy the situation unless the words used are indisputably narrower than the original words. Should a word be so imprecise that its boundaries cannot be found, then a pre-grant amendment might not be possible at all. Critically, it is also possible for a proprietor, after registration, to surrender the mark in relation to some goods or services⁶⁶ or for a court (or trade mark office) to declare it invalid in relation to only some of the goods or services.⁶⁷ Both of which also restrict the specification.

Although rarely discussed in the trade mark context,⁶⁸ the applicant cannot broaden her specification of goods and services after the date of filing.⁶⁹ So once an application has been filed, right through to the end of the registered mark’s life, the scope of the protection can never be any broader than it was in the original specification. The distinction put forward by the Advocate-General would make far more sense if the line he drew was the date of filing of the application and not the date of grant; as there no substantive difference in terms of modifying the scope of protection before or after grant, only the procedures are different.

The Advocate-General’s reasoning, therefore, does not set out the full picture in at least two respects. First, he unnecessarily places the obligation as to the precision of the list of goods and services on the Office and not the applicant. While there are systems of registration where objections can arise only prior to grant,⁷⁰ the obligation to satisfy that requirement is still placed on the applicant. Additionally, contrary to what the Advocate-General stated, once the trade mark application has been granted, the Office (or a court) may have to consider the specification once more if an application is made to invalidate the mark (or when deciding whether to accept a surrender). Secondly, as already mentioned, the specification of goods and services can be restricted at any time after filing, and so there is no material difference as between pending applications and registered marks.

Nevertheless, the Court of Justice held in LAMBRETTA that the class heading rule from IP TRANSLATOR applies only to pending applications so as preserve legal certainty.⁷¹ However, as has been discussed and contrary to the Advocate-General’s opinion, there is no material difference in terms of certainty before and after grant and so there is no need to apply IP TRANSLATOR to pending applications only. Indeed, it is illogical to do so.

Unfortunately, the drafting of Art. 33 of the EUTM Regulation,⁷² which essentially codified IP TRANSLATOR, also provides conflicting indicators. On the one hand, there are temporal references suggesting that the requirement of certainty only applies before grant: “protection of the trade mark is sought”⁷³; and specific provision for the Office to reject terms which are unclear or imprecise⁷⁴ (but, as explained, this could apply to the Office’s duties as they apply both before and after grant). Conversely, the requirement itself is “to enable the competent authorities and economic operators … to determine the extent of protection”. While this could be narrowly construed to mean economic operators should know the extent of protection for the purposes of opposition only (i.e. while it is still an application), this would seem unlikely, particularly as in C-420/13 Netto Marken-Discount⁷⁵ the Court of Justice held that the requirements of certainty for third parties applies to both applications and registrations.

It is unlikely the Court of Justice will depart from its approach in LAMBRETTA, but it is suggested that because the premise upon which it is built is weak, the Court should not construe it to be a straightforward rule applying IP TRANSLATOR only to applications. Such a reading would mean that a registered trade mark cannot be declared invalid on the ground that the goods and services are not precise. Rather the Court should read its earlier judgment so that it draws a “line in the sand” and applies to all applications for marks which were pending at, or filed after, the date of the judgment and continues to apply to those marks after registration (the view adopted by Arnold J in Skykick⁷⁶). This “line in the sand” approach fits with the mechanism in Art. 33(8) of the EUTM Regulation to regularise old applications which used class headings; although it does not provide a whole answer because, contrary to the normal rules, the mechanism allows for a proprietor of an existing mark to broaden the scope of protection.⁷⁷

3.4 An Existing Answer … but Is It Suitable?

It would be possible for the Court of Justice simply to adopt the General Court’s current approach to imprecise terms. Essentially, where the specification is too vague to undertake a proper comparison of trade marks, the offending words are read-out of the specification. For instance, in T-162/08 Frag Commerce Internacional,⁷⁸ the Court accepted that where the type of retail services had not been specified,⁷⁹ the Board of Appeal was entitled to not undertake a comparison in relation to those services.⁸⁰ In other words, temporarily crossing-out⁸¹ out the relevant words. The General Court has subsequently held repeatedly that there are circumstances where comparisons are not possible, such as where there were apparent typographical errors (“advertising sentence” it is supposed for “advertising services”⁸²) or where additional information would be needed to work out what was covered by the goods (such as “accessories”).⁸³ It similarly held that an applicant should not benefit from failing to set out the goods in a precise way.⁸⁴ It appears, therefore, that where the specification lacks clarity (and so is contrary to IP TRANSLATOR), the court or tribunal might in effect be able to cross-out the unclear words.

This presents two issues. The first is what to do in cases of fundamental uncertainty – that is where the words can never have a clear meaning. The second is what to do about so-called “fuzzy” edges. In cases of fundamental uncertainty, that is where the offending word is always uncertain and so would inevitably have to be “crossed-out” each time a comparison is made, the General Court’s approach, in effect, makes the trade mark (partially) invalid. The words have no legal effect whether the trade mark is declared invalid or not. The proprietor of the trade mark therefore has a worthless part of her specification and there can be no good reason not to expunge the relevant words.⁸⁵ Conversely, there are very good reasons to remove the inoperative words. Other traders, following a search of the register, would not necessarily be aware that the words are uncertain and so may give them a meaning (and as the terms are unclear, it might be a very broad meaning) leading that person to avoid using a mark she would lawfully be entitled to use. The retention of the words, therefore, allows the more unscrupulous trade mark proprietor to assert protection in relation to those words, notwithstanding that she knows those words would be made inoperative by crossing-out should the matter reach court.⁸⁶

In relation to “fuzzy” edges the question is more difficult.⁸⁷ In some cases, where it is unclear whether another good falls within the boundaries of the description in the specification, it may not be possible to work out whether that good is identical, but the later user’s goods are probably going to be similar. In such a case, a view could be taken to permit the fuzzy edges to remain and allow a similarity comparison to be undertaken (with the trade mark proprietor facing a higher hurdle, proof of likelihood of confusion, to avoid her benefiting from the imprecision). In other cases, where it is not clear whether the goods are even similar, but it is clear other goods not relevant to the current decision would be similar, it may be that crossing-out the imprecise words provides a temporary solution. However, legal certainty and the accuracy of the register would lend itself to making the result final even where there are only fuzzy edges. It is therefore submitted that if the specifications lack clarity the better solution is to find the relevant words to be invalid and to remove them from the register unless there is an acceptable narrowing amendment.

3.5 Certainty of Representation

The question of precision in a trade mark application has been explored in greater detail, both academically and by the courts, in relation to the representation of the mark than it has in relation to the specification of goods and services. By looking at these discussions it is possible to show that even if a clarity requirement were imposed for the specification it would allow some flexibility. In other words, the scope of protection granted by the representation shows certainty is a relative matter.

The representation of the trade mark defining the scope of the right, rather than recording it, arose as registration becoming a bureaucratic process.⁸⁸ As the Court of Justice held in C-273/00 Sieckmann the “representation of the trade mark must determine the precise subject matter of protection afforded”⁸⁹ and now the Directive and Regulation provide that “the sign must be represented on the registered in a manner which enables the competent authorities and the public to determine the clear and precise subject matter of the protection afforded to its proprietor”.⁹⁰ The failure to provide a proper representation is a substantive requirement and as Burrell and Handler say⁹¹:

The explanation for the success of trade mark law in enclosing the intangible also seems to lie in the extent to which the subject matter of trade mark law has been capable of being reduced to a visual representation: unlike patentable inventions, traditional trade mark subject matter, such as words, simple phrases and device marks, can be reproduced “as such”. Historically this has meant that there has not been the same sense of distance between the official documentation and the “essence” of the property.

While the property is represented on the register, contrary to the statement in Sieckmann, accurately reproducing the representation does not determine the precise subject matter of the protection afforded.⁹² Indeed, imprecision arises in at least two ways. First, it is immediately apparent that a person can infringe a trade mark if the infringing sign is identical or if it is similar⁹³ (and similarly, where goods or services are identical or similar and, in some cases without reference to the goods at all). Thus, while the register might tell us what the property “is” – it does not tell us what it can do, that is, the scope of the protection it affords.⁹⁴ The identification is therefore a useful mechanism for defining the right in terms of transactions and licensing (the “is” question), but it is only the beginning of the matter when it comes to what is similar. It becomes even more complicated as some marks which have been used in the marketplace can develop so-called “enhanced” distinctiveness,⁹⁵ which means that another mark is more likely to be confused with it or, put another way, distinctiveness broadens the “penumbra” of the protection afforded to the mark.

Secondly, it is now the case that the representation itself is no longer sacred. In C-252/12 Specsaver⁹⁶ the trade mark owner had registered a black and white image of the mark, but usually used a colour image in the marketplace. The Court of Justice took the view that it would be illogical to use the mark as registered to assess infringement. So rather than use the black and white representation,⁹⁷ the Court said the coloured mark actually used in the marketplace determined the scope of protection.⁹⁸ While this ruling may be confined to monochrome representations or maybe just the colours used, it moves away not only from the register defining the scope of protection but goes further so that what is registered does not actually define the representation of the mark.

The register does not, therefore, set the boundary of the scope of protection. One might ask, therefore, why does the representation have to be “clear and precise”? And by extension why is it being argued that the specification should also meet this requirement? If the boundaries of protection are blurred and fluid and can be influenced by use in the marketplace, cannot the requirements for specifications remain imprecise? No. What the clarity requirements are really about is relative certainty rather than absolute certainty. Even if the degree of certainty required in the specification is itself flexible and uncertain, as will be explored in the final part, greater certainty is still a desirable thing. We must now move on to the second issue which was before the Court of Justice in the Skykick reference.

4 Bad Faith

The concept of use is central to trade mark law from registration through to exploitation.⁹⁹ Normatively, requiring use of a trade mark avoids the administrative burdens involved in examining and registering the marks which are not going to be used,¹⁰⁰ and it keeps the marks free for those who will actually make use of the mark.¹⁰¹ The mechanism of revocation for non-use¹⁰² requires a potential user of the mark to wait five years before the mark becomes free. However, there has long been uncertainty over whether, under EU law, there was also another check on non-use, namely requiring an applicant to have an intention to use the mark on the date of filing.¹⁰³

This uncertainty led to Arnold J asking the Court of Justice in Skykick whether an application can be refused, or declared invalid, on the basis the application was filed without any intent to use the mark, that is, does the absence of such an intent amount to filing in “bad faith”?¹⁰⁴ In the United Kingdom, the Trade Marks Act 1994 provides that an applicant for a trade mark must state on the application form¹⁰⁵ whether she is using the mark or has a bona fide intention to use the mark.¹⁰⁶ Where an applicant makes a false declaration, the application is made in bad faith.¹⁰⁷ It might be said, therefore, that it is the falsity of the declaration, rather than the lack of an intent to use which has been used to refuse (or invalidate) applications on the grounds of bad faith in the United Kingdom.¹⁰⁸

In the absence of such a declaration, such as in relation to an application for an EUTM, the British courts have been more circumspect; however, as the domestic approach will be overtaken by the decision of the Court of Justice, it need not be explored further. The position in relation to EUTMs originates with a decision of the OHIM First Cancellation Division in TRILLIUM,¹⁰⁹ where the absence of a declaration of intent to use was noted, and the decision went on to hold¹¹⁰:

In general, and as a matter of principle, it is entirely left to the applicant to file a list of goods and services as long as he sees fit, i.e. a list exceeding his actual scope of business activity, and try later to expand his activities in order to be able to show genuine use of his CTM or face revocation under Article 50(1)(a) CTMR [now EUTMR, art 58(1)(a)] and other sanctions, respectively. It is exactly this “liberal” concept which underlies Articles 15 and 50 CTMR [EUTMR, arts 18 and 58] because otherwise a grace period of five years would make little sense, if any ….There may be cases where an applicant files a list of goods and services where all or part of it does not have the slightest connection with his actual economic activity, and where it might even appear unimaginable that said applicant would ever be able to expand. If in such case the holder of the CTM immediately took action, based on “remote” goods or services, against third parties, it might be worth considering Article 51(1)(b) CTMR [EUTMR, art 58(1)(b)]. But this can be left undecided because it is not the case in the present proceedings before us.

4.1 Lindt and All That

This is the only time any EU tribunal has expressly considered whether the absence of a bona fide intention to use a mark amounts to bad faith.¹¹¹ And it firmly came down against. However, the leading case on bad faith, C-529/07 Lindt,¹¹² is sufficiently ambiguous to make this answer decidedly suspect. In Lindt, the Commission submitted to the Court that bad faith should be restricted to registrations where the applicant had no intention of using the mark,¹¹³ and that bad faith was a corollary to revocation on the grounds of non-use. While the Advocate-General did not accept the Commission’s restricted position, he was convinced that filing without a bona fide intention to use would amount to bad faith.¹¹⁴ Unfortunately, the Court of Justice did not expressly address intention to use, save maybe by implication.¹¹⁵

What the Court did say in Lindt was that in determining whether an application was made in bad faith, certain factors could be taken into account: (a) the fact that the applicant knows or must know that a third party is using, in at least one Member State, an identical or similar sign for an identical or similar product capable of being confused with the sign for which registration is sought; (b) the applicant’s intention to prevent that third party from continuing to use such a sign¹¹⁶; and (c) the degree of legal protection enjoyed by the third party’s sign and by the sign for which registration is sought.¹¹⁷

Importantly, the Court’s finding was very narrow. It held that the determination of whether an application was made in bad faith must be made at the date of filing,¹¹⁸ which would fit in with a requirement for an intention to use.¹¹⁹ It also stated that bad faith requires proof of an intention to prevent a third party continuing to use the sign. The finding in Lindt is thus confined to steps taken to prevent an existing user¹²⁰ continuing what she has done before. In its reasoning, however, the Court said that that “the intention to prevent a third party from marketing a product may … be an element of bad faith”¹²¹ and subsequent conduct may be evidence that the intention at the time of filing was to prevent a third party entering the market.¹²² Maybe significantly, the Court did not refer to the product entering the market under a particular trade mark, but entering the market at all.

It might be assumed that any reference to marketing a product meant doing so under a particular mark or sign,¹²³ particularly as the answer to the referred question was in terms of preventing the use of the sign. This is certainly a more coherent view. Nevertheless, the peculiar facts of Lindt could enable the Court of Justice to read it very narrowly. This is because Lindt was about a shape mark – bunny shapes – being used in relation to chocolate. Accordingly, a registration of the bunny shape as a trade mark might have precluded the sale by other traders of any bunny shaped chocolate (a product) which the Court saw as problematic.¹²⁴ Different considerations might apply where the product could be sold by another trader albeit under a different mark.¹²⁵ Indeed, many traders already have to localise their trade marks for cultural or linguistic reasons, and it may be that the Court of Justice considers that modifying the mark to enter a market due to earlier rights is no different.

Finally, there is a somewhat perplexing suggestion in Lindt that when considering whether a mark was applied for in bad faith, the reputation the applicant enjoys in her mark might justify an interest in ensuring a wider legal protection of the sign. A broad reading of this suggestion would allow the proprietors of marks with a reputation to seek wider protection than those without. Applying this to the specification of goods would, apparently, endorse defensive marks of some sort (which as we will see cannot be right). It is probable, however, that a narrower meaning was intended by the Court enabling an applicant with existing unregistered rights to register those rights to strengthen any claim against a new market entrant.¹²⁶

While the Court of Justice considered bad faith once more in C-320/12 Malaysia Dairy Industries,¹²⁷ little of its reasoning adds much to what was said in Lindt. Nonetheless, bad faith has been considered by the Court again in a different context. In C-569/08 Internetportal und Marketing GmbH v. Schlicht¹²⁸ it was considered in relation to domain name registration. In contrast to trade marks, the relevant regulation¹²⁹ specifically provides for certain conduct to be bad faith,¹³⁰ and the Court of Justice was asked whether this list was exhaustive. It held it was not.¹³¹ Once this was established, the Court made reference to Lindt to state the assessment of bad faith must involve an overall assessment of the relevant factors.¹³² Critically, the Court concluded that registering a mark without the intention of using it other than to support an application for an “.eu” domain name can amount to bad faith (in relation to the domain name registration).¹³³ This finding certainly lends support (and little more) to the proposition that obtaining a mark with no intention to use can be bad faith. But for the references to Lindt it might be argued that Internetportal has no relevance to trade mark applications at all as it is an entirely different regime. In any event, the relevant prohibition has two elements: (a) filing without an intent to use as a mark; and (b) the sole purpose of the filing is to obtain a domain name registration. Accordingly, the judgment cannot be said to be authority for (a) in the absence of (b); nevertheless, it is at least an indication of the Court of Justice’s thinking. Finally, the actual finding of bad faith rested on numerous other factors as well, and so it may be that the absence of the intention to use was just one factor in a wider analysis.

4.2 The View of the General Court

The issue of intent to use has been mentioned, albeit obliquely, by the General Court on a handful of occasions. In T-507/08 Psytech International¹³⁴ the General Court was asked to consider whether the application should be rejected as being made in bad faith because the number of goods and services covered by the application was too large and the applicant had no intention to use the mark in relation to the entire list of goods or services covered. The court found that there was no rule which meant that by reason the number of goods or services covered by the application alone could mean the application was made in bad faith.¹³⁵ This must be right. The number of goods or services cannot, in itself, suggest bad faith. Some applicants may use the mark in relation to hundreds of different goods and services¹³⁶ and it must be proper for them to obtain a registration covering such use.

Subsequently, in T-33/11 Peeters Landbouwmachines,¹³⁷ the General Court indicated that a reason the particular application was not made in bad faith was the applicant’s intention to use the mark; or more precisely it had not been proved there was no such intention (suggesting, the absence of an intention might be bad faith).¹³⁸ In T-136/11 Pelicantravel.com¹³⁹ the General Court implicitly indicated that if evidence is provided that an applicant did not intend to use a mark at the time of filing, then this might amount to bad faith.¹⁴⁰ In addition to these oblique general statements of the law there are instances where it has been held that applying for a mark with no intention to use the mark and an intention to block a third party from entering the market amounts to bad faith (essentially following Lindt very closely).¹⁴¹ The conclusion Arnold J takes from these cases in Skykick was that the commercial logic for making the application is relevant for determining bad faith.¹⁴²

It can be seen from the decisions of the Court of Justice and General Court that there might be a requirement for an intention to use. Importantly, however, there is more than enough room when the Court of Justice considers the question (whether in Skykick or otherwise) to come to the opposite conclusion. It is therefore worth considering the legislative history and how the Court of Justice has dealt with the related issue of defensive marks. Both of these lend support to there being a requirement of a bona fide intention to use.

4.3 Legislative History

The prohibition on filings in bad faith in the Community Trade Mark Regulation¹⁴³ originated with a proposal from the German¹⁴⁴ delegation in October 1984.¹⁴⁵ It proposed that the applicant having a bona fide intention of using the trade mark should be a condition of registration.¹⁴⁶ This would, accordingly, prevent trafficking in marks and the register becoming clogged up.¹⁴⁷ The German proposal was accepted by the Working Party in April 1985, which took the view that a lack of bona fide intention to use the mark should become a ground of invalidity.¹⁴⁸ The Working Party, however, acknowledged the difficulties with proving a lack of the relevant intention and the majority took the view that intention should not be checked by the Office prior to registration (Germany dissenting and the UK and Ireland having an “open mind”).

The requirement to have a bona fide intention to use was therefore included¹⁴⁹ in a draft¹⁵⁰ of Art. 41(1)(b).¹⁵¹ By the 44th meeting of the Working Group in November 1986 the German and Dutch delegations had agreed to review the wording of the provision.¹⁵² The proposal subsequently put forward by the Germans¹⁵³ reflects the final wording¹⁵⁴ of the Regulation and so just refers to “bad faith”. Importantly, the Danish delegation took the view that the provision should be restricted to “where it becomes clear from the behaviour of the proprietor … that he did not, when filing the application, have the bona fide intention of using the trade mark”¹⁵⁵ and said there should be a statement as to what “bad faith” meant.¹⁵⁶ The Dane’s subsequent statement defining bad faith¹⁵⁷ included a lack of any intention to use the mark. Eventually, however, the Working Group took the view that there was no need for this clarification¹⁵⁸ and raised various unrecorded difficulties with the definition. One expressed difficulty was that it was not necessary due to the (WIPO) Explanatory Memorandum to Art. 6^bis of the Paris Convention. This statement clearly relates to concerns regarding parallel rights abroad, rather than bona fide intention to use. The final text was therefore adopted and no additional statements were made.¹⁵⁹

Tsoutsanis takes the view that the replacement of an explicit requirement to have a bona fide intention to use a mark with simply “bad faith” was done to exclude the requirement of intention to use from the Regulation (and Directive).¹⁶⁰ An alternative, and it is suggested better view, is that the replacement of the explicit requirement with the more general “bad faith” was done to broaden the scope of the provision, in that it was believed it encompassed a bona fide intention to use and other types of bad faith. There are at least five reasons why this view is preferable.

First, the Germans proposed the original requirement to have a bona fide intention to use and subsequently the final “bad faith” text. Tsoutsanis puts this down to a surprising change of mind¹⁶¹ by the Germans. While such a change of mind is possible, there is nothing in the travaux préparatoires¹⁶² which explains why the German’s chose new language. It is reasonable to assume that they did not have a change of mind, but wished to broaden the range of improper conduct. Secondly, there is nothing in any of the travaux, after the introduction of a lack of bona fide intention to use, indicating that the Working Group no longer believed it should be a ground of invalidity. Such an interpretation would require the reader to assume that a broader term necessarily excludes a potentially narrower one. Thirdly, the UK government, who would have been represented at all relevant meetings and did not dissent from the move to bad faith, must have thought that it included lack of an intention to use (otherwise they would be legislating contrary to the Directive).¹⁶³ Fourthly, if the Germans were proposing to exclude lack of intention to use from the scope of bad faith it would involve a departure from a rule of the then current German trade mark law.¹⁶⁴ While possible, this would be very uncommon in international negotiations.¹⁶⁵ Finally, the Danish government must have thought lack of intention to use was included as a type of bad faith because the grounds it put forward for clarification included it. We know difficulties were raised with the proposed clarificatory document, but there is nothing saying these were connected to bona fide intention to use and the reference to Art. 6^bis suggests they were more likely concerned with parallel rights.

It appears, therefore, that the drafters of the Directive and Regulation perceived bad faith as including (but not restricted to) a lack of bona fide intention to use the mark. Indeed, when over 20 years later the Max Plank Institute reviewed trade mark law as part of the lead up to the recent reforms, while it was acknowledged that there was no clear authority, it still put forward the view that filing a trade mark without intent to use was an act of bad faith.¹⁶⁶ In short, the history of both the original Directive, and in its recast form, there is strong support for the proposition that a lack of intention to use is an example of bad faith. Further support comes from the stance taken by the Court of Justice in relation to defensive marks.

4.4 Defensive Marks

Before the harmonisation of European trade mark law some Member States allowed the registration of so-called reserve marks.¹⁶⁷ These largely arose in systems where there was no requirement to use the mark for it to remain on the register.¹⁶⁸ For example, in Germany, a person could register a trade mark to keep it for later provided there was an intention to use the mark at the time of registration.¹⁶⁹ The number of such marks, and length of time of non-use permitted, depended on what was justified by the industry in question.¹⁷⁰ Eventually, the requirement of intention to use was modified by the courts so that a long period of non-use could be used as evidence there was no such intention to use at the date of filing.¹⁷¹

A defensive mark was a different matter. These marks were not being saved for later, but were registered (without intention to use) to shadow used marks so as to broaden the overall scope of protection.¹⁷² The origin of such marks was usually an unduly formalistic approach to the trade mark comparison.¹⁷³ While allowing reserve marks, the German courts were more sceptical regarding defensive marks,¹⁷⁴ eventually holding that defensive marks could only be infringed where: (a) the defensive mark was similar to the main mark; and (b) the alleged infringing mark must be similar to the main mark as well.¹⁷⁵ Thus, confusing similarity is not required between the main and infringing mark, only similarity, with the confusion being established with the defensive mark.

While defensive marks were largely concerned with registering signs which were slightly different from the mark being used, the same principles equally apply to the specification being broadened to cover goods or services very similar to those in respect of which the mark is being used. In any event, in C-234/06 Il Ponte Finanziaria SpA v. OHIM¹⁷⁶ the Court of Justice took the view that “defensive” trade marks, which were still recognised in Italy,¹⁷⁷ had no basis under the EU trade mark regime. It was therefore necessary to substantiate use in the normal way. This was confirmed in relation to national marks in C-553/11 Bernard Rintisch v. Klaus Eder.¹⁷⁸ This means that use as a defensive mark is not genuine use and so, at the very least, after five years the mark can be revoked. The judgments did not go further and address whether different rules apply in relation to defensive marks before that period has elapsed, but the court clearly considered defensive marks to be inappropriate within the EU system.

4.5 Excluding the Brand Creator

In addition to the arguments against permitting marks to be filed without an intention to use, there is a counter-argument for allowing, in some circumstances, a broader specification in the absence of a current intention to use. It would be predicated on the role of consumers in developing the brand, and potentially, the range of goods and services. In this context it must be remembered that a trade mark is little more than part of a brand.¹⁷⁹

The determination of what a brand “is” has become more difficult and descriptions have ranged from it being a “comprehensive concept” accompanying a range of “associated meanings”¹⁸⁰ to being described as a “gut feeling” about a good, service or company.¹⁸¹ Indeed, this approach to branding has led to the trade mark itself becoming the valuable commodity,¹⁸² so that “the identifier is itself the product”.¹⁸³ Critically for our purposes, there is now an acknowledgment of what has been called the “Brand Flip”¹⁸⁴ that is the role that consumers play¹⁸⁵ in the creation of the brand around new products,¹⁸⁶ or more pertinently expanding the brand to cover new goods or, more commonly, services. This acknowledgment is part of the increasing exploration of the role consumers play in developing brands – some brands become the product.¹⁸⁷

It might be unusual for most brands to have product ranges expanded by consumers. Yet there are clearly instances where the customer provides services which the supplier of the base product did not envisage.¹⁸⁸ A common instance might be a user group, which provides technical or customer support for a product using the trade mark. But there are also those who adapt products, such as computer games, which are the taken up and sold by the brand owner. Once it is acknowledged that the consumer, or more usually groups of consumers, can be the innovators and developers of the brand itself¹⁸⁹ the question arises as to what effect this has on the trade mark itself. Trade mark law has struggled with this broader communicative function of brands¹⁹⁰ as well as the role played by the consumers in broadening use in trade mark law. Indeed, there is clearly a much broader question than can be considered here, but if a brand owner is aware that consumer innovation is possible and, maybe even, seeks to direct it,¹⁹¹ any new good or service provided by a consumer¹⁹² is, probably, impliedly licensed by the trade mark owner and so counts as her use. Accordingly, going back a stage, can an intention to use be based on consumer development, and if it can, should arguments based on bad faith arising from lack of an intention to use take this into account?

One formalistic answer would be Art. 5(1) and (2) of the Directive should not be used to protect brands (only trade marks), particularly when a brand is still being developed. While the Directive allows for protection of broader brand rights, or merchandising rights, under Art. 5(3) this includes various checks and balances – due cause, damage to distinctiveness or repute or unfair advantage. To allow trade mark owners to obtain brand-like protection, including potential expansion, by claiming goods where there is currently no intention to use undermines the balance that was sought to be achieved. A more nuanced approach, addressing the role of brand creators is to assess whether at the time of filing the applicant had a reasonable belief that there will be an expansion instigated by consumers. Only where this is the case can there be said to have been an intention to use. This might make the enquiry more complex or it may be that it even negates the enquiry in some cases, but it should not affect the underlying arguments as to whether a filing is in bad faith or not.

5 Consequences

These two issues – clarity and intention to use – have engaged with a fundamental yet simple question, that is whether a trade mark applicant should already know, at the time of filing the application, how the mark will be used throughout its life. An intention to use requirement enables a brand to be developed before it is used. It avoids wasted investment costs being incurred before the launch of a product simply because someone else gets there first. Permitting broad and vague claiming, and having no current intention to use, reduces the cost of obtaining trade marks for an applicant. Nevertheless, even if a brand owner needs time to work out the exact parameters of the use that will be made of a trade mark, the law does not merely require use within five years or the mark is lost. It requires a third party to spend time and money seeking to revoke a mark for non-use. Without such action the mark remains valid and on the register and able to scare off other traders without the resources or inclination to undertake extensive searches to determine whether use has been made of the mark; such parties simply move on to another mark. Problematically, the system itself accepts the difficulty a third party has in proving use by reversing the burden of proof¹⁹³ and even then the proof of use required is quite low.¹⁹⁴ Many trade mark proprietors will, therefore, be able to keep competitors away from their brand perpetually even if there was never an intention to use a significant number of the goods or services covered by the mark. Where a burden is largely on third parties, with speculative benefit for the proprietor, it is difficult to justify imposing it.

5.1 Fence Posts or Signposts?

As the scope of protection of a trade mark includes goods or services which are similar to those included in the specification and, as discussed above, the representation on the register does not accurately represent the maximum extent of that protection, the move towards precision and narrow claiming might be thought to be an unnecessary complication. An analogy which is sometime used is that the register should provide signposts rather than fences.¹⁹⁵ Another way of expressing the matter is distinguishing between claims which set the boundaries of the subject matter (so-called peripheral claiming), compared with those which set out the central features (so-called central claiming).¹⁹⁶ The basic distinction between these two approaches can be explained by considering a field. To mark ownership of a field a person can put a fence around it or they can put a signpost in the middle saying it is her field. The usual difficulty with this analogy is that it suggests the edges of the field are always clear and can be precisely enclosed, but as we have seen, this is not true in relation to the scope of protection.

If the analogy with fence posts marking out the metes and bounds of the trade mark right is inappropriate,¹⁹⁷ so is that of signposts. Signposts point you in the direction of travel; they can mark places far off or near and they can be frequent or irregular. A more appropriate metaphor is calling a cab.¹⁹⁸ Thus, a person who gives a very detailed description of themselves tells the dispatcher they are in “London and want to go to Heathrow Airport.¹⁹⁹ The description “London” means that it would take a very very long time (and be very expensive) for the cab to collect the passenger. Should that person say they are in Westminster the time and complexity of being collected diminishes greatly (although no real cabby would bother with such vague directions). If the person says they are on Abbey Orchard Street²⁰⁰ then it might take a minute or two for the taxi driver to find her fare, but it will happen. But if the passenger says she is outside the Intellectual Property Office then they are certain of collection quickly. The start of the journey is therefore precise. Even if the end of the journey, which Heathrow terminal the passenger wants to go to, requires more information. Simply put, the less precise the collection point, the longer and more complicated it is to start the journey. So while the precision suggested here is not a full address, as it were, it would still require the applicant to be precise enough, following the metaphor, to identify the street.

It is on this basis, the provision of a known starting point, which can be used to justify a requirement for a precise representation and, in turn, a precise specification. While a representation or term in the specification may take a few steps to the left or right of this point, it enables the court and others to have a very good idea of what the rights are. The less precision in the starting point, the higher the cost for competitors to determine the scope of protection. Importantly, the absence of complete precision does not negate the advantages of more precision. This brings us to our final question: how does one determine how precise the specification must be.

5.2 Guidance for Precision

The answer to the final question should begin with the rules already in place (even if yet to be endorsed by the Court of Justice) for establishing whether the “use” requirement²⁰¹ has been satisfied in relation to some, but not all, goods or services. Why this is so important is it explains how the bits of a trade mark which have not been used are carved out from those bits which have. It spells out the level of specificity that ideally would be required for the mark to be valid and, in any event, how precise it needs to be after five years.

In T-126/03 Reckitt Benckiser (ALADIN),²⁰² the Court of First Instance (General Court) set out the basic approach to partial use of a good or service. The Court identified the two forces pulling in the scope of use. Where a trade mark has been used on only some goods or services then it cannot be considered to be used across all the goods or services for which it has been registered.²⁰³ The approach the Court took was to consider particular goods in terms of categories and sub-categories. Use in a category is only sufficient to maintain the entire category where it cannot be divided into sufficiently distinct sub-categories²⁰⁴ other than in an arbitrary manner.²⁰⁵ Thus, what needs to be done is determine whether or not a category includes independent sub-categories. The Court must then decide whether use has been proved only in respect of that sub-category of goods and services or, on the other hand, where no sub-category can be identified, use can be established across the whole category.²⁰⁶ So for instance, “AC adapters” is a sub-category and so does not sustain use across “power distribution or control machines and apparatus”²⁰⁷; “therapeutic preparations for respiratory illnesses” is a sub-category of “pharmaceutical and sanitary preparations; plasters”²⁰⁸ whereas “polish for metals”²⁰⁹ was an acceptable sub-category.

5.3 Conclusion … and a Proposed Answer for the Court of Justice

It is suggested, therefore, that intention to use should mirror revocation on the grounds of non-use.²¹⁰ So, where at the date of filing, the intended use of a mark would not be sufficient to survive a revocation action five years later, a court or tribunal should conclude that the application was made in bad faith. Nothing would change during examination, the Trade Mark Office would still have no obligation to determine whether there was an intention to use on filing,²¹¹ rather it would be left to third parties to apply to invalidate the mark after grant.²¹² In practice this would (usually) mean that only where a third party actually wants to use the unused mark would invalidity be sought; otherwise, the unused marks (including those with no intention to ever use) would simply remain on the register (as they do now).

As intention to use should, so far as possible,²¹³ mirror revocation for non-use, where there was no intention to use some of the goods and services, then the registration should be invalidated only in relation to those goods or service (and the rest of the registration should survive).²¹⁴ Partial invalidity for bad faith should mirror partial revocation so that an intention to use is required in relation to each and every sufficiently distinct sub-category of goods or services. There is no logical basis for permitting wider categorisation of goods or services for intention to use (bad faith) during the first five years of registration than that used one day later for revocation on the grounds of non-use.

While there will be some sorts of bad faith where the nature of the conduct means the entire mark should be invalidated, extending this to intention to use²¹⁵ would break the link with non-use and make too many registrations vulnerable.²¹⁶ So, for example, where an applicant applies for a mark in relation to alcoholic beverages, but intends to use it only for perry (pear cider), the specification should be partially invalidated so as to enable the registration for perry to remain. Accordingly, the General Court’s passing suggestion that a finding of bad faith invalidates an entire application²¹⁷ should apply only other types of bad faith.

In contrast to bad faith, the requirement of clarity would continue to be judged by the respective Trade Mark Offices. If clarity is confirmed as a substantive requirement, the only change of practice would be third parties could seek to (partially) invalidate existing and future registrations. This would not create a crisis on the register provided the imprecise term can be clarified by a narrowing amendment; meaning that, until challenged, the terms would remain on the register (as they do for old unused marks). Importantly, there is nothing in IP TRANSLATOR requiring the terms used in the specification to be based on sufficiently distinct sub-categories and so there is no reason why broad, but clear terms, cannot be used. However, absent use (and intention to use) across the whole of a broad category should any person apply to revoke or invalidate the mark, the specification would eventually have to be drafted in narrow terms in any event.

5.4 Final Thoughts

The requirement of precision (through an obligation of clarity and intent to use) would not be too onerous and could even be seen as lenient compared to other offices. We have seen how the path towards a more precise specification has been set by the history of the legislation and the case law of the European courts. To end, one can simply endorse the typically robust language of Sir Robin Jacob when in 2000 he said the problem of overly broad specifications needs resolution “sooner, rather than later … [so as] to stop this nonsense”.²¹⁸ It has taken some time, but the Court of Justice should take the opportunity it has been given in Skykick to make the register more accurate, less cluttered and more certain.