Dr. Dev Gangjee (London School of Economics) has a primary research interest in intellectual property, including trademarks, geographical indications and domain names. He is also an IP Osgoode Research Affiliate.
Can a building be a trade mark? Should a building be a trade mark?1 In recent years, trade mark registries and courts have grappled with applications for silhouettes, shapes, smells, textures, tastes, short cartoons, single colours, body movements, technical solutions, Tarzan’s yell, the positions of labels on products and brooding abstractions as trade marks. On the one hand, national, multilateral and international legal regimes have embraced an open-ended definition of a trade mark, which emphasises the functional rather than ontological status of a sign. Any sign which performs the communicative work of a trade mark can, in principle, be registered as one. On the other hand the trade mark system (legislation, doctrine, institutional architecture and conceptual apparatus) has accreted around paradigmatic subject matter: a conventional or traditional trade mark that is visual and consists of words, figurative devices or a combination of the two.2
Attempts to reconcile these competing imperatives – the system is open to all comers and its business as usual, yet the system was designed with a particular category of subject matter in mind – have resulted in the emergence of ‘coping techniques’ for non conventional marks (NCMs), which may have far reaching conceptual implications for trade mark doctrine as a whole. This post is part of a broader research project tracing the range of techniques, strategies and conceptual realignments within EU trade mark law, that are deployed at both institutional and doctrinal levels in response to this challenge. While outlining the emerging themes, the focus here is on recent efforts to register abstractions and concepts as trade marks.
The starting point for this story of pragmatic and conceptual accommodation is the open-ended definition of a trade mark found in modern regimes, which avoids being subject matter specific. National and international legislation refers to an open-ended definitional provision. Article 15(1) of TRIPs acknowledges that any ‘sign, or any combination of signs, capable of distinguishing the goods or services of one undertaking from those of other undertakings, shall be capable of constituting a trade mark.’ The Canadian definition is found in S 2 of the Trade-Marks Act 1985 (T-13) and refers to ‘a mark that is used by a person for the purpose of distinguishing… wares or services manufactured, sold, leased, hired or performed by him from those manufactured, sold, leased, hired or performed by others,’ whilst also including the broader concept of a ‘distinguishing guise’, as subsequently qualified in S 13. This echoes US statutory language: A trade mark is a designation, ‘any word, name, symbol, or device, or any combination thereof,” which serves “to identify and distinguish [the] goods [of the mark’s owner] … from those manufactured or sold by others and to indicate the source of the goods, even if that source is unknown.’3 The US Supreme Court has held that since ‘human beings might use as a “symbol” or “device” almost anything at all that is capable of carrying meaning, [the statutory definition] read literally, is not restrictive.’4 The ECJ has joined this chorus on a harmonious note in its Sieckmann decision5 which interpreted the equivalent definition in Art 2 of the EU’s Trade Marks Directive. Considered cumulatively, these definitions emphasise the communicative ability that a potential candidate must possess. The sign has to be distinctive; it should indicate origin and thereby distinguish or differentiate the goods or services upon which it is used, from others on the marketplace. This functional definition coupled with marketplace experience – shapes, product packaging and colours do sometimes act as indications of commercial origin – has led to formal recognition of NCMs as potential subject matter. Recent work by WIPO has documented the tribes of signs prepared to take advantage of this hospitality.6 These have been divided along sensory perception lines into visible signs (3D marks, colours, holograms, slogans, titles of media works, motion or multimedia signs, position marks and gestures) and non-visible signs (sounds both musical and non-musical, olfactory marks, tastes, textures).
Unsurprisingly, these new categories of subject matter have precipitated mixed responses. On the one hand trade mark law is expected to keep pace with consumer perceptions. If consumers experience colours or scents as indications of commercial origin, the underlying ‘marketplace communication enablement’ logic of trade mark protection demands that this is incorporated into registration practice and doctrine. Then again, it is easy to see why granting legal monopolies of potentially unlimited duration in colours or musical sounds would generate discomfort. This opens up three broad lines of enquiry.
Emerging Issues
(1) Procedural Accommodation
Representative registration is a key feature of contemporary trade mark registries. The register serves as a reference point for defining the scope of the protected subject matter, for allowing applications for marks to be examined and processed by the registry with certainty and also to enable those who consult the register to know what is protected and thus off limits. Historically the registration process has been premised on visual signs as its subject matter and therefore required graphical representation. While some non-visual signs can be represented graphically (a musical sound can be represented via sufficiently detailed musical notation, thereby enabling the recreation of the sign under protection), for others it’s less straightforward. For instance a registration for ‘the sound of a dog barking’ leaves readers speculating as to whether it’s a Pekinese or an Alsatian, whether it’s a welcoming or warning bark, its duration etc. Certain obstacles can be overcome by technology, such as uploading a digitised sound sample that is accessible online, or playing a short video clip for a motion mark. However for scents and tastes the competing demands of precision and accessibility have proved a serious impediment. The more scientifically accurate a representation of a scent (e.g. its chemical formula), the more inaccessible it becomes to the layperson and vice versa.
(2) Tensions Addressed by Substantive Law
Do consumers perceive certain categories of subject matter as trade marks? Even if they do, are there valid reasons for keeping certain signs off the register? Here the debates swirl around the core concept of distinctiveness and the functionality prohibition, which is explicitly based on policy grounds. Beginning with distinctiveness, if a colour or shape is perceived only fleetingly, or understood to be doing something else apart from indicating the product’s trade origin (for e.g. simple decoration or protective packaging for the product inside) it will not satisfy the distinctiveness requirement. A more controversial question is whether certain categories of signs will ever be able to satisfy the inherent distinctiveness requirement. The essential enquiry here is concerned with assessing whether the relevant public encountering the sign ‘cold’, for the first time, will understand it to be performing a trade mark function and the exemplars for this are invented words such as ‘XEROX’. Some jurisdictions (such as the US) have held that applicants for certain categories of signs, for instance colours, will always have to produce real world evidence of acquired distinctiveness or secondary meaning in the marketplace. Others (such as the EU) have determined that while the same distinctiveness test applies impartially to all signs, in practice consumers are unlikely to see certain types of signs as trade marks without some prompting, thereby leaving the door unlocked in theory but fairly hard to budge in practice.
When it comes to functionality, the focus shifts to the downstream effects of granting a single producer exclusive rights to NCM subject matter. For e.g. if a particular product shape is chosen because it produces a technical result, such as the face plate of the electric razor which produced a closer shave in the ECJ’s Philips decision,7 the result is an undesirable monopoly in a technical feature leading to an anti-competitive advantage for a single producer. Such signs are generally excluded from registration regardless of whether they are distinctive and in fact serve to indicate trade origin. Here the ingredients of the test for functionality and whether it extends to signs which incorporate stylistic or aesthetic features of a product continue to generate disagreement. This also leads a more broad based review of the overlaps between various other categories of intellectual property and trade mark law. If a short piece of classical music in the public domain can be trade marked as a sound mark, how will that affect the ‘limited monopoly’ balancing act of copyright law?
(3) Conceptual Realignments in the Future?
The enquiry here is more subtle and is only just beginning. It concerns the conceptual coherence and underpinnings of trade mark doctrine. What sort of feedback effects will emerge from the formal commitment to equality for all subject matter before trade mark law? Is it truly possible to apply the same tests for distinctiveness (under registrability) or likelihood of confusion or dilution (under infringement) across all categories of marks? Will the manner in which distinctiveness is assessed for NCMs have effects on the doctrinal presumptions for words and figurative marks? These are early days but the questions are worth considering.
Outer Limits?
Let me conclude with a case study which demonstrates the conceptual conundrums that NCMs potentially raise for trade mark law. So far we’ve looked at discrete categories of NCMs – shapes, smells, sounds – but since the definition of a trade mark is truly open ended, why stop there? What about themes, concepts, combinations of elements and other abstractions? Does trade mark law now lack any spatial or even temporal outer limits as regards its subject matter? Some of these issues were explored by various EU decision makers in the Dyson litigation, which illustrates the range of techniques used by registries and courts in response to the discomfort generated by some NCMs.
According to the applicant, the mark consisted ‘of a transparent bin or collection chamber forming part of the external surface of a vacuum cleaner as shown in the representation.’ The representation included two photographs of two separate models of vacuum cleaners with the externally visible transparent collecting bin displayed. In effect, Dyson appeared to be claiming the concept of a transparent collecting bin on vacuum cleaners as their trade mark, bolstered by the fact that they had enjoyed a period of patent enabled monopoly for those vacuum cleaners in the marketplace. The UK hearing officer8 objected on the basis that since the sign communicated a functional message it would be unlikely to communicate an origin indicating message to consumers and therefore lacked inherent distinctiveness. A transparent bin was a feature indicating (i) the cleaner was bagless and (ii) it indicated the vacuumed up contents of the container. It is interesting to note that the technical function of the bin related to both a lack of distinctiveness and to possible functionality objections. However the latter could not be independently raised as an objection since the application was pitched at the abstract level of a product feature across the range of vacuum cleaners and specifically not as a bin of particular shape or (transparent) colour, whereas the EU functionality objection is limited specifically to shapes.9 Furthermore, since the bins had not been promoted specifically as individual trade marks, there was no acquired distinctiveness. Customers made loose or temporary associations between the product feature and the first company to market it, but that was inadequate for trade mark purposes.
On appeal, the High Court10 grappled with the nature of the signs being applied for (aspects or features of the product itself) and was unpersuaded by the applicant’s arguments that a feature which was unique or striking was also inherently distinctive. While most ‘consumers sampled spoke of the striking and unusual appearance created by the transparent bin and of why they found that a useful feature’, this was not sufficient for distinctiveness. The signs under consideration were ‘descriptive of what the bin, or the machine as a whole, is or does. They say nothing about the machine’s commercial origin. The fact that it is a new type of cleaner does not of itself serve to associate it with any particular trader or manufacturer.’ However the court was uncertain about the existence of acquired distinctiveness in a situation where a patent monopoly for the product allowed Dyson to be the only producers on the market with that feature. The underlying concern here related to whether Dyson could build on a patent monopoly in order to acquire trade mark rights. So the court referred the matter to the ECJ, asking about the sorts of associations – and under which conditions – that would satisfy acquired distinctiveness. In particular did the transparent bins have to be actively promoted as trade marks during this period or would a looser association between feature, product and source suffice?
In a surprising development, the Advocate General and ECJ11 evaded the questions on distinctiveness by the referring court and hurtled out of the left field in their responses, first setting up and then dismantling the claim that this application qualified as a ‘sign’. As ‘a concept which appeals to the imagination’ and exists in headspace alone, this was not a sign. The ECJ went on to develop the test for separating abstract concepts from more tangible signs by relying on whether the subject matter could be perceived by one of the five senses. Since according to the EU definition of a trade mark12 the application fell at literally the first step on the road to registration, it couldn’t be saved by acquired distinctiveness, which only kicks in as an option at a later stage. The AG’s opinion also hints at a broader concept of functionality which operates beyond the narrow ambit of policing shapes and trade dress alone.
Finally, in parallel Community Trade Mark proceedings, it was decided that the Dyson application failed the graphical representation requirement as the written description and accompanying images did not convey a clear and unambiguous representation of the mark.13 The EU registry was concerned by the possibility that innumerable permutations and combinations of transparent bins were squeezed into that one written description with the two rather lonely accompanying photographs.
What is common to all these decisions is the tidal bore of unease that surges through them, driven by the prospect of allowing the applicant to gain a potentially permanent legal monopoly in a functional concept. Perhaps (judicial) nature abhors a vacuum but the manner in which this is translated into a legal test or rationale has consequences for the rest of trade mark law. In this illustration alone we see the requirement that the subject matter is specific enough to be a sign, graphical representation, inherent and acquired distinctiveness and functionality all explored as reasons for excluding problematic subject matter. With applicants intrigued by the possibilities of NCMs in a semantically crowded marketplace, conventional trade mark law looks like it will be severely tested in the days ahead. Trade mark law has a visually based communicative model and a related paradigmatic consumer at its core, which will require some serious rethinking.
1The Rock and Roll Hall of Fame v Gentile Productions 134 F.3d 749, 45 USPQ 2d 1412 (1998) (The claimant unsuccessfully argued that the design for the Rock and Roll Hall of Fame and Museum in Cleveland, Ohio – effectively the building itself – was a trade mark.)
2 WIPO Secretariat ‘New Types of Marks’ (SCT/16/2) 1 Sep 2006 at [3] (‘The types of signs that are nowadays considered as being capable of constituting a trade mark have expanded beyond words or figurative devices.’).
3 15 USC § 1127.
4 Qualitex Co. v. Jacobson Prods. Co., 514 U.S. 159, 162 (1995).
5 Sieckmann v. Deutsches Patent- und Markenamt (C-273/00) [2003] ETMR 37 (ECJ).
6 See WIPO Standing Committee on the Law of Trademarks, Industrial Designs and Geographical Indications (Sessions 16-19) At: http://www.wipo.int/meetings/en/topic.jsp?group_id=63
7 Philips Electronics v Remington Consumer Products [2002] ETMR 955 (ECJ)
8 Notetry Ltd’s Application (O29502) UK Trade Mark Registry, 23 July 2002 (Allan James)
9 As reflected in S 3(2) of the UK Trade Marks Act 994. In effect the functionality objection was converted into a distinctiveness objection.
10 Dyson Ltd v Registrar of Trade Marks [2003] EWHC 1062 (Ch); [2003] ETMR 77 (Patten J)
11 Dyson Ltd v Registrar of Trade Marks (C-321/03) [2007] ETMR 34 (ECJ).
12 Art 2 of the Trade Marks Directive 89/104/EEC stated: ‘A trade mark may consist of any sign capable of being represented graphically, particularly words, including personal names, designs, letters, numerals, the shape of goods or of their packaging, provided that such signs are capable of distinguishing the goods or services of one undertaking from those of other undertakings.’
13 See Dyson Ltd’s Application (R 655/2001-1) 2 Jul 2002 (Clear Bin). In response, Dyson unsuccessfully attempted to argue before OHIM that the application related to a series of marks.