Printer Friendly

DISENTANGLING FUNCTIONALITY, DISTINCTIVENESS AND USE IN AUSTRALIAN TRADE MARK LAW.

CONTENTS


  I Introduction: Managing Functional Trade Marks
 II How Functionality, Distinctiveness and Trade Mark Use Became
    Entwined
      A How Did Australia End Up without an Explicit Functionality Bar
        To Registration?
      B How Functionality Was Read into the Tests for Trade Mark Use
        And Distinctiveness
          1 Functionality and Trade Mark Use Entwined
          2 Functionality and Distinctiveness Entwined
          3 Problems with Relying on Trade Mark Use and
            Distinctiveness to Screen out Functional Marks
          4 The Alternative Approach: Allowing Factual Distinctiveness
            To Override the Need to Keep Functional Signs Free
III Do Distinctiveness and Trade Mark Use Provide Adequate
    Functionality Safeguards? Looking to Registration and Infringement
    Outcomes
      A Weeding Out Functional Marks at the Registration Stage
      B Interpreting the Infringement Test to Enable the Use of
         Functional Signs
 IV The Skewing Effect: How Focusing on Functionality Has Distorted the
    Tests for Distinctiveness and Trade Mark Use
      A Distorting Distinctiveness
      B Distorting Trade Mark Use
  V Recalibrating Distinctiveness and Trade Mark Use, and Rethinking
    The Shape of a Functionality Exclusion
      A Recalibrating Distinctiveness and Trade Mark Use to Deal with
        Non Traditional Marks More Effectively
        B Rethinking the Functionality Exclusion
          1 The Competing European Union and United States Models
          2 The Under-Inclusiveness of the European Union Model
          3 The Challenge of 'Aesthetic Functionality'
          4 Which Way Forward for Australia?
 VI Conclusion


I INTRODUCTION: MANAGING FUNCTIONAL TRADE MARKS

One of the most significant features of the trade mark provisions of the Agreement on Trade-Related Aspects of Intellectual Property ('TRIPS Agreement') is the stipulation in art 15(1) that '[a]ny sign ... capable of distinguishing the goods or services of one undertaking from those of other undertakings ... shall be capable of constituting a trademark ... [and] shall be eligible for registration'. (1) For many World Trade Organization ('WTO') Members, this involved recognising that non-traditional subject matter such as aspects of product packaging, the shape of goods themselves or colours used on goods or in conjunction with their marketing could constitute and be registered as trade marks. (2) At the same time, most WTO Members quickly appreciated that protecting or permitting the registration of non-traditional marks posed significant problems.

One difficulty was working out how longstanding tests for determining trade mark distinctiveness, which had been developed in the context of word and device marks, would map on to such new subject matter. A key concern here was to ensure that the tests would apply so that protection was not granted too readily over product features that would rarely be recognised as indicating trade origin so that the ability of other traders to use similar features in manufacturing and promoting their goods would not be unduly hampered. (3) An even more telling problem with permitting the registration of packaging features or elements of goods themselves, such as their configuration or colour, was recognised to be that some of these signs serve 'functional' purposes. Allowing the monopolisation of such signs was thought to raise the spectre of brand owners securing de facto monopolies over the functions such signs perform, causing significant competition concerns. (4)

There are a number of ways in which a non-traditional sign might be said to be functional. Most obviously, the sign might have a technical purpose. Examples include the shape of a toothbrush containing S-shaped bends in its handle to make it more flexible; a moulded ergonomic chair with the back and legs positioned to promote good posture; or a bright colour used on a rescue device to enhance its visibility in emergency situations. (5) A sign might also be functional if it conveys non-source identifying meaning or information about the goods, such as their ingredients or intended purpose, as with the use of red to indicate that caution is needed in the handling of the product, or a light and dark brown colour combination used on beverage packaging to connote a mocha flavour. (6) More controversially, it might even be said that where a nontechnical design feature of a product adds such aesthetic value that consumers might want to purchase the product because of that value (that is, they are prepared to pay some premium over the cost of the product attributable to its trade origin), the feature is serving an 'aesthetic function. The common concern in all of these cases is that it is entirely conceivable that the sign might have been used exclusively by a trader--whether by virtue of the fact the trader owns or owned patent or design rights covering the sign, or simply by virtue of being the first mover in a market where competitors did not choose to adopt the sign--such that it has become a factually distinctive trade mark. Granting potentially perpetual trade mark rights in such cases would clearly impose costs on present and future competitors who might otherwise wish to produce goods with those functional features or similar variants. These anti-competitive effects are thought to be particularly troubling if they have the effect of trampling over the in-built safeguards in the more obvious legal means of protecting functional product features--namely, patent or design laws--which provide for limited terms of monopoly protection, recognising the economic importance of allowing such features to fall into the public domain. (7)

Even though it might be difficult to agree on a clear set of principles for determining when, exactly, a sign is functional, there is widespread, and justifiable, agreement around the world that trade mark laws should contain limitations to ensure that such signs cannot be monopolised. The vast majority of WTO Members have sought to achieve this policy goal by categorically excluding functional signs from registration. That is, they have established outright bars to the registration of functional trade marks, which apply even if such marks satisfy the separate test for distinctiveness. (8)

A significant number of jurisdictions have considered that functional shape marks raise the most acute problems. Most significantly, the European Union ('EU'), in its 1988 Trade Marks Directive, established grounds of refusal and cancellation of registration for

signs which consist exclusively of:

* the shape which results from the nature of the goods themselves, or

* the shape of goods which is necessary to obtain a technical result, or

* the shape which gives substantial value to the goods ... (9)

These exclusions are said to have been motivated by an explicit desire to ensure that the life spans of patent and design rights are not indefinitely extended through trade mark laws and, at least in the case of the first two exclusions, that competitors have access to product features incorporating technical solutions. (10) Many countries have followed the EU's lead.

For example, Singapore, (11) India, (12) China (13) and Hong Kong (14) transposed the tripartite EU bar into their domestic laws. Sri Lanka (15) and South Africa (16) adopted only the first and second exclusions, although South Africa extended those exclusions to apply to the 'shape, configuration or colour' of goods. (17) Other countries have gone further in adopting bars that are not tied to particular types of sign. The most notable is the US, which has a longer history than most countries of protecting trade dress and product configuration, and of denying protection to functional marks. In 1998, the US amended its federal Trademark Act of 1946 to provide a ground of refusal of registration for a mark that 'comprises any matter that, as a whole, is functional'. (18) Registrations are refused on the basis of what is sometimes termed 'utilitarian functionality' (focusing on technical or utilitarian features of trade dress or product design) and 'aesthetic functionality' (focusing more on features that provide other, non-technical competitive advantages). The EU has recently moved closer to the broader US position: in both its 2015 Trade Marks Directive (19) and the amended EU Trade Mark Regulation, in force from 23 March 2016, (20) the three exclusions contained in the earlier 2008 Trade Marks Directive (21) have been extended to cover signs that consist of 'the shape, or another characteristic' of goods. (22) Notwithstanding the variety of approaches outlined above, what is clear is that a significant number of WTO Members have chosen, in enacting functionality exclusions, to prioritise competition concerns over the private interests of those traders whose signs have acquired distinctiveness and over limiting the consumer confusion that might result from multiple non-traditional signs being used in a market. (23)

So what then are we to make of the handful of countries that have not adopted explicit bars to the registration of functional marks in their domestic laws? Australia is perhaps the most notable example. Notwithstanding its close legal and historical ties to the United Kingdom ('UK'), and its tendency to look to other British Commonwealth nations in matters of intellectual property law reform, (24) Australia did not go down the same path as these countries on the issue of trade mark functionality when it came to reform its law to ensure compliance with the TRIPS Agreement. Australia's Trade Marks Act 1995 (Cth) ('TMA') adopted a broad, open-ended definition of 'sign' that specifically includes an 'aspect of packaging, shape, colour, sound or scent', (25) and a test for distinctiveness that contemplates that any mark can potentially acquire distinctiveness on the basis of use. (26) But it made no explicit provision for functional non-traditional signs. (27) On the face of it, this suggests that Australia sought to mediate the tensions between distinctiveness and functionality differently from most other countries, post-TRIPS Agreement. Indeed, it might be thought that Australia would provide an interesting, 20year long case study of the consequences of privileging the commercial interests of owners of non-traditional marks by affording them rights that potentially extend to prevent the third party use of functional product features, and of the impact of this deliberate policy choice on competition in particular markets.

Australia's experience with non-traditional marks is worth analysing, but for quite different reasons. It ends up being much more of a cautionary tale of how failing to be clear, in both legislation and case law, as to how competing policy concerns are to be ordered and managed within a trade mark registration system can result in the distortion of key doctrines within that system.

This article seeks to unpack the extraordinarily complex way in which Australian trade mark law has evolved over the last two decades in attempting to deal with some of the problems raised by functional non-traditional marks, and argues that reform of the law is long overdue. As will be seen in Part II, the history of the TMA suggests that the Australian government did intend, at least initially, to adopt a functionality bar to registration when it sought to reform the law in the mid-1990s, and the reasons for the absence of such a bar in the TMA are obscure. It then traces how functional trade marks were viewed by the Federal Court in the first few years of the TMA, showing how the Court, clearly troubled by the anti-competitive consequences of interpreting the law in a manner that would allow for the monopolisation of functional product features, attempted to find ways of incorporating functionality considerations into the TMA. It did so by 'shoehorning' such considerations into the doctrines of distinctiveness and trade mark use. The positive and negative consequences of this are explored in the next two Parts.

Part III investigates how the Trade Marks Office ('Office') has interpreted the TMA and internalised Federal Court case law on non-traditional marks, looking both to the Office's published decisions and registration outcomes involving shape and colour marks. It also considers how defendants have fared in Federal Court infringement proceedings involving registered functional marks. The analysis reveals that, judged in terms of outcomes, the 'shoehorning' approach to functionality appears to have been largely successful, such that it could be said that Australia has ended up with something close to a de facto functionality exclusion. Applications for registration of functional marks have, in many cases, been filtered out through a robust interpretation of the test for acquired distinctiveness, and courts have generally held defendants' use of functional matter to be non-infringing on the basis of a competition-friendly interpretation of the use 'as a trade mark' requirement for infringement. (28) However, it is argued that this de facto functionality exclusion is more unstable than has been appreciated, and is a suboptimal way of managing the problems raised by functional marks. Part IV deals with a more unexpected, and much more pressing, set of costs with the 'shoehorning' approach. By focusing so much on interpreting distinctiveness and trade mark use by reference to the problems raised by functional subject matter, the courts and the Office have skewed these doctrines as they apply to non-functional non-traditional marks. As a result, they have set the distinctiveness thresholds too low for such marks, and have found nonfunctional signs to be used as trade marks too readily in infringement proceedings. This has distorted the key policy goals sought to be achieved by the doctrines of distinctiveness and trade mark use in the TMA. Collectively, Parts II to IV show that functionality needs to be disentangled from distinctiveness and trade mark use in Australian law.

Part V then turns to consider what reform of the TMA should look like. While this involves rethinking aspects of the tests for distinctiveness and trade mark use, the key reform that needs to occur relates to functionality. Specifically, the legislature needs to confront functionality directly, and adopt an outright functionality bar to registration. The primary focus of this Part is determining what type of bar Australia should adopt. This, in turn, involves assessing whether Australia would be better off looking to the more detailed EU functionality rules or the more open-ended US standard. Although both the EU and US have been right to set up high bars that cover both technical and aesthetic functionality, it is argued that there are significant and ongoing problems with the language of the EU test, which is both over- and under-inclusive in what it prohibits from registration. Notwithstanding the problems with, and even incoherence of, a good deal of US case law on functionality, the open-textured nature of the US legislative standard makes it the preferable option for Australia.

II HOW FUNCTIONALITY, DISTINCTIVENESS AND TRADE MARK USE BECAME ENTWINED

A How Did Australia End Up without an Explicit Functionality Bar to Registration?

Understanding the problems with Australia's approach to functional trade marks first requires an appreciation of how Australia, unlike most other countries, ended up with legislation that lacks an explicit functionality bar to registration.

Before the TMA came into force at the start of 1996, Australia took a highly restrictive approach to the registration of non-traditional trade marks. (29) Under the Trade Marks Act 1955 (Cth) ('1955 Act'), a 'trade mark' was defined as a 'mark used or proposed to be used in relation to goods for the purpose of indicating ... a connexion in the course of trade between the goods' and the owner. (30) A 'mark' was defined non-exhaustively, but the illustrative examples were limited to traditional signs such as words, names, letters and devices. (31) Working with these definitions, Australian courts considered that a mark had to be 'something distinct from the goods in relation to which it is used', (32) or something 'extra' or 'added to the goods for the purpose of denoting the[ir] origin'. (33) Thus, an application for registration of the appearance of pharmaceutical capsules, consisting of differently coloured halves, was rejected on the basis that this did not meet the definition of a trade mark. (34) Under similar UK legislation, the House of Lords rejected an application for registration of the 'contour' shape of the Coca-Cola bottle on the same basis, (35) an approach thought to represent the law in Australia. (36)

In the early 1990s a working party, consisting of several interested groups (the 'Working Party'), was asked by the Australian government to review Australia's trade mark laws. (37) The Working Party accepted that the scope of registrable trade mark subject matter ought to be expanded. (38) Specifically, it was thought that the definition of a trade mark should encompass a wider range of signs, including shapes, colours, sounds and scents. These recommendations were made with more than one eye on the then-draft TRIPS Agreement, as well as recent developments in the EU. (39) At the same time, the Working Party recognised the importance of ensuring that technically functional or naturally occurring shapes and packaging could not be registered, although it did not consider whether shapes that added 'substantial value' to goods should also be prevented from registration. (40) It separately expressed a concern that the tests for inherent distinctiveness (or 'capacity to distinguish') should not be applied too liberally in relation to non-traditional signs. (41)

The Working Party vacillated, however, on how these concerns should be taken into account in any new legislation. In its discussion paper in 1991 it proposed a definition of a 'sign' that included 'colour, 'non-functional shape' and 'non-functional aspects of packaging'--that is, it sought to manage the issue of functionality via the definition of a trade mark. (42) But in its final report in 1992, it put forward an expansive, non-exhaustive definition of 'sign' that did not refer to non-functional features, (43) instead recommending the adoption of two discrete grounds of refusal of registration. One was for 'signs which are not capable of distinguishing the applicant's goods or services from those of other persons' (the distinctiveness ground), (44) the other was for 'signs which consist ... of elements, such as shape, which necessarily result from the nature of the specified goods or which are necessary to obtain a technical result' (the functionality bar). (45)

The Working Party's recommendations were influential. The Commonwealth Parliament, in late 1994, passed the Trade Marks Act 1994 (Cth) ('1994 Act'), before public comment on an exposure draft of the 1994 Bill had been fully considered, so that Australia could become a member of the WTO and comply with its TRIPS Agreement obligations by the start of 1995. (46) The 1994 Act adopted the Working Party's open-ended definition of a 'sign' (adding 'sound' as a further example) (47) and defined a 'trade mark' to be 'a sign used, or intended to be used, to distinguish goods or services dealt with or provided in the course of trade by a person from goods or services so provided or dealt with by any other person'. (48) It provided for a ground of refusal for lack of distinctiveness in s 40, and an outright functionality bar in s 39, applying where the trade mark the subject of the application for registration

consists wholly or principally of:

(a) the shape, or some other characteristic, possessed, because of their nature, by the goods; or

(b) a shape, or some other characteristic, that the goods must have if a particular technical result is to be obtained. (49)

Thus, by this point, Australia's position was clear. Previously unregistrable signs, such as shapes, colours or other features of goods, were now to be considered registrable if they were distinctive, but with an exception for 'naturally occurring' and 'technically functional' signs, which were to be excluded from registration outright. This reflected a policy, as previously articulated by the Working Party, to prioritise the interests of present and future competitors in having access to functional product and design features.

However, the 1994 Act never commenced operation. The following year, the government decided, after taking into account comments on the exposure draft of the 1994 Bill, to repeal the 1994 Act and replace it with the TMA. The TMA is similar in most respects to the 1994 Act. It contains the same, expansive definitions of 'sign' (50) and 'trade mark'. (51) It contains a comparable test for distinctiveness, albeit a more permissive one that contemplates that any mark can conceivably pass the test on the basis of evidence of acquired distinctiveness. (52) But the TMA does not contain an equivalent to s 39 of the 1994 Act.

The reasons for the disappearance of the functionality bar in the TMA are wholly unclear. There was no mention of the excision in the parliamentary debates on the 1995 Bill--indeed, the relevant Ministers responsible for the Bill claimed that it continued to reflect the Working Party's unanimous recommendations, one of which was the recommendation of a functionality bar to registration. (53) If documentary evidence exists that explains why the government changed its position, this has never been made public.

It is exceptionally difficult, therefore, to know what to make of the issue of functionality based on the text of the TMA and its history. On the face of the TMA, it seems that the only relevant ground on which a non-traditional trade mark can be refused registration is if it lacks distinctiveness under s 41, (54) with functionality not forming a separate barrier. It is entirely plausible that this is what Parliament intended. (55) Indeed, it is worth noting that the peak body representing the Australian legal profession, in making submissions on both the Working Party's 1991 discussion paper and the 1994 exposure draft Bill, advocated for the deletion of the functionality bar and argued that registration should hinge entirely on distinctiveness. (56) Such submissions might well have been persuasive, but if the government intended to go down this path it would have represented a major policy shift. In the alternative, it is arguable that because Parliament saw itself as giving effect to the Working Party's unanimous recommendations in enacting the TMA, and did not point to the excision of s 39 of the 1994 Act as being significant, it might have believed that a functionality exclusion had been embedded in one or more other sections of the TMA. The former interpretation chimes better with ordinary Australian canons of statutory interpretation, (57) even if there is little direct evidence in the history of the TMA to support it. The latter interpretation, despite its sound policy basis, begs the question of precisely how such an exclusion had been accommodated within the TMA.

As will be explained in the following section, the Full Court of the Federal Court has tended to favour the latter interpretation, at least in the case of shape marks, although judges have offered different views as to how, exactly, a functionality bar is contained within the TMA. (58) The Office, on the other hand, has more consistently adhered to the former interpretation, an approach that has found favour in a first instance Federal Court decision in a case involving the distinctiveness of a colour mark. (59) The ongoing uncertainty over this issue has come at a significant cost to the integrity of the Australian registration system as a whole.

B How Functionality Was Read into the Tests for Trade Mark Use and Distinctiveness

In the first four Federal Court cases to consider non-traditional marks under the TMA, decided between 1999 and 2002, the Court showed itself to be aware of the anti-competitive effects of allowing the monopolisation of functional shapes and colours, and keen to find ways of limiting the protection available to such signs. But none of the cases provided an especially good vehicle for a full consideration of all of the relevant issues, and the ways in which some of the Justices attempted to read functionality considerations into the 'trade mark use' and 'distinctiveness' doctrines in the TMA were problematic. (60)

1 Functionality and Trade Mark Use Entwined

In the first case, Koninklijke Philips Electronics NV v Remington Products Australia Pty Ltd ('Philips v Remington (Trial)'), (61) Philips contended that its registered mark, which consisted of a two-dimensional representation of the shape of its three-headed rotary shaver, had been infringed by Remington's marketing and sale of its three-headed rotary shaver. (62) There was no issue as to the validity of Philips' registration; the case turned solely on whether Remington had used the shape of its shaver 'as a trade mark', this being the threshold requirement for infringement under s 120 of the TMA. Lehane J accepted that under the new Act it was possible to infringe by using the entire shape of goods as a trade mark. (63) But his Honour found that Remington was not using its shape as an indicator of origin. Applying the orthodox test of 'trade mark use'--which looks to the context of the defendant's use of the sign to ask whether it is being presented as an indication of the origin of the defendant's goods (64)--Lehane J considered that the specifics of Remington's marketing of its shaver, especially the way the 'Remington' brand name was emphasised, meant that it was not using the shape as a mark and had therefore not infringed. (65)

On appeal, the Full Court agreed with this outcome ('Philips v Remington (Appeal)'), (66) but based on different, and convoluted, reasoning. Burchett J, with whom Hill J and Branson J agreed, was deeply perturbed by the prospect of Philips securing a perpetual monopoly over functional product features that would put it in a better position than a patentee or design owner, and would put competitors at a commercial disadvantage. (67) This strongly influenced his Honour's approach to the question of trade mark use. It was considered that even though the expanded definition of 'sign' in the TMA encompassed 'shape', the position under the 1955 Act, where a 'mark' needed to have a separate identity from the marked goods, remained unchanged. (68) Thus, it was held that Remington's manufacture and sale of goods with a functional shape did not involve a use of the shape in a trade mark sense: there was nothing 'added' to or severable from its goods that could be considered to be a trade mark. (69)

Burchett J recognised that this finding on 'use' in the context of infringement would have an impact on the potential registrability of shape marks under the new Act, an issue considered in extensive obiter dicta. His Honour suggested that because the old law on 'severability' remained good law, neither a naturally occurring nor technically functional shape 'could distinguish the goods of one trade source from the similar goods of another; and therefore such a shape could not function as a trade mark'. (70) He went on to say:
   Indeed, it is hard to imagine how such a shape of the goods
   themselves could be used, or be intended to be used, for the
   purpose set out in s 17 [of the TMA, which defines a 'trade mark'
   to mean 'a sign used, or intended to be used, to distinguish
   goods']--its use would inevitably be nothing other than part of the
   use of the commodity itself. (71)


It was on this basis that Burchett J explained away the removal of s 39 of the 1994 Act. It was thought that a standalone bar to registration would have been redundant in the TMA, since naturally occurring or technically functional signs could never be 'used', and thus never meet the new statutory definition of a 'trade mark'. (72) In fact, Burchett J's reliance on the severability doctrine to reach this conclusion means that any sign that is not severable from the specified goods (whether functional or not) would fail to meet the s 17 definition of a 'trade mark' and be barred from registration accordingly.

2 Functionality and Distinctiveness Entwined

The Philips v Remington (Appeal) approach to the registrability of functional marks was, however, reshaped by a differently constituted Full Federal Court two years later in Kenman Kandy Australia Pty Ltd v Registrar of Trade Marks ('Kenman Kandy'). (73) That case concerned the distinctiveness of an imaginary, six-legged 'bug shape' for confectionery, a shape that did not serve a technical purpose. This meant that the Justices' comments about the registrability of functional marks were, again, obiter. Stone J, with whom French J broadly agreed, (74) echoed the concerns expressed in Philips v Remington (Appeal) about the dangers of affording trade mark monopolies over functional signs, (75) but her Honour sought to cabin Burchett J's findings about the 'severability' doctrine. Stone J agreed that a mark had to be distinct from the 'inherent form' of the marked goods, but asserted that the inherent form of goods 'can only refer to those aspects of form that have functional significance.' (76) Her Honour also agreed that it was never intended for functional features of goods to be registrable under the TMA, but took a different view as to how s 39 of the 1994 Act had been absorbed into the TMA. For Stone J, marks that would have been caught by the former s 39 would never meet the test for distinctiveness under s 41 of the TMA. (77)

3 Problems with Relying on Trade Mark Use and Distinctiveness to Screen out Functional Marks

The Full Federal Court's decisions Philips v Remington (Appeal) and Kenman Kandy were remarkable interventions. Not only were the Justices unequivocal in their views that protection for functional marks was simply unavailable under the TMA, but they also managed to identify two distinct doctrines in which this exclusion had taken form. Judged purely in policy terms, the decisions are commendable for recognising the harms that might result from protecting functional marks, at both the registration and infringement stages, and for the clear attempt to read competition considerations into the fabric of the TMA. But this came at the expense of a detailed engagement with the question of whether Parliament had, by deleting the former s 39, shifted the policy balance in the TMA towards safeguarding the commercial interests of those who had built up distinctiveness in their functional, non-traditional marks. The more immediate problem, however, was with the way the Justices sought to build functionality exclusions into the 'trade mark use' and 'distinctiveness' tests in the Act. (78)

There are two significant difficulties in relying on 'use' to screen out functional marks in the manner suggested by the Full Court in Philips v Remington (Appeal). The first relates to the continued reliance on the severability doctrine. The problem goes beyond the artificial attempt in Kenman Kandy to limit the 'inherent form' of goods to 'aspects of form that have functional significance', (79) since the doctrine was never so confined under former law. (80) It is that the applicability of the doctrine depends entirely on the level of abstraction at which the 'goods' or their 'inherent form' are defined. It is only if the inherent form of goods is defined as being coterminous with the functional shape or other functional feature that it makes sense to say that they are inseparable. But once that inherent form is defined at a higher level of abstraction, a distinction can more clearly be drawn between that form and the product feature in question. Thus, in Philips v Remington (Appeal), the only way it could be said that Philips's shape was inseparable from its goods is if the goods were categorised as 'electric shavers with three rotary shaving heads arranged in an equilateral triangle'. But if the goods were classified generically, as 'shavers' or even 'electric shavers', it must be possible to say that a particular configuration of the shaver heads is different or separable from the 'goods', and could therefore be a sign 'used in relation to the goods'. (81) To be clear, the outcomes reached in applying the severability doctrine are entirely justifiable--indeed, such a broadly expressed doctrine ensures that market space is created for competitors who wish to copy the exact functional features in question, even if technical alternatives (of dissimilar appearance) are available. (82) Yet it cannot be pretended that the doctrine does much more than assert a desirable policy outcome. (83)

The second, related problem is that it is counterintuitive to find that a functional sign can never be 'used' as a trade mark. Even if a design feature serves a functional purpose, it can still be presented to consumers in such a way that, over time, they come to associate it exclusively with the manufacturer and rely on that feature as a source indicator. There are complex issues here about the desirability of, say, allowing a former patentee or design owner to be able to point to a period of monopoly use of the product under the patent or design as evidence that can go towards establishing acquired distinctiveness in the configuration of the product. (84) But the key point is that a sign is clearly being used as a trade mark if the result is that consumers come to rely on the sign as an indicator of trade origin. Indeed, the evidence presented by Philips in the litigation, in the context of its separate argument that Remington had engaged in passing off, indicated that consumers had come to associate the triangle shaver head shape with Philips, this being strong evidence of use of its shaver configuration as a trade mark. (85) This is not to suggest that the Court should therefore have found that the defendant, Remington, had used its shape as a mark: on the contrary, the trial judge was entirely correct to find that the specifics of Remington's use meant that it had not infringed. Indeed, Lehane J's approach offers a much more nuanced approach to the infringement inquiry than the Full Court's categorical exclusion of functional signs from the rubric of trade mark use.

The Full Court's retreat from the above position in Kenman Kandy, where it sought to incorporate a functionality bar within the distinctiveness test, is even more problematic. Both French J and Stone J asserted that a functional mark 'would not be capable of distinguishing' the specified goods, (86) but without explaining why this would be the case, or how such a decision could invariably be made under s 41 of the TMA. This section does not simply ask 'is the mark capable of distinguishing?'. Rather, it sets up a highly structured two-step inquiry. The decision-maker must first assess whether the mark is prima facie distinctive; if it is not, the decision-maker must then consider whether the mark will acquire or has acquired distinctiveness. This second step sets up a factual inquiry. If it can be established that the mark has been used to such an extent that as a factual matter it does distinguish such goods, it is to be taken to be capable of distinguishing. (87) To make sense of Stone J's claim, it has to be assumed that a functional sign can never as a factual matter be used as a mark, such that there will never be any 'use' that can go towards establishing acquired distinctiveness. As argued above, this can only be done on the basis of the flawed severability doctrine, and it flies in the face of marketplace reality. (88)

It is worth noting that the majority in Kenman Kandy did not consider the trade mark laws of the EU, the US or any other country that had adopted a functionality bar to registration, all of which had been structured around the possibility that a functional mark could acquire distinctiveness. (89) Closer to home, this point had also been raised by Lehane J in Philips v Remington (Trial) (90) and in commentary on that case. (91) In short, the majority in Kenman Kandy was in error in asserting that an outright functionality exclusion could be built into the distinctiveness test in s 41. (92)

4 The Alternative Approach: Allowing Factual Distinctiveness to Override the Need to Keep Functional Signs Free

Stone J's consideration of functionality through the lens of distinctiveness (rather than as something going to the concept of 'use' in the TMA s 17 definition of a 'trade mark') did, however, move the courts closer to the approach to shape marks that had been taken by the Office since the TMA came into force in 1996.

The Office had been looking exclusively to s 41 in considering the registrability of such marks, including where these were technical, or even aesthetically, functional. (93) But, in contrast to the view that came to be adopted in Kenman Kandy, the Office saw functionality as going only to the first step of the distinctiveness inquiry: that is, whether the mark is prima facie distinctive. Under Australian law, the test for prima facie distinctiveness has long been explicitly based on competition concerns. In the High Court case of Clark Equipment Co v Registrar of Trade Marks ('Clark Equipment'), Kitto J held that it is to be assessed by looking to
   the likelihood that other persons, trading in goods of the relevant
   kind and being actuated only by proper motives--in the exercise,
   that is to say, of the common right of the public to make honest
   use of [signs] forming part of the common heritage, for the sake of
   the signification which they ordinarily possess--will think of the
   [sign] and want to use it in connexion with similar goods in any
   manner which would infringe a registered trade mark granted in
   respect of it. (94)


Since most shape marks would fail this test of prima facie distinctiveness, the Office's assessment would invariably turn to the second step of the distinctiveness inquiry: that is, whether the mark had become factually distinctive. Here, the Office was not prepared to find that a functional shape could never be used or never acquire distinctiveness. Rather, it saw the main problem as working out what evidence of use would be sufficient to show that the functional shape mark had become factually distinctive. (95) Strikingly, the Office did not respond to the Full Federal Court's obiter comments in either Philips v Remington (Appeal) or Kenman Kandy by changing its practice and attempting to read an outright functionality bar to registration into the TMA (even if, as we will see in Part IV, these court decisions have come to influence Office practice more subtly). (96)

The Office's approach, in which functionality concerns have a direct role to play only in the assessment of prima facie distinctiveness, is also reflected in the third Federal Court decision to address the issue of functional nontraditional signs. Philmac Pty Ltd v Registrar of Trade Marks ('Philmac'), decided four months after Kenman Kandy, involved an attempt to register 'the colour terracotta as applied to the connecting insert of polypipe fittings. (97) The evidence showed that pipe fittings, which were generally all black, had been sold in retail stores with different companies' products mixed together. Philmac was the first company to recognise that applying a colour to its fittings made them stand out to consumers. (98) The only issues for the Court were whether the colour mark was prima facie distinctive and, if not, whether Philmac had demonstrated acquired distinctiveness under s 41.

Mansfield J cited Clark Equipment and the standard, competition-focused tests for prima facie distinctiveness, (99) and then considered a range of circumstances in which it might be said that other traders would not wish to use a single colour in marketing their goods. It was held that this would be so where:

* the colour does not serve a utilitarian function: that is, it does not physically or chemically produce an effect such as light reflection, heat absorption or the like;

* the colour does not serve an ornamental function: that is, it does not convey a recognised meaning such as the denotation of heat or danger or environmentalism;

* the colour does not serve [a natural] function: that is, it is not the naturally occurring colour of a product and registration of that colour in respect of that product would not thereby submit competing traders to extra expense or extraordinary manufacturing processes in order to avoid infringement; [or]

* the colour mark is not sought to be registered in respect of goods in a market in which there is a proven competitive need for the use of colour, and in which, having regard to the colour chosen and the goods on which it is sought to be applied, other properly motivated traders might naturally think of the colour [and wish to] use it in a similar manner in respect of their goods. (100)

Philmac's colour did not serve any of the functions identified in the first three bullet points, but Mansfield J held that there was a proven competitive need for the use of colour in the market for pipe fittings, such that the mark was not prima facie distinctive. Based on evidence of almost four years' use, the sale of several million fittings, and high levels of consumer recognition that the terracotta colour served a trade mark function, Philmac's mark was found to have acquired distinctiveness. (101)

Mansfield J's articulation of the various functions a colour might serve is laudable. Importantly, it represented the first acknowledgement by an Australian court that functional marks are not limited to those consisting of 'characteristic[s], possessed, because of their nature, by the goods' and 'characteristics] ... the goods must have if a particular technical result is to be obtained' (these being the two categories listed in s 39 of the 1994 Act and considered in Philips v Remington and Kenman Kandy). In referring to 'ornamental function', Mansfield J appreciated that some colours might convey non-source identifying information, connotations or meanings, and that competitors would be unfairly disadvantaged if they were unable to use such colours to access these meanings (something that might in other jurisdictions fall to be considered under an 'aesthetic functionality' doc trine). (102)

However, the most striking feature of Philmac is that Mansfield J did not attempt to set up an outright functionality bar to the registration of colour marks. His Honour did not suggest that when a colour mark is functional for one of the three identified reasons ('utilitarian, 'natural' or 'ornamental'), it should somehow be disqualified from protection--either because the colour is inseparable from the 'inherent form' of the goods or because the colour could never be distinctive. Rather, it was recognised that, under s 41 of the TMA, those functionality concerns could count only towards the assessment of whether or not the mark is prima facie distinctive. (103) If the mark had acquired distinctiveness, however, it could be registered--just as the Office had recognised with shape marks.

The Philmac-Office approach to the registrability of non-traditional marks that serve functional purposes has become the settled position in Australia. (104) However, as we will see, the impact of Philips v Remington and Kenman Kandy has continued to be felt in unexpected and problematic ways.

III DO DISTINCTIVENESS AND TRADE MARK USE PROVIDE ADEQUATE FUNCTIONALITY SAFEGUARDS? LOOKING TO REGISTRATION AND INFRINGEMENT OUTCOMES

It might be thought, based on the discussion in Part II, that the absence of a functionality bar in the TMA, and the limited role that functionality has come to play in the distinctiveness test, would have had profoundly negative consequences. In particular, it might be assumed that in the 20-odd years in which the TMA has been in force, a substantial number of functional nontraditional marks would have been registered on the basis of acquired distinctiveness, and that there would have been cases where parties had infringed by using functional signs that were sufficiently similar to registered, functional marks. Part III tests this assumption. It shows that although it is possible to point to the existence of registered functional shape and colour marks, the way in which the acquired distinctiveness test has been interpreted has meant that the distinctiveness doctrine has done a good job of keeping functional marks off the Register. Further, the infringement test has been applied in a pro-competitive manner in cases where defendants have sought to use functional product features. Yet a closer look at the state of the Register and the infringement cases suggests that we should not be too sanguine about the ability of the distinctiveness and trade mark use doctrines to do the job of an outright functionality exclusion.

A Weeding Out Functional Marks at the Registration Stage

The most useful starting point in considering how functional marks have fared in practice is with the published decisions of the Office (that is, decisions in cases where an applicant requested a hearing before a delegate of the Registrar after a distinctiveness ground of rejection had been maintained in examination, or where an opposition to registration was based on a lack of distinctiveness). (105) Up to the end of 2017 there had been 22 published decisions on s 41 of the TMA in which the Office considered that the shape or colour mark in question was wholly or predominantly functional, (106) with 19 involving technically functional marks. (107) In every one of these 22 decisions the Office found that the mark was not prima facie distinctive. In 21 of these decisions it was found that the mark had not acquired distinctiveness; (108) in the other decision the issue did not fall for consideration. (109)

The most interesting feature of these decisions is the way in which the Office applied the test for factual distinctiveness. While in some cases the evidence of use was non-existent or clearly deficient, (110) in others applicants were able to point to extensive sales of goods bearing the shape or colour in question, and considerable advertising and marketing expenditure, often over substantial periods of time. In these cases the Office--clearly conscious of the functional nature of the marks in question--sought to apply the acquired distinctiveness test in a robust manner. It emphasised that what was needed was more than evidence of a substantial period of exclusive use of the shape or colour, but rather use of the shape or colour as a trade mark. That is, it demanded evidence that showed that by the manner of its use, the shape or colour had come to indicate that the specified goods came from a single trade source. In doing so, the Office drew on British authority that suggested that it would not be enough to show that consumers might associate the sign with the owner due to its use, but that they would need to 'rely' on it as an indication of origin. (111) Adopting a reliance-based approach meant that the Office felt comfortable downplaying evidence of purported acquired distinctiveness where the shape or colour had been used consistently with primary branding, which served to push out any source significance the shape or colour might have had. (112) More generally, it meant that the Office could dismiss a good deal of evidence of 'use' as showing little more than the fact that goods bearing the shape or colour had been offered for sale, with associated marketing efforts that accentuated features of shape or colour doing little more than indicating that the goods had those features. (113)

Looking only at published Office decisions, it might therefore be said that the distinctiveness test has, in practice, done an excellent job of screening out functional marks. A somewhat mixed picture emerges, however, from an assessment of the state of the register. (114)

Between 1 January 1996 and 31 December 2017, there were 3,159 applications for registration of marks classified by the Office as 'shape' marks, resulting in 1,225 registrations (237 of which were subsequently removed for non-use or were not renewed). The overwhelming majority of these registrations were for aspects of packaging, especially bottle shapes, and numerous marks classified as 'shape' marks have in fact consisted of combinations of shapes, labels, colours and/or brand names. For 'colour' marks, 1,259 applications for registration were made over the same period, resulting in 358 registrations (52 of which were removed for non-use or not renewed). Again, a significant number of these were for colours combined with a shape or device, or for colour combinations. Whether functionality was considered in the examination of the distinctiveness of these 'shape' and 'colour' marks is impossible to assess from the public record. However, based on the representations of the marks on the Register and their specifications, it is possible to identify numerous 'shape' and 'colour' marks that give rise to functionality concerns. Most notably, the shape of the Rubik's cube, (115) the four-finger Kit Kat chocolate bar, (116) Stokke's 'Tripp Trapp' high chair, (117) and the 'non-slip' handle design of Yoshida's kitchen knife (118) have all been registered in Australia. This is so notwithstanding that these marks have either been refused registration on functionality grounds in the EU (119) or in other jurisdictions that have copied the EU's bar, (120) or at least have given rise to serious functionality concerns in those jurisdictions. (121) Other examples of Australian registrations of arguably functional shapes entered on the Register since 2014 alone have included a container with a curved handle featuring a number of ribbed ridges, (122) a toilet cleaning product containing four balls in a cage, (123) a basic cylindrical bottle with a silver lid, (124) a clasp on a plastic container, (125) a furniture support or foot intended to sit under the leg of a piece of furniture, (126) and a 'nasal support device' for animals. (127) Older examples are provided in Parts III(B) and IV(A) below. Illustrative examples of functional registered colour marks include black applied to a container for medicated cream, (128) green for chocolate beverage preparations, (129) cerise for horse riding clothes, (130) a gold disc for audio equipment, (131) orange for vehicle recovery devices, (132) and a series of seven colours applied to packaging for tampons. (133) A survey of the register thus demonstrates that the distinctiveness test is, at the very least, an imperfect mechanism for weeding out technically functional shape and colour marks.

B Interpreting the Infringement Test to Enable the Use of Functional Signs

Even if some functional marks have found their way on to the register, a more significant question arises: in what circumstances will the owners of such marks be able to enforce rights in them? As seen in Part II(B), Philips v Remington (Appeal) is authority for the proposition that, in the infringement context, a defendant's use of a functional sign will not be considered to be 'use as a trade mark', which is a threshold requirement for infringement under s 120 of the TMA. This decision has been embraced in a number of subsequent cases involving allegations of infringement of functional shape marks that had been registered under the TMA.

Mayne Industries Pty Ltd v Advanced Engineering Group Pty Ltd ('Mayne Industries'), from early 2008, provides the best example. (134) The registered mark consisted of an S-shaped 'fence dropper' (a device that clips to a wire fence). The fence dropper had been the subject of a patent that had expired in 1984; the application for registration of the shape as a trade mark was made in 2001. After that time the defendant copied the product and marketed it under a different brand name. Given the functional nature of the fence dropper, Greenwood J had little difficulty applying Philips v Remington (Appeal) in finding that the defendant had not engaged in trade mark use. (135) Even though the defendant had sought to draw attention to the shape of its fence dropper in its marketing, this was held to have been for the purpose of advertising the functional features of the product, and not to indicate trade origin. Like Burchett J in Philips v Remington (Appeal), Greenwood J was particularly concerned about what a finding of trade mark infringement would mean in circumstances where patent law had limited the monopoly available to the inventor of a functional product. (136)

The interpretation of trade mark use in Philips v Remington (Appeal) and Mayne Industries now appears to be entrenched. In the later case of Sebel Furniture Ltd v Acoustic & Felts Pty Ltd ('Sebel Furniture'), (137) the owner of a registered shape mark for an ergonomic chair sought an interlocutory injunction to restrain the sale of a similar-looking chair. Foster J considered that there was not even a serious question to be tried as to whether the defendant had engaged in trade mark use of the shape of its chair. (138)

The effect of these cases--especially when read in light of the fact that the test for factual distinctiveness will often filter out functional marks from registration--is to create something close to a de facto functionality exclusion in Australian trade mark law. Given the strong policy basis for such an exclusion, as outlined in Part I, this is a positive outcome. However, two notes of caution should be sounded.

The first is that the Philips v Remington (Appeal) line of cases might be said to go too far in setting up a categorical exclusion from liability for the use of functional signs. The same results could have been achieved in the cases on a more conventional basis. In each of the cases the defendant was using primary branding on its goods ('Remington' on the shaver; 'ezy-lock' on a swing tag attached to the fence dropper; 'Titan' branded chairs) and was not doing anything specifically to suggest that the shape of its goods was indicating origin. But, as we saw in Part II(B)(4), it is problematic to suggest that a functional sign can never be used as a trade mark, and at the distinctiveness stage the Court has shown a degree of comfort with the idea that such signs can be used as trade marks, at least in the context of colour marks. It is therefore possible to envisage a future infringement case with more complex facts than Philips v Remington (Appeal), Mayne Industries or Sebel Furniture where the defendant makes an overt attempt to trade off the functional shape or colour in question as a badge of origin. Just as courts have recognised that a word may have a descriptive element to it but still be used as a trade mark, (139) it is conceivable that a court could make a similar finding in relation to the use of a functional sign. (140)

Having said this, something that might justify categorically excluding the use of functional signs from the concept of 'trade mark use' is the special role that this concept has come to play in the test for infringement in Australian law. From very soon after trade mark use was developed in the UK in the mid-1920s, as a means of circumscribing the scope of the registered owner's monopoly, courts started interpreting the doctrine flexibly to give effect to important policy considerations. Thus, the sort of use held to fall outside the scope of trade mark use came to include use in comparative advertising, (141) the parallel importation and sale of both legitimate branded goods (142) and legitimate second-hand goods, (143) and some types of nominative use. (144) A similar carve-out for functional signs is therefore entirely consistent with the way the trade mark use requirement has long operated in Anglo-Australian law. This flexible interpretation of trade mark use should be embraced, but it is important to note that it is under threat. Over a series of cases under the TMA, the Federal Court has come to find that the parallel importation and sale of legitimate trade marked goods, and by implication the sale of secondhand trade marked goods, constitute use of those trade marks by the importer or reseller and are prima facie infringing. (145) This reading is defensible, but it overlooks the complex way in which trade mark use evolved over the twentieth century to accommodate a broad range of policy concerns. (146) What is most important for present purposes is that this recent trend in the interpretation of trade mark use calls into question the stability of the foundations of the Philips v Remington (Appeal) line of cases.

A second note of caution about the present state of Australian law relates to the costs of having 'false positives' on the Register. If a primary justification for having a Register is that it provides valuable information, verified by a bureaucratic process, as to the signs that third parties should and should not be able to use, it must be the case that most third parties will understand that registration is a clear signal to steer clear of the use of that mark or a similar variant. A well-advised trader might be told that if it chooses to adopt a functional sign that is closely similar to a registered but functional mark the chances of a successful infringement action being brought against it are small. But even a well-advised trader might not wish to take such a risk. This is not only because of the difficulty of determining whether or not a registered mark is truly functional, or whether the trader might be liable if it copies both functional and non-functional elements of a registered mark. (147) It is also because the costs of adopting a different functional product and suffering any associated commercial disadvantage in so doing might seem to be better able to be absorbed than the costs and inconvenience involved in successfully defending Federal Court infringement proceedings. (148) In other words, even if the law seems to be clearly on the side of parties seeking to make use of functional trade marked product features, the chilling effects of registration, and the potentially negative impact on competition that might result, still need to be taken seriously. (149)

IV THE SKEWING EFFECT: HOW FOCUSING ON FUNCTIONALITY HAS DISTORTED THE TESTS FOR DISTINCTIVENESS AND TRADE MARK USE

The analysis in Part III demonstrated that 'distinctiveness' and 'trade mark use' have ended up performing important gatekeeper roles in ensuring that functional product features can rarely be registered, and even more rarely enforced against third party competitors. However, as suggested at the end of Part III, there are problems with the way in which functionality, distinctiveness and use have become intertwined, and the Australian model cannot be said to provide the same degree of comfort to those wishing to access and use functional product features as similarly placed competitors in overseas markets. But there is perhaps an even more significant problem with the way in which functionality concerns have been 'shoehorned' into the distinctiveness and use doctrines at both the registration and infringement stages. This relates to the way in which the distinctiveness and use doctrines have ended up being distorted as they apply to non-traditional marks more generally.

A Distorting Distinctiveness

Kenman Kandy provides the clearest example of how the shoehorning approach outlined above has ended up bending the distinctiveness test out of shape. Notwithstanding that the 'bug shape' mark under consideration was not technically functional, Stone J (with whom French J agreed) considered technical functionality in great detail. In doing so, the majority seemed to believe that only technically functional shape marks give rise to sufficient competition concerns that would justify the distinctiveness test being interpreted strictly, so as to impose a barrier to the registration of such marks. But when the judges finally turned their attention to the distinctiveness of the bug shape before them, they seemed to believe that because the shape could be described as being 'invented' (150) or 'concocted' (151) it should be treated as if it were an invented or concocted word. On that basis, the majority found the shape to be prima facie distinctive. Both majority judges were unconcerned about the impact of registering such a mark without the need for the applicant to show factual distinctiveness, with French J asserting that it would be 'speculative, absent evidence, to draw conclusions ... [as to] whether the particular arrangement [of the features of the shape] has any significant impact upon the access of other traders to the use of insect like shapes as trade marks'. (152)

Kenman Kandy is a deeply problematic decision that draws a false ontological equivalence between invented words and invented shapes, and puts Australian law badly out of step with other jurisdictions, most notably the US. (153) The key problem for present purposes is, as Mark Davison has noted, that by seeing the functionality of a shape as a factor that makes it non-distinctive, the Court mistakenly assumed the obverse of this proposition: that non-functional features of a shape must therefore make it distinctive. (154) This misunderstands the pro-competition policy that underpins the Australian test for prima facie distinctiveness. In asking whether other traders wishing to make and sell the specified goods would wish to use the shape of goods (or a similar shape) for the sake of its 'ordinary signification', the answer will invariably be yes, as this signification is simply that the goods have a particular shape, and this is something that other traders are generally free to copy. The competition concerns that underpin the test for prima facie distinctiveness remain the same for functional and non-functional shape marks alike. (155) It is just that for functional marks, the competition concerns are arguably so overwhelming that they should continue to take precedence over a finding of acquired distinctiveness, hence the arguments in favour of an outright functionality exclusion. (156) But Kenman Kandy has lowered the bar for distinctiveness of non-functional shape and packaging marks, and rides roughshod over such competition concerns--it allows a first mover in a market to secure potentially perpetual trade mark rights over a new product design even before it commences use of the design, denying future competitors the ability to compete in the market for similarly-shaped goods by using that or a similar design. To justify this outcome on the basis of a lack of 'evidence' that registration would have a 'significant impact upon the access of other traders to the use of [similar] shapes as trade marks' (157) is doubly problematic: not only does this misapply High Court authority on how prima facie distinctiveness is to be assessed, (158) but it fails to appreciate the forward-looking nature of the distinctiveness test that aims to preserve competition in future markets, where 'evidence' of impeded market access does not yet exist.

The Kenman Kandy distortion of the test for prima facie distinctiveness has, inevitably, had a negative impact on Office decisions and especially on registration outcomes. In the Office's decision in Application by FreshFood Holdings Pte Ltd, (159) the applicant had sought to register an irregular hexagonal shape of a coffee jar. The hearing officer, following Kenman Kandy, emphasised that the jar was 'complex' and 'not a common shape', thus finding it to be prima facie distinctive. (160) More worryingly, a significant number of shape marks have been entered on the register since the Kenman Kandy decision without an endorsement indicating that the mark was accepted upon a showing of acquired distinctiveness. Noteworthy examples of currently registered shape marks that were accepted as being prima facie distinctive include a kennel on wheels (for animal care related goods and services), (161) a trophy shaped like drag racing starting lights (for entertainment and sporting services), (162) and various liquor and perfume bottle shapes. (163) Similar shape marks whose registrations were not renewed include the sole of a sports shoe, (164) a wine bottle with a vine leaf and grapes embossed on the exterior, (165) a container in the shape of an ice cream cone (for ice cream), (166) a disc with a grooved surface (for CDs and DVDs), (167) pasta pieces shaped like a human skeleton, (168) a drop-shaped loudspeaker, (169) an aquarium in the shape of a wooden barrel, (170) a citrus juicer (171) and a square tile with a fin-shaped protrusion (for building tiles). (172) The fact that all of these marks were accepted for registration without the applicant needing to show any evidence of use is greatly concerning.

Further, the Kenman Kandy distortion of the test for prima facie distinctiveness has caused an unusual feedback effect that has made it easier for a larger number of descriptive non-traditional marks, and even some functional non-traditional marks, to be registered. The reason for this is complex. Australian law recognises two categories of non-prima facie distinctive marks: those that are 'not to any extent inherently adapted to distinguish' the specified goods or services, (173) and those that are 'to some extent, but not sufficiently, inherently adapted to distinguish' such goods or services. (174)

These are unusual categories, (175) and it remains difficult to pinpoint exactly when a non-prima facie distinctive mark will fall into one category rather than the other. The distinction makes more sense in the context of word marks, where, for example, entirely descriptive words will fall into the 'not to any extent inherently adapted' category, whereas less descriptive words, or descriptive words represented in a syntactically unusual or stylised manner, will tend to fall into the 'not sufficiently inherently adapted' category. (176) The consequences of the classification are significant. A mark falling in the 'not sufficiently inherently adapted' category can be registered if, on the basis of the evidence of use or intended use, it has or will become factually distinctive--that is, it is possible for such a mark to be registered even if it has not acquired distinctiveness at the time of assessment. In contrast, a mark falling in the 'not to any extent inherently adapted' category can be registered only on the basis of pre-filing date use that demonstrates that the mark had acquired distinctiveness by that date. (177)

For non-traditional marks, a good case can be made on first principles that because the ordinary significations of such signs are simply that they are the shape of the goods, an aspect of the packaging of the goods, the colour of the goods, etc, they should always fall into the 'not to any extent inherently adapted' category--that is, they are the equivalent of a word mark that entirely describes the goods. However, the effect of Kenman Kandy, with its focus on the appearance of the goods and the extent to which their shapes might be described as being 'invented' or 'concocted', has meant that in subsequent cases the courts and the Office have tended to classify a large proportion of non-traditional marks as being 'not sufficiently inherently adapted' if there is something slightly unusual or striking about their appearance. (178) From the date of the Kenman Kandy decision up to the end of 2017, at least 181 non-prima facie distinctive shape marks were entered on the Register. (179) One hundred and forty-nine of these marks, or 82%, were entered with an endorsement indicating that they had been classified in examination as being to some extent, but not sufficiently, inherently adapted to distinguish. (180) Of the 127 non-prima facie distinctive colour marks so registered, 94 marks, or 74%, were classified as not sufficiently inherently adapted to distinguish. (181) To be clear, it is not possible to determine whether these marks were ultimately registered on the basis that they had in fact acquired distinctiveness by their filing dates, or that they merely would do so in the future. The problem is more that such a high proportion of non-traditional marks are being classified in the 'not sufficiently inherently adapted' category at all, meaning that the owner at least has the option of trying to make a sufficiently persuasive case that the mark will acquire distinctiveness at some point in the future. This in turn means that there must be a likelihood that marks that are not factually distinctive, and indeed might never acquire distinctiveness, find their way on, and remain on, the Register. (182)

More troublingly, the effect of this approach is to place some functional marks in the 'not sufficiently inherently adapted' category, simply because such marks might be 'striking' or have some 'visual appeal'. (183) Examples of technically functional shape marks that were considered to be 'not sufficiently inherently adapted' that were entered on the Register post-Kenman Kandy include a toothbrush with a flexible handle, (184) a plastic fastener, (185) a coffee capsule, (186) the Yoshida knife handle, (187) a rectangular card dispenser, (188) the hook portion of a clothes hanger (189) and a hinged, foldable electronic device cover. (190) Registered functional colour marks that were classified as 'not sufficiently inherently adapted' include a brown colour combination applied to a basic bottle shape (for dairy products and chocolate based beverages) (191) and a gold coloured disc (for audio equipment). (192) That functional marks are potentially able to be registered based on arguments that they will acquire (but have not in fact acquired) factual distinctiveness raises further questions about the ability of the distinctiveness test to screen out functional signs from registration.

B Distorting Trade Mark Use

There are similar problems with shoehorning functionality considerations into the 'use' test. An analysis of key infringement decisions post-Philips v Remington (Appeal) reveals that the courts have been far too quick to find that a defendant's use of a non-functional, non-traditional sign constitutes trade mark use.

The best illustration of this is Global Brand Marketing Inc v YD Pty Ltd ('Global Brand'), (193) decided in 2008 shortly after Mayne Industries. The two registered marks in question consisted of the configuration of a shoe and the sole of a shoe. These marks had been registered on the basis of prima facie distinctiveness, but arguably only because both shapes bore a prominent stylised 'D' logo (which would have been prima facie distinctive in its own right). With no requirement under Australian law for marks that consist of distinctive and non-distinctive elements to be registered subject to disclaimers over the exclusive use of the non-distinctive elements, (194) the marks were registered without the owner needing to show that the configuration or the sole in isolation bore any source significance. The defendant sold shoes with an allegedly similar sole pattern and overall appearance, albeit marked with its 'yd' logo and not the trade mark owner's 'D' logo.

In considering the threshold question of whether the defendant's use of its sole and its shoe configuration involved trade mark use, Sundberg J drew a distinction between functional features of goods and those features that might be 'striking' or 'make goods more arresting of appearance and more attractive, with the latter being likely to indicate trade mark use. (195) The defendant's shoe configuration was thought to present something of a challenge, in that it consisted of a mix of functional or common features (stripes on the side, a square toe cap, a facing and a padded collar) and non-functional features (the 'yd' label, a piece on the heel, a flat sole, a rubber panel on the instep and the pattern on the sole). Sundberg J ultimately concluded that the non-functional elements 'overwhelmed]' the functional elements, such that the entire shape was being used as a trade mark. (196) As for the defendant's sole, which Sundberg J emphasised was not technically functional, it was held that because the pattern was 'distinctive' and would be noticed by shoppers, it was also being used as a trade mark. (197)

Global Brand suffers from a similar problem to Kenman Kandy in its assumption that non-technically functional features of a shape will almost invariably serve a trade mark purpose. By setting up this false dichotomy, the Court ended up taking the wrong approach in assessing whether nonfunctional product design features are being used in a trade mark sense. It simply cannot be decisive that such features are 'attractive' or 'arresting of appearance, because this could equally support a finding that the features are being used in a purely ornamental or decorative (and non-trade mark) manner. The Court ended up glossing over the factors that would help in a determination of whether features of shape are being used in a trade mark sense, rather than as mere decoration. For example, there was little evidence in the case that the defendant had marketed its shoes in a manner that sought to present their shapes, and the sole patterns in particular, as having source significance. Equally, there was little evidence that consumers did in fact recognise these features as indicating the origin of the defendant's shoes--the limited evidence that consumers might have noticed the overall appearance and sole pattern of the defendant's shoes before purchase was clearly insufficient in this regard. There was no consideration of the fact that the registered owner had not established that the configuration and pattern of its shoes, apart from its 'D' logo, were distinctive, which would be relevant to the question of whether consumers would ascribe source significance to such features when used by competitors. And what ought to have been a decisive factor--the presence of the primary 'yd' brand on the defendant's shoe, which must have served to push out any source identifying role the overall configuration or sole pattern performed--was simply treated as a 'nonfunctional' element to be weighed against the presence of the functional elements. (198)

By focusing too much on functionality considerations, the Federal Court has thus distorted the test for trade mark use as it applies to non-traditional marks generally. This can be seen in more recent infringement cases, such as Coca-Cola Co v PepsiCo Inc [No 2] ('Coca-Cola v PepsiCo'), (199) involving a dispute over PepsiCo's use of an allegedly similar bottle shape to the famous 'contour' Coke bottle shape. A finding that PepsiCo had not used its bottle shape as a trade mark ought to have been straightforward, given the presence of prominent 'Pepsi' labelling. Instead, the Court asserted that because PepsiCo's bottle was 'distinctive' in appearance and consumers would associate it with PepsiCo, it was being used as a trade mark. (200) When combined with other recent decisions, a line of infringement cases thus seems to be emerging that should give rise to real concerns about the limited scope for third parties to engage in descriptive or decorative use of non-traditional signs under Australian law. (201)

V RECALIBRATING DISTINCTIVENESS AND TRADE MARK USE, AND RETHINKING THE SHAPE OF A FUNCTIONALITY EXCLUSION

The analysis in the previous three Parts has demonstrated that there are serious deficiencies in Australia's current approach to functional and nonfunctional non-traditional marks. In many respects, the problems stem from the legislature's failure in enacting the TMA to be clear as to how the issues of distinctiveness and functionality were to be prioritised and managed. This was a particular problem given that the position under the former law was one of hostility to the very idea that signs such as the colour or shape of goods could ever serve a trade mark purpose. In light of that history, and the lack of clarity in the TMA as to how, exactly, non-traditional signs were to be assessed, the judges in Philips v Remington (Appeal) and Kenman Kandy are hardly to be faulted for continuing to believe that functional signs were to be excluded from protection under the TMA. As the vast majority of countries around the world had come to recognise, there are very strong policy reasons for such an outcome. The problem in Australia was the lack of an obvious mechanism to give effect to such policy concerns. The results of the Federal Court's attempts to force functionality considerations into the distinctiveness and trade mark use doctrines have, as we have seen, caused significant problems with the operation of those doctrines more generally. Reform is long overdue.

This Part sets up a framework for unravelling the tangled doctrines of functionality, use and distinctiveness. The key reform that needs to occur is the adoption of an explicit functionality bar to registration. But before considering what form this bar should take in Australian law, it is worth considering the likely impact of such a proposal on the distinctiveness and trade mark use doctrines, and any associated changes that need to be made to those doctrines.

A Recalibrating Distinctiveness and Trade Mark Use to Deal with Non-Traditional Marks More Effectively

If functional signs are to be excluded from registration through a new statutory bar, this would take the pressure off both the distinctiveness and trade mark use doctrines as the mechanisms for dealing with the problems raised by such signs. More subtly, this should, in turn, force a re-evaluation of the distinctions that have been drawn in the case law on both distinctiveness and trade mark use between functional and non-functional signs, and improve the way the Office and courts manage non-functional signs.

Looking first at distinctiveness, the distinction made at the prima facie distinctiveness stage between functional signs on the one hand, and concocted or visually attractive signs on the other hand, (202) must be rethought. The courts and the Office have consistently asked the wrong question at the prima facie distinctiveness stage by focusing on the presence or otherwise of these non-functional elements, since this overlooks the fact that the 'ordinary signification' of a feature of shape, colour, or an aspect of packaging is simply that it is a feature of shape, colour, or an aspect of packaging, things that any trader might wish to use for non-trade mark purposes. The Australian test for prima facie distinctiveness has always had competition concerns at its heart, (203) and a revitalised approach to assessing the distinctiveness of nontraditional marks needs to ensure that these concerns are prioritised. And although such an approach would invariably lead to a finding that such marks are not prima facie distinctive, this does not in any way impose an insuperable barrier to the registration of such marks. If it is the case that the feature of shape, colour or packaging is particularly striking or visually attractive, this might well assist the owner in making its case that the mark has been used, and has come to be recognised, as a source identifier. But this is a factor that is relevant only to an assessment of acquired distinctiveness and not to the anterior question of the mark's inherent adaptation to distinguish.

The Full Federal Court or High Court should therefore overrule Kenman Kandy. The problem, however, is that neither Court is likely to have the opportunity any time soon. Remarkably few cases on the registrability of shape marks have ended up going to court, in part because the Kenman Kandy rule is so friendly to brand owners. As such, a more efficient response might be for the legislature to amend s 41 of the TMA to overturn the effect of Kenman Kandy. Building on suggestions made more than a decade ago by Mark Davison and Craig Smith, the legislature could do so by deeming that certain specified 'signs' such as shapes, aspects of packaging, colours, sounds and scent are never to be considered to be prima facie distinctive. (204) This would be an unusual provision in the sense that legislation around the world rarely singles out certain types of sign for this sort of treatment. However, it is a change that not only reflects the way in which such signs ought to be assessed under Australian law, but is also consistent with the approach taken by courts in other key jurisdictions. For example, the effect of the US Supreme Court's decisions in Qualitex Co v Jacobson Products Co Inc ('Qualitex') (205) and Wal-Mart Stores Inc v Samara Brothers Inc (206) is that single colours and product configuration, respectively, will be registrable only upon a showing of acquired distinctiveness. (207)

Legislative reform to achieve these ends would clearly be resisted by brand owners. But this could only be on the grounds that denying such parties the opportunity to register product features as trade marks on the basis of prima facie distinctiveness strips them of valuable legal protection against free-riders either before they have started to use such products, or during the period of time before those products' features have acquired distinctiveness. This merely begs the question of why trade mark law should intervene to provide this protection. Such owners would need to make an empirical case that there would be an underinvestment in the development of new product features as a result of any potential free-riding, (208) a particularly challenging case to make given that other intellectual property regimes, most notably the registered designs system, would seem to provide sufficient ex ante incentives for exactly this sort of product-feature development.

Two further legislative reforms would be desirable. One is the reintroduction of mandatory disclaimers over non-distinctive parts of marks. This would recognise the particular problems that arise where an applicant seeks to register a mark consisting of a combination of 'traditional' elements such as words or devices and 'non-traditional' elements such as shape or colour. While it still ought to be possible to find certain 'combination' marks prima facie distinctive, applicants should be required to disclaim any non-distinctive elements of such signs. This would ensure that the Register provides a clearer signal as to the more limited scope of the rights in such combination marks. It would also help avoid a case like Global Brand being litigated under the problematic assumption that the registered owner of a combination mark in fact owns rights that extend to prevent the use of similar design features when it has not established distinctiveness in such features themselves. (209)

The other legislative reform that ought to be considered is the removal of the distinction currently drawn in s 41 of the TMA between the two types of non-prima facie distinctive marks. As argued elsewhere, there are no longer sound justifications for allowing the registration of signs that are 'not sufficiently inherently adapted' to distinguish goods or services on the basis that they will become distinctive at some unspecified point in the future. (210) All non-prima facie distinctive marks raise competition concerns: they are marks that have been determined (or, as suggested above, should be deemed) to consist of signs that other traders would wish to use for the sake of their ordinary significations. They should be registrable only on the basis that they have acquired distinctiveness. Making this change would bring Australian law into alignment with the approach taken in the US, the EU and those countries that have modelled their trade mark laws on EU law, such as India, New Zealand and Singapore.

Turning to the infringement inquiry, a key benefit of a functionality bar to registration is that it would allow for a reconsideration of the scope of the 'trade mark use' requirement in s 120 of the TMA. More specifically, and similar to what ought to happen at the distinctiveness stage, there needs to be a re-evaluation of the problematic distinction that has been drawn between functional and non-functional signs in the test for infringement. It is vitally important that the courts move away from the simplistic approach taken in cases like Global Brand and Coca-Cola v PepsiCo, and think more carefully about when a shape, colour or aspect of packaging is being used decoratively or descriptively, rather than as a trade mark. Drawing this distinction is something that courts have managed successfully when considering two-dimensional marks, such as logos and slogans, particularly because the courts have been mindful of the full context of the use of those signs. That is, the fact that such signs have been used in conjunction with other branding elements has often been instrumental in courts' determinations that the logos or slogans were not serving a trade mark purpose, but were instead being used decoratively. An approach that gives more weight to such contextual factors in considering non-traditional subject matter, rather than one that assumes that a non-functional, visually striking feature points towards trade mark use, needs to be adopted. If this makes it more difficult for owners of nontraditional marks to succeed at this first stage of the infringement inquiry, this is entirely consistent with the way the trade mark use safeguard has long operated more generally. For example, it has long been recognised that even if an owner manages to register its descriptive word mark on the basis of factual distinctiveness, this does not give it unfettered rights over that term, as a third party's use of the sign for descriptive purposes will not be considered to be trade mark use. (211)

B Rethinking the Functionality Exclusion

The final issue that needs to be considered is how best to deal with the issue of functionality. Most Australian commentators who have touched on this issue have suggested that Australia should 'reintroduce' s 39 of the 1994 Act. (212) To reiterate, this would set up a bar to the registration of a trade mark that consists 'wholly or principally' of a shape or other characteristic that is 'possessed, because of their nature, by the goods' or that 'the goods must have if a particular technical result is to be obtained' (that is, drawing on the taxonomy set up in Philmac and described in Part II(B)(4), if the mark serves a 'natural' or a 'utilitarian' function). This would certainly be a major improvement on the current position. However, with very few exceptions, commentators have not sought to explain why the former s 39, as drafted, represents the optimal solution to the problems raised by functional marks. Nor have they considered whether Australia should look further afield, specifically to the approaches to functionality that have been adopted in the EU and the US, and critically assess whether one of these models, or perhaps a hybrid version of them, might provide a better way forward. (213) In addition, they have not considered whether other reforms, such as the introduction of mandatory disclaimers for functional parts of combination marks, ought to be considered. Such a consideration forms the basis of this final sub-Part.

1 The Competing European Union and United States Models

The EU model takes a similar approach to s 39 of the 1994 Act, in that it sets up a number of 'rules' as to when a mark will be refused registration on the grounds of functionality. The current EU bar contains exclusions for signs that consist 'exclusively' of 'the shape, or another characteristic' of goods that serve one of three functional purposes, (214) the first two of which are broadly coterminous with the former s 39. That is, the first two EU exclusions apply to a shape or characteristic 'which results from the nature of the goods themselves' (215) (ie serving a 'natural' function) and to a shape or characteristic of goods 'which is necessary to obtain a technical result' (ie serving a 'utilitarian' function). (216) The most important point of difference for present purposes is the third EU exclusion, which covers a shape or characteristic 'which gives substantial value to the goods',217 thus going beyond the scope of the former s 39.

In contrast, the US has taken what might be termed a 'standards' based approach to functionality, in that it excludes from registration a mark that 'comprises any matter that, as a whole, is functional', without further elaboration. (218) As such, it has been left to the courts to establish a set of principles for determining when a mark is 'functional'. Following a series of Supreme Court decisions, culminating in TrafFix Devices Inc v Marketing Displays Inc ('TrafFix') in 2001, it has been established that a feature of trade dress will be functional 'if it is essential to the use or purpose of the article or if it affects the cost or quality of the article', (219) a test set up primarily to deal with utilitarian functionality. This test does not impose a requirement of 'competitive necessity', in the sense that a feature will be functional only if other traders need to use the feature in question to compete effectively: rather, a feature can be functional even if viable design alternatives exist. (220) And even if a feature of trade dress is not 'essential to the use or purpose of the article or if it affects the cost or quality of the article', it can still be held functional if its exclusive use 'would put competitors at a significant non-reputation-related disadvantage'. (221) The Supreme Court has indicated that this second test is to be applied in cases of what it termed aesthetic functionality, and a finding that a mark is functional on this basis does require competitive necessity. (222)

It might be thought that there would be significant advantages for a country in Australia's position choosing to adopt a rule-based functionality bar. This is particularly the case given that the bar would operate in a primarily bureaucratic setting, where a set of rules might well provide examiners and hearing officers in the Office with a greater degree of clarity as to what is and is not to be excluded from registration than a broadly-expressed 'functionality' exclusion. It also needs to be remembered that the US standard is perhaps more rule-like in its operation, since it has been fleshed out by decades of case law and Registry-level decision-making; something that Australia lacks.

However, it is simplistic to conclude that a set of rules always generates more efficient outcomes than standards, even in a bureaucratic context, and a particular challenge with adopting a set of rules is making sure they are sufficiently well-drafted so that they are neither under- nor over-inclusive in operation. (223) In this regard, there are three significant problems with the rules contained in the former s 39 and the EU provision. Some of these problems could potentially be overcome with more careful drafting. However, a stronger argument can be made that the protean nature of 'functionality, and the difficulty of defining its boundaries ex ante, suggest that Australia might be better off adopting a general, US-style functionality exclusion.

2 The Under-Inclusiveness of the European Union Model

The first problem with the former Australian and current EU rule-based approach relates to the way in which utilitarian functionality is addressed. Under the second EU exclusion and the former s 39 there is a requirement that the feature of the goods be 'necessary,' or something the goods 'must have, for the achievement of a technical result. This language strongly suggests that the exclusion is to have a limited operation: that if there are other ways of achieving that technical result, the feature cannot be said to be necessary and the mark will not be excluded from registration. (224) In other words, it seems to set up the sort of 'competitive necessity' requirement that has been rejected by the US Supreme Court in cases of utilitarian functionality. Notwithstanding the language of the EU provision, the Court of Justice has shifted EU law closer towards the US position. A sign will be caught by the second EU exclusion if its 'essential characteristics perform a technical function, where to allow the mark to be registered would limit competitors' 'freedom of choice in regard to the technical solution they wish to adopt. (225) It was thought that there was 'nothing in the wording' of the exclusion that meant that the ground of refusal could be overcome by evidence of other shapes that could achieve the same technical result. (226) Turning the test of 'necessity' into one that simply looks to the existence of a causal relationship between the shape or characteristic and the technical outcome sought to be achieved is a desirable outcome in policy terms. It broadens the scope of the exclusion and ensures that a wider range of signs will be left free for third party use and also sets up a much more workable test in bureaucratic terms. (227) But it cannot be overlooked that such an approach glosses over the language of 'necessity' in the exclusion. The real concern with Australia adopting a utilitarian functionality bar expressed in the language of the EU exclusion or the former s 39 is therefore whether it would be under-inclusive in operation. It has to be accepted that there is a real risk that an Australian court would interpret the language of 'necessity' or 'must have' as setting up a true 'competitive necessity' test. If this were the case, then even the shape of Philips' three-headed rotary shaver, which so troubled the Full Federal Court in Philips v Remington (Appeal), would not be excluded on utilitarian functionality grounds, given evidence of the availability of alternative designs to achieve the same technical outcome.

A second problem of under-inclusiveness in the EU provision and the former s 39 is one of structure. Both provisions set up different functionality exclusions as alternatives: for example, the mark will be barred from registration if it consists exclusively of features that serve a natural function or if it consists exclusively of features that serve a utilitarian function. It is, however, entirely possible to conceive of a sign that has multiple essential characteristics, some of which have a technical effect, with others resulting from the nature of the goods. This is, in fact, how the characteristics of the four-finger Kit Kat chocolate bar were categorised in the UK. (228) In these cases, the Court of Justice has interpreted the language of the EU provision strictly, finding that a mark will be excluded from registration only if all its essential characteristics serve one of the three identified functions. (229) This approach permits the registration of marks that could accurately be described as 'functional' and, like the first problem, has the potential to undercut the competition policy that underpins the exclusions. (230)

3 The Challenge of 'Aesthetic Functionality'

The two problems identified above could possibly be overcome with more careful drafting of the exclusions. (231) However, the third problem I wish to explore is more fundamental, since it goes to the question of how far the functionality bar should extend beyond cases of natural and utilitarian functionality, and whether a legislative rule can accurately set the boundaries of what is 'functional.

In Australia, the drafters of the former s 39 (and the Working Party before it) did not consider there to be any need to deal with cases other than of natural and utilitarian functionality. (232) In contrast, the EU adopted an additional exclusion that applies where a shape or characteristic 'gives substantial value to the goods.' (233) According to the Court of Justice in its 2014 decision in Hauck GmbH & Co KG v Stokke A/S ('Hauck'), this covers those shapes and characteristics that provide 'artistic or ornamental value', (234) which is to be assessed by reference to consumer perception and taking into account
   the nature of the category of goods concerned, the artistic value
   of the shape in question, its dissimilarity from other shapes in
   common use on the market concerned, a substantial price difference
   in relation to similar products, and the development of a promotion
   strategy which focuses on accentuating the aesthetic
   characteristics of the product in question. (235)


The EU rule is understood as setting up a type of 'aesthetic functionality' exclusion. But it needs to be asked whether (a) it captures all of the circumstances in which such an exclusion ought to apply; and (b) it is drafted so broadly that it might unjustifiably prevent certain non-traditional signs from registration. That is, is such a provision both under- and over-inclusive?

Understanding the limitations of the EU provision requires some appreciation of the circumstances in which the aesthetic functionality doctrine has been applied in the US, and the policy factors that have animated the doctrine there. In one line of US aesthetic functionality cases, the registration of colour marks has been rejected, or the protection of colour marks denied, on the basis that the colour conveyed a recognised non-source identifying meaning (236) --what Mansfield J in Philmac termed an 'ornamental function'. (237) Examples of such marks include the colour black for outboard motors (making the goods appear smaller), (238) the shape and colours of frozen ice cream 'beads' (signifying various flavours) (239) and black used for floral packaging services (connoting formality and condolences). (240) In another line of cases, the key factor was that the marks consisted of features of goods that made them compatible with other products that consumers might have purchased. (241) Such marks included the appearance of various lighting fixtures that were designed to be compatible with a range of exterior and interior household finishes, (242) and a metal roof vent cover that resembled the shape of roof tiles and was designed to be 'unobtrusive when installed' alongside such tiles. (243) In both lines of cases, the policy reasons for denying registration or protection are clear: there are no obvious alternatives available to competitors wishing to use the signs to convey the precise meanings in question, or for compatibility purposes, and thus there is a clear 'competitive necessity' for such aesthetically or ornamentally functional signs to remain free for general use. (244)

It is much less obvious, however, how these sorts of aesthetically functional signs would fall to be considered under an EU-style exclusion that asks whether a shape or other characteristic gives 'substantial value' to the owner's goods. The precise rationale for this 'substantial value' exclusion is unclear, (245) but it makes most sense if viewed as a mechanism to ensure that the limited monopoly afforded by design law is not unduly extended. (246) That is, it seems to contemplate that a trader's goods might be sufficiently visually striking, such that consumers would be prepared to pay a premium for that design over and above what they would be willing to pay for goods without that design, with this surplus value not being attributable to the fact the design has come to be recognised as the trader's mark. While it might be appropriate for that surplus value to be captured through a design registration for a limited period of time, the exclusion from trade mark registration suggests that such capture should not be permanent. Putting this scenario to one side for the time being, it is less clear when a shape or characteristic might add 'substantial value'. It is possible to conceive of the 'compatibility' aspect of a shape or characteristic of goods as something that gives those goods an additional, non-reputation-related 'value'. Whether the same can be said for signs that convey non-source identifying information is less certain. But the bigger problem is that an exclusion based around the language of 'substantial value' does not seem particularly apt to manage the competition concerns raised by the attempted monopolisation of these types of aesthetically functional sign. Requiring decision-makers to look to the extent of the value that is added to the owner's goods in the meaning or compatibility cases not only raises difficult empirical questions of consumer preference and what they are willing to pay premiums for. It also steers attention towards a peripheral set of factors (such as price differences and promotion strategies) and away from what should really be the central question: whether other traders will be unfairly disadvantaged by allowing one party to monopolise signs for which ready alternatives do not exist. For this reason, there is a risk that a 'substantial value' exclusion will be under-inclusive.

Moving beyond the meaning or compatibility cases in the US, it is exceptionally difficult to pin down the scope of the aesthetic functionality doctrine. Other cases in which owners failed in unfair competition actions because their trade dress was held to be aesthetically functional show the challenges involved in confining the scope of the doctrine. In the early 1950s, it was held in Pagliero v Wallace China Co Ltd ('Pagliero') that if a design element was an 'important ingredient in the commercial success of the product' it would be unprotectable. (247) On this broad reasoning a pattern used on china was held to be aesthetically functional, (248) as was a visually striking shoe sole design that appealed to consumers and was a reason the shoes were purchased. (249) However, these older decisions are impossible to square with the modern test for aesthetic functionality set up by the Supreme Court in TrafFix that looks for a 'competitive necessity' and the absence of alternative designs. (250) This has substantially narrowed the scope of the doctrine, and led to disagreement among lower courts as to when, exactly, aesthetic features are competitively necessary. (251) And little guidance can be obtained by looking at outcomes of more recent cases where aesthetic functionality was found: in these cases it is sometimes hard to escape the conclusion that the doctrine was being used to deny protection to 'marks' that were in fact only decorative features and which had not acquired distinctiveness. (252)

Notwithstanding the vagueness of the aesthetic functionality doctrine in the US, the case law offers a vital lesson as to the problem with adopting an exclusion based on the wording of the EU exclusion. To ask whether the shape or characteristic provides 'substantial value' to the goods seems to set up a test similar to the one that has effectively been rejected in the US: that is, whether the shape or characteristic is an 'important ingredient in the commercial success of the product.' The forerunner to the 1988 Trade Marks Directive--the Benelux Uniform Law on Trade Marks, which came into force in 1971 but whose form was settled by the early 1960s (253)--excluded from registration shapes 'which affect [the] intrinsic value' of the goods. (254) Annette Kur has argued that the drafters of the Benelux Uniform Law on Trade Marks were likely to have been influenced by contemporaneous developments in the US and especially the Pagliero approach to aesthetic functionality. (255) That is, the Benelux exclusion, and the subsequent EU exclusion, seemed to be based on an understanding that if the commercial appeal of an article could be attributed in some way to its aesthetic appeal, the mark should be prevented from registration. Two noteworthy examples of the exclusion in application in EU law are the General Court's decision in Bang & Olufsen A/S v Office for Harmonisation in the Internal Market (Trade Marks and Designs) ('Bang & Olufsen'), which affirmed the rejection of registration of a mark consisting of the striking and unusual shape of an audio speaker, (256) and the decision of the Board of Appeal of what is now called the European Union Intellectual Property Office in Consorzio Origini v Vitra Collections AG, in which the Community Trade Mark registration for the shape of an Eames lounge chair was cancelled. (257) The concern here is that the EU test sets up a potentially over-broad exclusion that runs the risk of excluding too many distinctive nontraditional signs from protection.

The logical problem in seeking to exclude a non-traditional mark from registration based on a criterion of 'substantial value' is the difficulty in separating what might be termed 'aesthetic' value from 'trade mark' value. (258) To understand this, it needs to be remembered that an outright functionality bar is only needed to deal with the potential problems caused by a nontraditional mark becoming factually distinctive. If the test for acquired distinctiveness is applied in a robust manner, the owner will have needed to show that consumers have come to rely on the shape or characteristic as an indication of source. That is, while a shape or other characteristic of goods might have had a purely aesthetic value when those features were adopted, over time that value will come to be a mixture of aesthetic and trade mark (or source signification) value. To adapt the facts of Bang & Olufsen, if the striking and unusual shape of an audio speaker has come to acquire distinctiveness, consumers will likely wish to purchase the speaker both because they know the shape indicates trade source and because of its attractive appearance. (259) A test that asks whether the sign adds 'substantial value', but seeks to extricate any 'trade mark value', seems to ask an impossible question when aesthetic and trade mark value have become entwined. (260) This can be seen in some of the factors identified by the Court of Justice in Hauck that supposedly help to isolate the 'substantial value' added by the shape or characteristic. The shape's 'dissimilarity from other shapes in common use on the market concerned' (261) is just as relevant to whether the mark is factually distinctive and thus provides a 'trade mark' value. And where the mark has acquired distinctiveness, it will be unrealistic to look for 'a promotion strategy which focuses on accentuating the aesthetic characteristics of the product in question', (262) when such a strategy will also have accentuated those same characteristics in a way that emphasised that they were serving a source-indicating purpose. (263)

The above indicates a more fundamental problem with the EU test. Assuming that a feature of shape or characteristic adds both aesthetic and trade mark value to goods, why might the presence of the former justify an outright exclusion from registration? Why is it that the owner cannot capture the value of a key ingredient in the commercial success of its product? What gets lost in focusing on identifying a mark's 'value' is a careful assessment of the impact on competitors of allowing the monopolisation of the mark. Denying competitors the ability to copy a technically functional product feature that has subsequently acquired distinctiveness seems to raise competition concerns of a quite different magnitude compared with denying them the ability to copy a merely aesthetically pleasing, factually distinctive design. In the latter case it is less obvious that a competitor ought to be allowed to use the design, unless the availability of alternatives that consumers would be prepared to accept is so small that it can be said that there is a competitive need for the copying of the design feature or a closely similar variant. (264) Apart from the meaning or compatibility cases outlined above, it is hard to pin down exactly when such a competitive need might arise, but a test that is focused on competitive necessity ensures that at least the right questions are being asked. The problem with a test that focuses on 'substantial value' is that it runs the risk of excluding from protection any sufficiently aesthetically pleasing shape, even if doing so would only have the most marginal impact on competition.

4 Which Way Forward for Australia?

The discussion in the two preceding sections shows that Australia should exercise great caution before 'reintroducing' s 39 of the 1994 Act, or transposing the three EU functionality exclusions into Australian law. The most significant problem is with the third EU exclusion, which is both under-inclusive (in that it is poorly set up to deal with marks that serve an 'ornamental function') and over-inclusive (in that it has the potential to bar visually attractive and factually distinctive marks from registration without a proper inquiry into the extent of the competitive need for the use of such designs). While the problems raised by 'aesthetically functional' marks are real, they are hard to capture with a bright line rule.

The better approach for Australia would therefore be to adopt a US-style, generally-worded functionality bar to registration and allow the courts and the Office to determine the circumstances in which marks are to be rejected. If it is thought that such an open-ended provision would provide too little guidance to the courts and the Office, such guidance could be provided in various ways. The Explanatory Memorandum to the Bill introducing the change could spell out that the bar is intended to exclude marks that serve a natural, utilitarian or an aesthetic function, the last of which includes an ornamental function. It could make clear that, in line with the US Supreme Court's approach in Qualitex and TrafFix, it is only in cases of aesthetic functionality that courts should apply a test based on competitive necessity, whereas a broader standard applies in considering natural or technical functionality. It could also clarify that a mark that consists of a number of characteristics that serve different functional purposes can be excluded on the basis that it is functional. In addition, non-binding legislative notes making some of these points (along the lines of the notes to s 41 of the TMA indicating the types of marks that are not inherently adapted to distinguish goods or services) could be added to the new section introducing the functionality exclusion. What is key is that some direction can easily be provided to assist the courts and the Office to define the core scope of the functionality exclusion, leaving them to outline the boundaries of the exclusion on a case-by-case basis.

A related reform that ought to be considered expands on my earlier recommendation for the reintroduction of mandatory disclaimers. Applicants for registration should be required to disclaim not only the non-distinctive elements of combination marks, but also any functional elements of such marks. (265) The reason for this is that a bar framed in terms of 'functionality' might be able to be circumvented by an applicant for registration of a mark that consists, by way of example, of a functional shape of a product bearing a distinctive brand name or logo. While in some cases it might be that the brand name or logo is so obscure in the representation that it could be ignored in the assessment of the registrability of the mark, (266) in many cases the presence of the brand name would ensure that the mark could be registered on the basis of prima facie distinctiveness. (267) A provision that required the applicant to disclaim the exclusive use of the non-distinctive features of such a mark might well be useful in this situation, but not if the applicant were able to demonstrate that the features had themselves acquired distinctiveness. The end result is that it is possible to imagine 'hybrid' marks, featuring distinctive and functional features, being entered on the register, with little guidance being afforded as to the scope of rights in such marks. This is of particular concern, given the way in which the Federal Court applied the test for 'trade mark use' in Global Brand, where the registered marks were of this 'hybrid' variety. In that case, Sundberg J did not seek to disaggregate the functional and distinctive elements of either the plaintiff's or defendant's marks, instead finding that the defendant had used its entire shoe shapes and entire sole shapes as trade marks, on the basis that the nonfunctional elements 'overwhelmed' the functional elements. This approach has the potential to afford over-broad protection for owners of hybrid marks. A provision that seeks to ensure that the functional features of such marks are identified ex ante and that such features are filtered out of the infringement analysis, therefore, has much to recommend it. (268)

VI CONCLUSION

Functionality has played an attenuated role in Australian trade mark law for too long. The government, at least until 1994, and the Federal Court, in the first few years of the TMA, appreciated the serious harm that would result from permitting the registration of technically functional non-traditional marks. But in the absence of a clear legislative bar to the registration of such marks in the TMA, the Court was forced to rely on related doctrines to screen out functional marks from the trade marks system. Australia's de facto functionality exclusion might appear to be a success--the Office has done a fair job of keeping functional marks off the Register and the courts have tended to interpret the infringement test so as to give competitors breathing room to make use of functional signs. However, forcing functionality concerns into the tests for distinctiveness and trade mark use has not only resulted in a de facto exclusion that is less efficient than a standalone exclusion, but it has distorted the operation of the distinctiveness and trade mark use tests as they apply to non-traditional marks more generally. In addition, it has meant that the courts and the Office have never been required to identify the full range of circumstances in which a mark might be 'functional' or think carefully about the different types of impact on competitors that result from permitting the registration of marks that serve a natural, technical or aesthetic function.

The time has come for Australia to take this task seriously. It needs to recalibrate the relationship between functionality, distinctiveness and use, liberalising the operation of the last two and detaching functionality so that it has a standalone sphere of operation. A functionality bar to registration along the lines of that originally proposed by the Australian government in 1994, or the current EU bar, would be an advance on the current position. But an open-ended and broadly-worded functionality exclusion, along the lines of the US provision, coupled with a new requirement that an applicant for registration disclaim any non-distinctive and/or functional elements of combination marks, represents a better way forward. Based on the international experience with managing functional marks, interpreting and applying such a US-style exclusion will not be straightforward. But, given the overarching importance of ensuring that present and future competitors in markets are not unduly hampered in their ability to access and use functional features of shape, packaging, colour and similar signs, it is a challenge that is worth embracing.

(1) Marrakesh Agreement Establishing the World Trade Organization, opened for signature 15 April 1994, 1867 UNTS 3 (entered into force 1 January 1995) annex 1C ('Agreement on Trade-Related Aspects of Intellectual Property Rights') art 15(1) (emphasis added) ('TRIPS Agreement').

(2) See generally Qian Zhan, 'The International Registration of Non-Traditional Trademarks: Compliance with the TRIPS Agreement and the Paris Convention' (2017) 16(1) World Trade Review 111, 113-19.

(3) See generally Robert Burrell, Huw Beverley Smith and Allison Coleman, 'Three-Dimensional Trade Marks: Should the Directive Be Reshaped?' in Norma Dawson and Alison Firth (eds), Trade Marks Retrospective (Sweet & Maxwell, 2000) 139, 146-58.

(4) Ibid 156-7. See also below nn 40-1 and accompanying text.

(5) These examples are registered trade marks in Australia: respectively, AU Trade Mark No 815934, filed on 2 December 1999 (Registered on 25 August 2003); AU Trade Mark No 1054076, filed on 6 May 2005 (Registered on 15 January 2007); AU Trade Mark No 1595348, filed on 5 December 2013 (Registered on 13 February 2015).

(6) See, respectively, AU Trade Mark No 1386911, filed on 5 October 2010 (Registered on 14 December 2012); AU Trade Mark No 1218000, filed on 7 January 2008 (Registered on 8 February 2010).

(7) See generally Graeme B Dinwoodie, 'The Death of Ontology: A Teleological Approach to Trademark Law' (1999) 84(4) Iowa Law Review 611.

(8) This is a qualification to art 15(1) of the TRIPS Agreement permitted by art 15(2). Article 15(2) provides that art 15(1) 'shall not be understood to prevent a Member from denying registration of a trademark on [grounds other than that the mark is not capable of distinguishing], provided that they do not derogate from the provisions of the Paris Convention (1967)', this being the Paris Convention for the Protection of Industrial Property, opened for signature 14 July 1967, 828 UNTS 305 (entered into force 26 April 1970) ('Paris Convention'). It might, on first glance, appear that a bar on the registration of functional marks that extends to marks that have been registered in other Paris Convention countries conflicts with art 6quinquies of the Paris Convention. This is because art 6quinquies(A)(1) provides that a trade mark 'duly registered in the country of origin shall be ... protected as is in the other countries of the Union, subject to the reservations indicated in this Article'. Of the three reservations in art 6quinquies(B), only the reservation for marks that are 'contrary to . public order' would seem to be relevant: see Annette Kur, 'Too Pretty to Protect?: Trade Mark Law and the Enigma of Aesthetic Functionality' (Research Paper No 11-16, Max Planck Institute for Intellectual Property and Competition Law, 28 September 2011) 18-21 <https://papers.ssrn.com/sol3/papers.cfm?abstract_id=1935289>, archived at <https://perma.cc/KDJ9-NWMA>. However, it has been argued convincingly that art 6quinquies(A)(1) does not apply at all in this context. It sets up only a narrow requirement that marks be accepted in the same form as their original registration (for example, in the same script) but does not otherwise derogate from art 6(1), which allows a country to set up its own conditions for registration such as a functionality bar: see Ng-Loy Wee Loon, 'Special Restrictions on Registration of Product Shape Marks: A Breach of International Law?' in Irene Calboli and Martin Senftleben (eds), The Protection of Non-Traditional Marks (Oxford University Press, forthcoming).

(9) First Council Directive 89/104/EEC of 21 December 1988 to Approximate the Laws of the Member States Relating to Trade Marks [1989] OJ L 40/1, art 3(1)(e) ('1988 Trade Marks Directive'). See also Council Regulation (EC) No 40/94 of 20 December 1993 on the Community Trade Mark [1994] OJ L 11/1, art 7(1)(e), setting up the same rule for European Community trade marks. The Court of Justice of the European Union has recently determined that the reference to 'shape' in art 3(1)(e)(iii) of Directive 2008/95/EC of the European Parliament and of the Council of 22 October 2008 to Approximate the Laws of the Member States Relating to Trade Marks [2008] OJ L 299/25 ('2008 Trade Marks Directive') (the 2008 codification of the 1988 Trade Marks Directive (n 9)) is limited to three-dimensional properties of goods and does not encompass other characteristics, such as a colour that is applied to a specific part of the goods: Louboutin v Van Haren Schoenen BV (Court of Justice of the European Union, C-163/16, ECLI:EU:C:2018:423, 12 June 2018) [24].

(10) See, eg, Hauck GmbH & Co KG v Stokke A/S (Court of Justice of the European Union, C-205/13, ECLI:EU:C:2014:2233, 18 September 2014) [18]-[19] ('Hauck').

(11) Trade Marks Act (Singapore, cap 332, 1998) s 7(3).

(12) Trade Marks Act 1999 (India) s 9(3).

(13) [phrase omitted] [Trademark Law of the People's Republic of China] (People's Republic of China) National People's Congress, Order No 59, 27 October 2001, art 12.

(14) Trade Marks Ordinance (Hong Kong) cap 559, s 11(3).

(15) Intellectual Property Act (Sri Lanka) Act No 36 of 2003, art 103(1)(a).

(16) Trade Marks Act 1993 (South Africa) s 10(5).

(17) Ibid. An even broader exclusion can be seen in South Korea: Trademark Act (Republic of Korea), as at 21 March 2017, art 34(1)(15) [tr Korea Legislation Research Institute, 'Trademark Act', Statutes of the Republic of Korea (Web Page, 2017) <http://elaw.klri.re.kr/ eng_service/lawView.do?hseq=42777&lang=ENG>, archived at <https://perma.cc/8E6HQB94>] (preventing from registration '[a]ny trademark consisting solely of the three-dimensional shape, color, combination of colors, sound, or smell, which is essential (in cases of services, referring to cases essential for the use and purpose thereof) to secure the function of goods intended to obtain trademark registration or of the package of such goods').

(18) Trademark Act of 1946, 15 USC [section] 1052(e)(5) (2016) (emphasis added). The reason for the belated inclusion of a functionality bar to registration is likely to be that in enacting the Trademark Act of 1946, Congress never intended that trade dress would be registrable on the Principal Register. It was only following the Supreme Court's acceptance in Qualitex Co v Jacobson Products Co Inc, 514 US 159 (1995) ('Qualitex') that 'almost anything at all that is capable of carrying meaning' is capable of registration on the Principal Register: at 162 (Breyer J for the Court), and subsequent case law preventing functionality from being raised as a means of challenging a registration, that Congress intervened to introduce the functionality bar to registration. See generally Glynn S Lunney Jr, 'The Trade Dress Emperor's New Clothes: Why Trade Dress Does Not Belong on the Principal Register' (2000) 51(6) Hastings Law Journal 1131.

(19) Directive (EU) 2015/2436 of the European Parliament and of the Council of 16 December 2015 to Approximate the Laws of the Member States Relating to Trade Marks [2015] OJ L 336/1 ('2015 Trade Marks Directive').

(20) Council Regulation (EC) No 207/2009 of 26 February 2009 on the Community Trade Mark [2009] OJ L 78/1 ('EU Trade Mark Regulation'), as amended by Regulation (EU) 2015/2424 of the European Parliament and of the Council of 16 December 2015 [2015] OJ L 341/21, art 1(9).

(21) 2008 Trade Marks Directive (n 9) art 3(1)(e).

(22) 2015 Trade Marks Directive (n 19) art 4(1)(e) (emphasis added); EU Trade Mark Regulation (n 20) art 7(1)(e) (emphasis added). A similar shift has occurred in Japan. Before April 2015, a mark 'consist[ing] solely of a three-dimensional shape of goods or their packaging which is indispensable for such goods or their packaging to properly function' was excluded from registration: Trademark Act (Japan) Act No 127 of April 13, 1959, as at 1 April 2012, art 4(1)(xviii) ['Trademark Act (Act No 127 of April 13, 1959, as amended up to Act No 63 of 2011)', World Intellectual Property Organization (Web Page) <http://www.wipo.int/wipolex/ en/text.jsp?file_id=299905>, archived at <https://perma.cc/4VHN-7QT6>]. With effect from 1 April 2015, this article was extended to apply to marks 'consist[ing] solely of features ... that are naturally provided to goods': Trademark Act (Japan) Act No 127 of April 13, 1959, as at 1 April 2015, art 4(1)(xviii) (emphasis added) ['Trademark Act (Act No 127 of April 13, 1959, as amended up to Act No 36 of May 14, 2014)', World Intellectual Property Organization (Web Page) <http://www.wipo.int/wipolex/en/text.jsp?file_id=369104>, archived at <https://perma.cc/57F4-7JJF>].

(23) See Patricia Loughlan, 'The Concept of "Sign" in Australian Trade Mark Law' (2005) 16(2) Australian Intellectual Property Journal 95, 101-2.

(24) See generally Kimberlee Weatherall, 'The Emergence and Development of Intellectual Property Law in Australia and New Zealand' in Rochelle C Dreyfuss and Justine Pila (eds), The Oxford Handbook of Intellectual Property Law (Oxford University Press, 2018) 291.

(25) Trade Marks Act 1995 (Cth) s 6 (definition of 'sign') ('TMA').

(26) Ibid s 41.

(27) Other countries without functionality bars to registration include Israel, Malaysia and New Zealand.

(28) TMA (n 25) s 120.

(29) It is worth noting that, at this time, Australian courts had left open the possibility that actions for passing off, and for breach of the prohibition on engaging in 'misleading or deceptive conduct' then contained in s 52 of the Trade Practices Act 1974 (Cth), were potentially available in cases where defendants had copied the plaintiff's trade dress or product configuration, even if these served technical functions. However, it was relatively easy for defendants to escape liability in these situations: courts were reluctant to find that plaintiffs had acquired sufficient reputation in their trade dress or product configurations, and if defendants used sufficiently clear labelling to differentiate their similarly-shaped goods, courts would not find that a misrepresentation had occurred. See generally Parkdale Custom Built Furniture Pty Ltd v Puxu Pty Ltd (1982) 149 CLR 191 and, for the equivalent approach in the United Kingdom in a passing off case, see Reckitt & Colman Products Ltd v Borden Inc [1990] 1 WLR 491.

(30) Trade Marks Act 1955 (Cth) s 6(1) (definition of 'trade mark') ('1955 Act').

(31) Ibid (definition of 'mark').

(32) Smith Kline & French Laboratories (Australia) Ltd v Registrar of Trade Marks (1967) 116 CLR 628, 639 (Windeyer J) ('Smith Kline & French').

(33) Ibid 639, quoting Charles Goodall & Son Ltd v John Waddington Ltd (1924) 41 RPC 658, 668 (Sargant LJ).

(34) Smith Kline & French (n 32) 640. Despite this decision, it was possible to point to a number of subsequent registrations consisting of colour combinations (such as the appearance of copper and gold on Duracell's batteries). However, it was also recognised that it would be difficult to register colour marks, given that most would lack distinctiveness. See DR Shanahan, Australian Law of Trade Marks and Passing Off (Law Book, 2nd ed, 1990) 14.

(35) Re Coca-Cola Co [1986] 1 WLR 695.

(36) Shanahan (n 34) 14-15.

(37) The working party consisted of members of the Trade Marks Office, the Law Council of Australia and the Institute of Patent Attorneys: Gail E Evans, Lawmaking under the Trade Constitution: A Study in Legislating by the World Trade Organization (Kluwer Law International, 2000) 133-4.

(38) See generally ibid 133-8.

(39) Possible Changes to the Australian Trade Marks Legislation (Discussion Paper, July 1991) 3-6.

(40) Ibid 6: 'It should ... be a requirement, if the sign constitutes some element of the shape or the packaging of the product, for the applicant to demonstrate that that element is not necessary for the proper functioning, or does not result from the nature ... of the product or its packaging'.

(41) Ibid: 'A sign would not be considered prima facie capable of distinguishing if it consisted of something that any trader in similar goods may need to use in producing or packaging those goods'.

(42) Ibid 7.

(43) Working Party to Review the Trade Marks Legislation, Recommended Changes to the Australian Trade Marks Legislation (Commonwealth of Australia, 1992) 38 (Recommendation 1B).

(44) Ibid 45 (Recommendation 6A(4)).

(45) Ibid 46 (Recommendation 6A(8)).

(46) The Act was said to reflect an acceptance of all of the Working Party's unanimous recommendations: Commonwealth, Parliamentary Debates, House of Representatives, 18 October 1994, 2225 (John Moore).

(47) Trade Marks Act 1994 (Cth) s 5 (definition of 'sign') ('1994 Act').

(48) Ibid s 16.

(49) Ibid s 39.

(50) TMA (n 25) s 6 (definition of 'sign') (adding 'scent' as a further example).

(51) Ibid s 17.

(52) Ibid s 41. Under the 1955 Act (n 30), it had been held that a mark that was not to any extent inherently adapted to distinguish the specified goods or services could never be considered to be capable of distinguishing, even if evidence could be adduced to show that the mark had acquired distinctiveness in fact (that is, the mark needed to be 'distinctive in law' as well as in fact): Clark Equipment Co v Registrar of Trade Marks (1964) 111 CLR 511, 513 (Kitto J) ('Clark Equipment'). Section 40 of the 1994 Act (n 47) seemed to leave open the possibility that a court or the Office could continue to impose a 'distinctiveness in law' requirement and prevent factually distinctive marks from registration. This possibility was removed in s 41(6) of the TMA (n 25), as enacted, which made clear that a mark that is not to any extent inherently adapted to distinguish the specified goods or services can be registered on a showing of acquired distinctiveness.

(53) Commonwealth, Parliamentary Debates, Senate, 30 March 1995, 2589-90 (Nick Sherry); Commonwealth, Parliamentary Debates, House of Representatives, 27 September 1995, 1911 (Michael Lee).

(54) It might appear that s 40, which provides that an application for registration can be rejected if the mark is incapable of being represented graphically, would also play a significant role. However, this ground has never been interpreted as setting up a requirement of 'adequate' representation. Cf the position under EU law: Sieckmann (C-273/00) [2002] ECR I-11737 (on olfactory marks); Shield Mark BV v Kist (C-283/01) [2003] ECR I-14313 (on sound marks). Provided the Australian application complies with the Office's bureaucratic requirements for representation of non-traditional marks, s 40 will be satisfied: see generally Robert Burrell and Michael Handler, Australian Trade Mark Law (Oxford University Press, 2nd ed, 2016) 18-21, 81-4.

(55) See Janice Luck, 'Distinctiveness, Deceptive and Confusing Marks under the Trade Marks Act 1995' (1996) 7(2) Australian Intellectual Property Journal 97, 100. It is worth noting that when New Zealand came to update its trade mark laws, its Trade Marks Bill 2001 (NZ) initially contained the transposed EU functionality bar. This was, however, rejected by the government's Commerce Committee, in part because it was thought that such a bar might 'unduly restrict and prevent registration of shapes that are distinctive': Commerce Committee, Parliament of New Zealand, Trade Marks Bill (Commentary, 8 March 2002) 4 <http://www.nzlii.org/nz/legis/hist_bill/tmb20021422131.pdf>, archived at <https://perma.cc/95YF-X9BV>. The New Zealand Court of Appeal has since accepted that a mark that is at least 'largely functional' can be registered if it is distinctive: Fredco Trading Ltd v Miller (2006) 11 TCLR 751 ('Fredco'). See generally Rob Batty, 'Is New Zealand's Trade Marks Law out of Shape?' (2011) 33(5) European Intellectual Property Review 281.

(56) Intellectual Property Committee, Law Council of Australia, 'Response to Working Party Paper: Possible Changes to the Australian Trade Marks Legislation' (Document, February 1992) 2-3; Law Council of Australia, 'Trade Marks Bill 1994 Exposure Draft: Law Council of Australia Recommendations' (Document, 31 August 1994).

(57) That is, words in a statute are to be given some effect, such that the absence of such words in amending legislation is significant: see Dennis C Pearce and Robert S Geddes, Statutory Interpretation in Australia (LexisNexis Butterworths, 8th ed, 2014) 62-4. Pearce and Geddes also discuss the ramifications of looking to repealed provisions of former legislation in interpreting existing legislation: at 121-4.

(58) See also Loughlan (n 23) 103; Andrew Stewart et al, Intellectual Property in Australia (LexisNexis Butterworths, 6th ed, 2018) 705-7.

(59) See Philmac Pty Ltd v Registrar of Trade Marks (2002) 126 FCR 525 ('Philmac').

(60) The Australian cases, and some of their problems, have been discussed extensively by commentators, with a particular focus on shape marks: see, eg, Megan Richardson, 'Shape Trade Marks in Australian Courts' (2001) 12(1) Australian Intellectual Property Journal 5; Janice Luck, 'The Registrability of Shapes of Goods as Trade Marks: A Commentary on This Part of the Full Federal Court's Judgment in Philips v Remington' (2001) 12(1) Australian Intellectual Property Journal 12 ('The Registrability of Shapes of Goods'); Julia Baird, 'The Registrability of Functional Shape Marks' (2002) 13(4) Australian Intellectual Property Journal 218; Julia Baird, '"This Mark Is So Attractive--It Should Be Free for All to Use!": An Australian Perspective on Functional Shape Marks' (2003) 52 Intellectual Property Forum 26; Trevor Stevens, 'The Protection of Trade Dress and Color Marks in Australia' (2003) 93(6) Trademark Reporter 1382; Jani McCutcheon, 'How Many Colours in the Rainbow?: The Registration of Colour Per Se under Australian Trade Mark Law' (2004) 26(1) European Intellectual Property Review 27; Jani McCutcheon, 'Monopolised Product Shapes and Factual Distinctiveness under s 41(6) of the Trade Marks Act 1995 (Cth)' (2004) 15(1) Australian Intellectual Property Journal 18 ('Monopolised Product Shapes'); Jani McCutcheon, 'The Registration of Sounds and Scents as Trade Marks under Australian Law' [2004] (2) Intellectual Property Quarterly 138 ('The Registration of Sounds and Scents'); Mark Davison, 'Shape Trade Marks: The Role and Relevance of Functionality and Aesthetics in Determining Their Registrability' (2004) 15(2) Australian Intellectual Property Journal 106; Loughlan (n 23); Craig Smith, 'Trade Mark Protection for Product Shape: Where Does the Public Interest Lie?' (2005) 63 Intellectual Property Forum 32 ('Trade Mark Protection for Product Shape'); Geoffrey Levy, 'The Registrability of Functional Shapes under the Trade Marks Act 1995: A Continuing Issue?' (2009) 77 Intellectual Property Forum 20; Ben McEniery, 'Trade Marks for the Design and Layout of Retail Premises' (2014) 24(3) Australian Intellectual Property Journal 167.

(61) (1999) 91 FCR 167 ('Philips v Remington (Trial)').

(62) Remington did not seek to raise the argument that the mere passage of the TMA (n 25) could not have served to enlarge the rights given to Philips in a two-dimensional device mark registered under the 1955 Act (n 30), without Philips establishing that it could now in fact register the three-dimensional version of its device as a shape mark. Such an argument later found favour in the High Court of Singapore: Nation Fittings (M) Sdn Bhd v Oystertec plc [2006] 1 SLR(R) 712, 726-8 (Andrew Phang Boon Leong J).

(63) Philips v Remington (Trial) (n 61) 173-6 [16]-[26], 183 [47].

(64) The Shell Co of Australia Ltd v Esso Standard Oil (Australia) Ltd (1963) 109 CLR 407, 425 (Kitto J).

(65) Philips v Remington (Trial) (n 61) 183-4 [49].

(66) Koninklijke Philips Electronics NV v Remington Products Australia Pty Ltd (2000) 100 FCR 90 ('Philips v Remington (Appeal)').

(67) Ibid 93 [1].

(68) Ibid 102-4 [14]-[16].

(69) Ibid 101-2 [12], 105 [18].

(70) Ibid 103 [15].

(71) Ibid.

(72) Ibid. For discussion of how Burchett J's interpretation might involve denying registration to a broader range of signs than contemplated by both the Working Party and s 39 of the 1994 Act (n 47), see Luck, 'The Registrability of Shapes of Goods' (n 60) 18.

(73) (2002) 122 FCR 494 ('Kenman Kandy').

(74) Ibid 509 [45].

(75) Ibid 528-9 [134]-[137]; see also at 513-14 [62] (Lindgren J).

(76) Ibid 529 [137] (emphasis added).

(77) Ibid 529-30 [137]; see also at 509 [43] (French J).

(78) My discussion in this section draws on arguments made in Burrell and Handler, Australian Trade Mark Law (n 54) 77-8, 135-6.

(79) Kenman Kandy (n 73) 529 [137] (Stone J).

(80) See also Davison (n 60) 107 n 6.

(81) In Kenman Kandy (n 73), Lindgren J raised much the same question: Was the inherent form of the goods 'confectionery', or '[b]ug-shaped items of confectionery'?: at 525 [112].

(82) For extensive discussion of the consideration of the 'right to copy' and the 'need to copy' in the utilitarian functionality doctrine under US law, see Mark P McKenna, '(Dys)functionality' (2011) 48(4) Houston Law Review 823; Robert G Bone, 'Trademark Functionality Reexamined' (2015) 7(1) Journal of Legal Analysis 183.

(83) McCutcheon has noted a more technical reason why the severability doctrine ought not to be used as a mechanism to exclude functional signs from the definition of a trade mark in s 17 of the TMA (n 25). This definition refers to 'a sign used, or intended to be used, to distinguish goods or services'. The severability doctrine under the former law hinged on the fact that a 'mark' needed, by definition, to be 'used or proposed to be used in relation to goods for the purpose of indicating ... a connexion in the course of a trade': 1955 Act (n 30) s 6(1) (definition of 'trade mark') (emphasis added). The absence of the 'in relation to' requirement in the current s 17 definition suggests that a sign that is inseparable from the goods, however this is defined, can still potentially be 'used', and thus meet this definition: McCutcheon, 'The Registration of Sounds and Scents' (n 60) 154-6.

(84) See, eg, Justin Hughes, 'Cognitive and Aesthetic Functionality in Trademark Law' (2015) 36(4) Cardozo Law Review 1227, 1270-3. Cf McCutcheon, 'Monopolised Product Shapes' (n 60).

(85) Philips v Remington (Trial) (n 61) 204 [110], 207 [119] (Lehane J).

(86) Kenman Kandy (n 73) 509 [43], [45] (French J), 529-30 [137] (Stone J).

(87) TMA (n 25) s 41(5)-(6), as enacted.

(88) See Davison (n 60) 108 on distinctiveness being a question of fact, not policy. See above n 52 on how the TMA (n 25) no longer contains a 'distinctiveness in law' barrier to registration.

(89) For example, in its June 2002 decision in Koninklijke Philips Electronics NV v Remington Consumer Products Ltd (C-299/99) [2002] ECR I-5475, the Court of Justice held that the functionality bar in the EU 1988 Trade Marks Directive (n 9) was 'intended to prevent the protection conferred by the trade mark right from being extended, beyond signs which serve to distinguish a product or service from those offered by competitors, so as to form an obstacle preventing competitors from freely offering for sale products incorporating such technical solutions or functional characteristics': at I-5516 [78] (emphasis added). This point had also been made in contemporaneous commentary: see, eg, William P Kratzke, 'The Supreme Court and Trade Dress: A Short Comment' (2001) 24(1) Hastings Communications and Entertainment Law Journal 73, 107: 'Distinctiveness does not preclude the possibility that a product feature is functional.'

(90) Philips v Remington (Trial) (n 61) 188-9 [66].

(91) Luck, 'The Registrability of Shapes of Goods' (n 60) 18.

(92) This criticism was made by commentators shortly after the decision: see, eg, Baird, 'The Registrability of Functional Shape Marks' (n 60) 225-6; McCutcheon, 'Monopolised Product Shapes' (n 60) 29-30; Davison (n 60) 108-9.

(93) The key Office decision that pre-dated Philips v Remington (Trial) (n 61) was Re Application by Chocolaterie Guylian NV (1999) 46 IPR 201, which involved an attempt to register a seashell shape for chocolate. The decision is noteworthy for its recognition that a non-technically functional shape might have 'pleasing design and will accordingly be bought ... largely on the success ... of [its] configuration. Where an aspect of shape is designed to capture market appeal that shape must then be regarded as functional': at 206 (Deputy Registrar of Trade Marks Hardie). This is a view that arguably reflects the 'shape which gives substantial value to the goods' ground of rejection under EU law: see below Part V(B)(3).

(94) Clark Equipment (n 52) 514. That this test applied under s 41 of the TMA (n 25) was first confirmed by the Federal Court in Blount Inc v Registrar of Trade Marks (1998) 83 FCR 50.

(95) See Michael Arblaster, 'Shape as a Trade Mark' (2000) 40 Intellectual Property Forum 30, 34-5 (on the Office's practices). Michael Arblaster was the Acting Registrar of Trade Marks at the time.

(96) See below Part III(A) on Office practice. The Office's examination practice has never been set up to look to the issue of whether the 'thing' sought to be registered does not comply with the definition of a 'trade mark' in s 17 of the TMA (n 25). For the position as at July 2018, see IP Australia, Trade Marks Office Manual of Practice and Procedure (Web Page, July 2018) pt 1.4 <http://manuals.ipaustralia.gov.au/trademarks/trade_marks_examiners_manual.htm>, referring to the Office's Product Quality Standards ('PSQ'). PSQ 2 and 3 show that only the grounds of refusal in ss 39-44 and a limited number of other issues (such as classification of the goods and services, and the adequacy of the representation of the mark) will be considered in examination. Further, in the current Statement of Grounds and Particulars form that opponents are required to lodge in opposition proceedings there is no facility for opponents to raise non-compliance with s 17 as a potential opposition ground: IP Australia, Statement of Grounds and Particulars (Web Document) <https://www.ipaustralia.gov.au/sites/g/files/ net856/f/tm00034_1116.pdf>, archived at <https://perma.cc/T29N-CCCF>. The Federal Court has recently confirmed that under s 57 opponents to the registration of standard trade marks may raise only the specified grounds under which an application may be rejected at the examination stage, namely those grounds contained in s 39 and ss 41-44, and reg 4.15A of the Trade Marks Regulations 1995 (Cth): Frucor Beverages Ltd v The Coca-Cola Co [2018] FCA 993, [99], [136]-[137] (Yates J) ('Frucor'). For background to this finding in Frucor, see Burrell and Handler, Australian Trade Mark Law (n 54) 46-7.

(97) Philmac (n 59) 527 [2] (Mansfield J).

(98) Ibid 535 [27] (Mansfield J).

(99) Ibid 540-3 [42]-[51].

(100) Ibid 546 [65]. The third function was, in fact, called an 'economic function'. This has been altered to indicate more clearly the type of function that Mansfield J sought to describe, and to differentiate it from the 'utilitarian function'.

(101) Ibid 548-51 [74]-[80].

(102) Ibid 543 [53].

(103) It is notable that Mansfield J referred to the Full Court's decisions in Philips v Remington (Appeal) (n 66) and Kenman Kandy (n 73) in the context of a discussion of the factors that would make a colour lack inherent adaptation to distinguish: ibid 544 [54].

(104) It would also seem to reflect the law on functionality and distinctiveness in New Zealand: see Fredco (n 55).

(105) See TMA (n 25) ss 33(4), 55. See also s 38, where the Registrar can revoke acceptance of an application for registration on the basis that s 41 should have been raised in examination.

(106) The 22 decisions are: Re Application by Chocolaterie Guylian NV (n 93) (seashell shape of chocolate); Re Application by Adcock (2000) 48 IPR 389 (shape of a traffic barrier); Re Application by Effem Foods Pty Ltd (2000) 49 IPR 139 (basic jar shape); Re Kenman Kandy Australia Pty Ltd (2000) 50 IPR 491 (bug shape of confectionery); Saramar LLC [2000] ATMO 117 (air freshener bottle shape); Dualit Ltd [2001] ATMO 6 (toaster shape); Re Application by Reemtsma Cigarettenfabriken GmbH (2001) 52 IPR 449 (shape of cigarette packet holder); Re Reemtsma Cigarettenfabriken GmbH [2002] AIPC [paragraph] 91-775 (shape and appearance of cigarette packet holder); Re Application by Conewich Enterprises Ltd Partnership (2002) 59 IPR 558 (cone-shaped bread); Re Australiana Group Holdings Pty Ltd [2003] AIPC [paragraph] 91-909 (basic bottle shape); Application by Multix Pty Ltd (2004) 64 IPR 128 (oval shape of an oven tray); Aldi Stores (A Limited Partnership) v Societe Des Produits Nestle SA [2008] AIPC [paragraph] 92-313 (shape of a four-finger Kit Kat chocolate bar); Re Gibson (2010) 89 IPR 414 (shape of a house); Re Application by Grove Fruit Juice Pty Ltd (2011) 91 IPR 380 (colour white for fruit juice packaging); Re Ferrero SpA (2012) 98 IPR 250 (shape of the Tic Tac container); Re Fin Control Systems Pty Ltd (2013) 102 IPR 488 (pattern of dots on a surfboard to indicate where fins were to be attached); Re Notaras (2013) 104 IPR 421 (shape of the Atomic brand coffee maker); Re Apple Inc (2015) 115 IPR 116 (the edges and rounded corners, oblong speaker slit and concave circular home button of an iPhone 4); Coca-Cola Co v Frucor Beverages Ltd (2016) 121 IPR 52 (colour green (Pantone 376c) for energy drinks); Re Van Ruth Pty Ltd (2016) 122 IPR 319 (industrial plug shape); Re Gram Engineering Pty Ltd (2017) 130 IPR 273 (shape of a slat); Reid [2017] ATMO 144 (stitching design).

(107) In Re Application by Chocolaterie Guylian NV (n 93) and Re Kenman Kandy Australia Pty Ltd (n 106) the hearing officers considered that the marks were functional because they consisted of attractive design features that added substantial value to the goods. It is noticeable that after the Full Federal Court's decision in Kenman Kandy (n 73), the Office has not considered marks to be 'aesthetically functional' on this basis. More recently, in Coca-Cola Co v Frucor Beverages Ltd (n 106), the hearing officer noted evidence that 'green' when used in relation to energy drinks could serve a technical function (to prevent light degradation) but also an ornamental function (providing information as to flavour or ingredients): at 63 [24]; see also at 79-81 [74]-[77] (Hearing Officer Kirov). The issue of functionality was not considered on appeal: Frucor (n 96). For further discussion, see below Part V(B).

(108) Three of the decisions were appealed to the Federal Court. In Frucor (n 96), on appeal from Coca-Cola Co v Frucor Beverages Ltd (n 106), Yates J (in obiter dicta) reached the same conclusion that the colour green had not acquired distinctiveness. The Hearing Officer's finding in Re Kenman Kandy Australia Pty Ltd (n 106) was overturned by the Full Federal Court in Kenman Kandy (n 73). The Court found the bug shape to be prima facie distinctive, a conclusion criticised in Part IV(A) below. The Hearing Officer's decision in Aldi Stores (A Limited Partnership) v Societe Des Produits Nestle SA (n 106), in which Aldi was successful in opposing Nestle's application for registration of its Kit Kat shape, was appealed to the Court. Before the hearing, however, Aldi withdrew its opposition and the Registrar indicated that it did not object to its decision being set aside, such that Nicholas J ordered the registration of Nestle's mark without considering the merits of the appeal: Societe Des Produits Nestle SA v Aldi Stores (A Limited Partnership) [2010] FCA 218.

(109) In Reid (n 106), the Registrar was required only to consider whether acceptance of the application should be revoked on the basis that a prima facie distinctiveness objection should have been raised in examination.

(110) Most of these decisions are from the period 2000-03, when it is possible that applicants were taking their chances under the then-new Act in seeking to persuade the Office that their marks were prima facie distinctive (and thus registrable without the need to provide any evidence of use).

(111) Unilever plc's Trade Mark Applications [2003] RPC 35, 662 [31]-[33] (Jacob J) (emphasis in original). This view has become entrenched in the UK, even after a reference to the Court of Justice: Societe des Produits Nestle SA v Cadbury UK Ltd [2018] 2 All ER 39. It has also been adopted by the Singapore Court of Appeal: Societe des Produits Nestle SA v Petra Foods Ltd [2017] 1 SLR 35.

(112) For recent examples, see Re Ferrero SpA (n 106) (giving weight to 'Tic Tac' branding); CocaCola Co v Frucor Beverages Ltd (n 106) (emphasising the impact of the primary 'V' brand); Re Van Ruth Pty Ltd (n 106) (considering that use of 'VRP' and 'Van Ruth' brands, and not the shape, distinguished the applicant's goods).

(113) The Office's approach is consistent with that taken by the Federal Court in considering acquired distinctiveness under the TMA (n 25) in relation to non-traditional (but nonfunctional) marks: see Woolworths Ltd v BP plc [No 2] (2006) 154 FCR 97 (BP unable to establish acquired distinctiveness in green used as the predominant colour of its service stations); Chocolaterie Guylian NV v Registrar of Trade Marks (2009) 180 FCR 60 (seahorse shape for chocolate not factually distinctive).

(114) All evidence of the state of the Register is available at IP Australia, Australian Trade Mark Search (Web Page) <https://search.ipaustralia.gov.au/trademarks/search/quick>. For an earlier empirical overview of registrations of non-traditional marks under the TMA (n 25), see Amanda Scardamaglia and Mitchell Adams, 'Registering Non-Traditional Signs as Trade Marks in Australia: A Retrospective' (2016) 26(3) Australian Intellectual Property Journal 149.

(115) AU Trade Mark No 707483, filed on 29 April 1996 (Registered on 4 December 1997).

(116) AU Trade Mark No 822780, filed on 9 February 2000 (Registered on 15 April 2010).

(117) AU Trade Mark No 895183, filed on 1 October 2001 (Registered on 17 January 2005).

(118) AU Trade Mark No 1059752, filed on 10 June 2005 (Registered on 19 September 2007).

(119) See Pi-Design AG v Yoshida Metal Industry Co Ltd (Court of Justice of the European Union, C-337/12 P to C-340/12 P, ECLI:EU:C:2014:129, 6 March 2014) and, on appeal from the remand decision, Yoshida Metal Industry Co Ltd v European Union Intellectual Property Office (Court of Justice of the European Union, C-421/15 P, ECLI:EU:C:2017:360, 11 May 2017) (design of the Yoshida knife handle necessary to obtain a technical result); Simba Toys GmbH & Co KG v European Union Intellectual Property Office (Court of Justice of the European Union, C-30/15 P, ECLI:EU:C:2016:849, 10 November 2016) (shape of the Rubik's cube necessary to obtain a technical result).

(120) Societe des Produits Nestle SA v Petra Foods Ltd (n 111) (Singapore Court of Appeal) (shape of the four finger Kit Kat bar 'necessary to obtain a technical result': at 83 [115] (Sundaresh Menon CJ for the Court)).

(121) Hauck (n 10) (considering whether the shape of the Stokke chair added substantial value to the goods).

(122) AU Trade Mark No 1537250, filed on 13 November 2012 (Registered on 22 October 2014).

(123) AU Trade Mark No 1577467, filed on 28 May 2013 (Registered on 2 March 2017); AU Trade Mark No 1578556, filed on 28 May 2013 (Registered on 2 March 2017); AU Trade Mark No 1578557, filed on 28 May 2013 (Registered on 2 March 2017); AU Trade Mark No 1583905, filed on 28 May 2013 (Registered on 2 March 2017). The functional nature of these marks was noted in Re Henkel AG & Co KGaA (2015) 112 IPR 33 ('Re Henkel').

(124) AU Trade Mark No 1660227, filed on 24 November 2014 (Registered on 9 February 2017).

(125) AU Trade Mark No 1709495, filed on 24 July 2015 (Registered on 14 December 2016).

(126) AU Trade Mark No 1737034, filed on 25 November 2015 (Registered on 23 November 2016) (the endorsement specifies that the descriptive word 'SMARTLEG' appears on the shape).

(127) AU Trade Mark No 1820186, filed on 12 January 2017 (Registered on 9 August 2017).

(128) AU Trade Mark No 780225, filed on 4 December 1998 (Registered on 19 November 2001).

(129) AU Trade Mark No 888415, filed on 7 September 2001 (Registered on 15 February 2006).

(130) AU Trade Mark No 971988, filed on 25 September 2003 (Registered on 24 October 2005).

(131) AU Trade Mark No 1466881, filed on 22 December 2011 (Registered on 16 September 2015).

(132) AU Trade Mark No 1595348 (n 5). Cf Re Orange Communications Inc, 41 USPQ 2d 1036 (TTAB, 1996) (orange held to be functional for public telephones and telephone booths, since the colour promoted the visibility of the goods in emergency situations).

(133) AU Trade Mark No 1159874, filed on 6 February 2007 (Registered on 9 March 2010). Cf Kasco Corporation v Southern Saw Service Inc, 27 USPQ 2d 1501 (TTAB, 1993) (green used on wrappers for saw blades was held to be functional, where green was one of six colours used in a system to identify different types of blade).

(134) (2008) 166 FCR 312 ('Mayne Industries').

(135) Ibid 338-9 [63]-[66].

(136) Ibid 340 [69]. Greenwood J also ordered that the registration of the shape mark be cancelled. This was not, however, because the mark was not distinctive at its filing date. Rather, it was on the basis that the shape was thought to be 'a sign that describes or is the name of' a formerly patented article and, more than two years after that expiry, 'the sign is the only commonly known way to describe or identify the article': at 340-8 [71]-[99], quoting TMA (n 25) ss 25(1), 87. For discussion of the Court's manipulation of the text of s 25 to achieve a desirable policy outcome, see Burrell and Handler, Australian Trade Mark Law (n 54) 79-81; Craig Smith, 'Trying to Fence in a Monopoly? Section 25 of the Trade Marks Act' (2008) 20(10) Australian Intellectual Property Law Bulletin 150.

(137) (2009) 80 IPR 244.

(138) Ibid 271-2 [144]-[147], 275 [162].

(139) Anheuser-Busch Inc v Budejovicky Budvar np (2002) 56 IPR 182, 228 [191] (Allsop J).

(140) The current editors of Shanahan's Australian Law of Trade Marks & Passing Off have suggested that it would be 'rare' for a functional trade mark to be infringed, but they do not rule out the possibility: Mark Davison and Ian Horak, Shanahan's Australian Law of Trade Marks & Passing Off (Lawbook, 6th ed, 2016) 658.

(141) Irving's Yeast-Vite Ld v Horsenail (1934) 51 RPC 110.

(142) Champagne Heidsieck et Cie Monopole SA v Buxton [1930] 1 Ch 330.

(143) Wingate Marketing Pty Ltd v Levi Strauss & Co (1994) 49 FCR 89.

(144) Musidor BV v Tansing (1994) 52 FCR 363.

(145) The relevant cases are Transport Tyre Sales Pty Ltd v Montana Tyres Rims & Tubes Pty Ltd (1999) 93 FCR 421; E & J Gallo Winery v Lion Nathan Australia Pty Ltd (2009) 175 FCR 386; Paul's Retail Pty Ltd v Sporte Leisure Pty Ltd (2012) 202 FCR 286; Paul's Retail Pty Ltd v Lonsdale Australia Ltd (2012) 294 ALR 72; Scandinavian Tobacco Group Eersel BV v Trojan Trading Co Pty Ltd (2016) 243 FCR 152.

(146) For critical analysis of this issue, see Robert Burrell and Michael Handler, 'A Requiem for Champagne Heidsieck: Trade Mark Use and Parallel Importation' (2016) 26(3) Australian Intellectual Property Journal 110.

(147) A point explored in more detail in Part IV(B) below.

(148) This is particularly given that the costs of a successful party are, except in rare cases, only partially recoverable.

(149) See generally Robert G Bone, 'Notice Failure and Defenses in Trademark Law' (2016) 96(3) Boston University Law Review 1245. It might be argued that the fact that, under Australian law, the owner of a functional mark can always bring proceedings for passing off and for breach of the prohibition on engaging in '[m]isleading or deceptive conduct ... in trade or commerce' under s 18 of the Australian Consumer Law, where a third party has adopted a similar mark, must lessen any chilling effects caused by the registration of that functional mark: Competition and Consumer Act 2010 (Cth) sch 2 ('Australian Consumer Law'). However, for the owner to be able, in a cease and desist letter, to point to a registration that clearly corresponds with the shape or colour the defendant is using or intends to use would likely have a much stronger impact than what would inevitably be a more abstract assertion that the owner has developed a sufficient reputation in its mark and that the defendant's use in all the circumstances would be misleading.

(150) Kenman Kandy (n 73) 535 [156] (Stone J).

(151) Ibid 536-7 [162] (Stone J); see also at 510-11 [49] (French J).

(152) Ibid 511 [50].

(153) See Wal-Mart Stores Inc v Samara Bros Inc, 529 US 205 (2000). For extended criticism of Kenman Kandy (n 73), see Burrell and Handler, Australian Trade Mark Law (n 54) 136-8.

(154) Davison (n 60) 109-10.

(155) A point recognised as long ago as 1934, where the Goschen Committee in the UK considered that 'the grant of monopolies for indefinite periods for the shape of say a bottle or jar might prove embarrassing to other traders and to the public generally': Board of Trade (UK), Report of the Departmental Committee on the Law and Practice Relating to Trade Marks (Cmd 4568, 1934) 7 [13].

(156) It is worth noting that the competition concerns in relation to non-functional signs do not disappear. Rather, as a matter of policy, it has been generally accepted that the interests of those who have acquired distinctiveness in such signs (and, arguably, consumers who have come to recognise that such signs operate as badges of origin) should be prioritised over the need to keep such signs free for general commercial use. But see Lisa P Ramsey, 'Descriptive Trademarks and the First Amendment' (2003) 70(4) Tennessee Law Review 1095 (arguing that descriptive signs should be unregistrable, even on a showing of factual distinctiveness, to avoid undue suppression of third party speech).

(157) Kenman Kandy (n 73) 511 [50] (French J).

(158) The test is not whether competitors would wish to use the sign 'as a trade mark', but simply whether they would wish to make any use of the sign 'in the ordinary course of their businesses': FH Faulding & Co Ltd v Imperial Chemical Industries of Australia and New Zealand Ltd (1965) 112 CLR 537, 555 (Kitto J); Clark Equipment (n 52) 516-17 (Kitto J).

(159) (2005) 64 IPR 607.

(160) Ibid 611-12 [19], 614 [27]-[28] (Hearing Officer Thompson).

(161) AU Trade Mark No 1456716, filed on 28 October 2011 (Registered on 5 June 2012).

(162) AU Trade Mark No 1760883, filed on 24 March 2016 (Registered on 3 November 2016).

(163) See, eg, AU Trade Mark No 1655901, filed on 31 October 2014 (Registered on 18 October 2016); AU Trade Mark No 1680031, filed on 10 March 2015 (Registered on 13 February 2017); AU Trade Mark No 1757877, filed on 25 January 2016 (Registered on 29 June 2016); AU Trade Mark No 1812778, filed on 2 May 2016 (Registered on 2 March 2017).

(164) AU Trade Mark No 802345, filed on 2 August 1999 (Removed on 16 October 2008).

(165) AU Trade Mark No 908216, filed on 15 February 2002 (Removed on 30 August 2012).

(166) AU Trade Mark No 957334, filed on 13 May 2003 (Removed on 9 January 2014).

(167) AU Trade Mark No 966145, filed on 14 August 2003 (Removed on 6 March 2014).

(168) AU Trade Mark No 1028331, filed on 5 November 2004 (Removed on 28 May 2015).

(169) AU Trade Mark No 1030836, filed on 19 November 2004 (Removed on 18 June 2015).

(170) AU Trade Mark No 1097622, filed on 7 February 2006 (Removed on 1 September 2016).

(171) AU Trade Mark No 1221817, filed on 21 November 2007 (Removed on 23 February 2017).

(172) AU Trade Mark No 1143015, filed on 7 August 2006 (Removed on 23 March 2017).

(173) TMA (n 25) s 41(3)(a).

(174) Ibid s 41(4)(a).

(175) For example, when compared with US law, they do not correspond with the 'descriptive' and 'generic' categories on the distinctiveness spectrum articulated in Abercrombie & Fitch Co v Hunting World Inc, 537 F 2d 4, 9-10 (Friendly J) (2nd Cir, 1976), since a mark that is 'not to any extent inherently adapted to distinguish' need not be as far along the spectrum as a generic mark.

(176) For further discussion and illustrative examples, see Burrell and Handler, Australian Trade Mark Law (n 54) 105-15.

(177) The ability to register a mark on the basis that it will acquire distinctiveness is broadly similar to the approach taken to registration on the Supplemental Register in the US: see J Thomas McCarthy, Westlaw, McCarthy on Trademarks and Unfair Competition, Fifth Edition (online at 4 July 2018) [section] 19:38.

(178) See, eg, Chocolaterie Guylian NV v Registrar of Trade Marks (n 113).

(179) The figure of 181 was determined by conducting a search at IP Australia, Australian Trade Mark Search (Web Page) <https://search.ipaustralia.gov.au/trademarks/search/quick> of the number of 'shape' marks entered on the Register between 28 August 2002 and 31 December 2017 with an endorsement containing the terms '41(6)', '41(5)', '41(4)' or '41(3)' (all of which indicate that the mark was registered only on the basis of factual distinctiveness).

(180) The figure of 149 was determined by conducting the search described above n 179, but with the search of the endorsement limited to the terms '41(5)' or '41(4)' (which indicate that the mark was initially considered by the Office to be to some extent, but not sufficiently, inherently adapted to distinguish).

(181) The total figure of 127 was determined by conducting the search described above n 179, but by reference to 'colour' marks. The figure of 94 was determined by limiting the search of the endorsement to the terms '41(5)' or '41(4)'.

(182) In addition, one of the anomalies of the Australian trade marks system is that it does not contain a ground for cancelling the registration of a mark that was classified as 'not sufficiently inherently adapted' to distinguish and was registered only on the basis that it would acquire distinctiveness, if such a mark does not in fact become distinctive after being registered for a particular period of time. For criticism, see Burrell and Handler, Australian Trade Mark Law (n 54) 317.

(183) For confirmation in recent Office decisions, see Re Notaras (n 106) 428 [24] (Hearing Officer Wilson); Re Henkel (n 123) 41 [20] (Hearing Officer Murray).

(184) AU Trade Mark No 815934 (n 5).

(185) AU Trade Mark No 902216, filed on 4 February 2002 (Registered on 30 October 2006).

(186) AU Trade Mark No 902483, filed on 28 December 2001 (Registered on 26 September 2005).

(187) AU Trade Mark No 1059752 (n 118). See also above n 119.

(188) AU Trade Mark No 1250780, filed on 7 July 2008 (Registered on 2 November 2010).

(189) AU Trade Mark No 1372905, filed on 16 July 2010 (Registered on 9 October 2013).

(190) AU Trade Mark No 1442181, filed on 11 August 2011 (Registered on 23 October 2013).

(191) AU Trade Mark No 1218000 (n 6).

(192) AU Trade Mark No 1466881 (n 131).

(193) (2008) 76 IPR 161 ('Global Brand').

(194) This has been recognised as a significant problem in a number of recent Australian inquiries: see Advisory Council on Intellectual Property, Review of Trade Mark Enforcement (Report, April 2004) 14-15; Productivity Commission, Intellectual Property Arrangements (Draft Report, April 2016) 335-6. See also below Part V(A).

(195) Global Brand (n 193) 175 [61], citing Coca-Cola Co v All-Fect Distributors Ltd (1999) 96 FCR 107, 118 [25] (Black CJ, Sundberg and Finkelstein JJ).

(196) Global Brand (n 193) 177 [71]-[73].

(197) Ibid 176 [66]-[67]. Sundberg J ultimately found for the defendant on the basis that its soles and shoe designs were not deceptively similar to the registered marks: at 180 [94].

(198) Ibid 177 [71] (Sundberg J). As a result, and quite counterintuitively, the presence of this primary brand ended up being treated as a factor that somehow counted towards the overall shoe configuration operating as a trade mark.

(199) (2014) 322 ALR 505.

(200) Ibid 544 [213] (Besanko J). As in Global Brand (n 193), the defendant succeeded on the basis that its bottle shape was deceptively similar to the registered mark. A broadly similar approach was taken in the parallel New Zealand infringement proceedings over the same bottle shapes: Coca-Cola Co v Frucor Soft Drinks Ltd (2013) 104 IPR 432, 461-3 [132]-[141] (Wylie J) (High Court of New Zealand).

(201) See also Adidas AG v Pacific Brands Footwear Pty Ltd [No 3] (2013) 308 ALR 74 (finding the defendant's use of stripe patterns on its shoes to have been trade mark use rather than decoration). But see Clipsal Australia Pty Ltd v Clipso Electrical Pty Ltd [No 3] (2017) 122 IPR 395, 425 [149]-[155] (Perram J) (finding the defendant's use of the non-functional shape of a light switch not to be 'trade mark use', focusing on the packaging of the defendant's switch and its use of a primary brand name).

(202) See above Part IV(A).

(203) This is so even after the High Court's recent consideration of the test of prima facie distinctiveness in Cantarella Bros Pty Ltd v Modena Trading Pty Ltd (2014) 254 CLR 337, which focused more on how the 'ordinary signification' of a sign is to be assessed: at 361-2 [70]-[71] (French CJ, Hayne, Crennan and Kiefel JJ). The case did not set up a test akin to that which exists in US or EU law that focuses explicitly on consumer perception in determining whether or not a mark is inherently distinctive or has distinctive character: see, eg, The H Marvin Ginn Corporation v The International Association of Fire Chiefs Inc, 782 F 2d 987, 990 (Rich J) (Fed Cir, 1986); SAT.1 SatellitenFernsehen GmbH v Office for Harmonisation in the Internal Market (Trade Marks and Designs) (C-329/02 P) [2004] ECR I-8317, I-8347 [24].

(204) See Davison (n 60) 111; Smith, 'Trade Mark Protection for Product Shape' (n 60) 47 (both suggesting a legislative requirement that all shape marks should be shown to be factually distinctive).

(205) Qualitex (n 18).

(206) 529 US 205 (2000).

(207) Although the Supreme Court has also found that non-product-configuration trade dress can be prima facie distinctive in Two Pesos Inc v Taco Cabana Inc, 505 US 763 (1992), this decision has been strongly criticised: see, eg, Kenneth L Port, 'Revisiting Park 'N Fly: In Pursuit of Constraints on Trademark Bullies' (2015) 16(1) Wake Forest Journal of Business and Intellectual Property Law 134, 153-5.

(208) See Glynn S Lunney Jr, 'Trademark Monopolies' (1999) 48(2) Emory Law Journal 367, 389-90. On the need to be careful in seeing free-riding as misappropriation, see generally Mark A Lemley, 'Property, Intellectual Property, and Free Riding' (2005) 83(4) Texas Law Review 1031; Dev Gangjee and Robert Burrell, 'Because You're Worth It: L'Oreal and the Prohibition on Free Riding' (2010) 73(2) Modern Law Review 282.

(209) For discussion in the context of EU law, see generally Dev S Gangjee, 'Paying the Price for Admission: Non-Traditional Marks across Registration and Enforcement' in Irene Calboli and Martin Senftleben (eds), The Protection of Non-Traditional Marks (Oxford University Press, forthcoming).

(210) Burrell and Handler, Australian Trade Mark Law (n 54) 158-9.

(211) For a discussion of illustrative cases, see ibid 376-9.

(212) See, eg, Baird, 'The Registrability of Functional Shape Marks' (n 60) 227; Davison (n 60) 106; McCutcheon, 'Monopolised Product Shapes' (n 60) 32; Levy (n 60) 32-3.

(213) Cf Smith, 'Trade Mark Protection for Product Shape' (n 60) 42-3, 47.

(214) 2015 Trade Marks Directive (n 19) art 4(1)(e).

(215) Ibid art 4(1)(e)(i).

(216) Ibid art 4(1)(e)(ii).

(217) Ibid art 4(1)(e)(iii).

(218) Trademark Act of 1946, 15 USC [section] 1052(e)(5) (2016).

(219) 532 US 23, 32 (Kennedy J for the Court) (2001) ('TrafFix'), quoting Qualitex (n 18) 165 (Breyer J for the Court), quoting Inwood Laboratories Inc v Ives Laboratories Inc, 456 US 844, 850 n 10 (O'Connor J for Burger CJ, O'Connor, Brennan, Blackmun, Powell and Stevens JJ) (1982).

(220) TrafFix (n 219) 33-4 (Kennedy J for the Court). Such an approach that downplays the importance of alternative designs has been criticised: see, eg, Dinwoodie (n 7) 721-8; Sandra L Rierson, 'Toward a More Coherent Doctrine of Trademark Genericism and Functionality: Focusing on Fair Competition' (2017) 27(4) Fordham Intellectual Property, Media & Entertainment Law Journal 691, 726-48. For recent consideration, see Moldex-Metric Inc v McKeon Products Inc (9th Cir, No 16-55548, 5 June 2018).

(221) Qualitex (n 18) 165 (Breyer J for the Court); TrafFix (n 219) 32-3 (Kennedy J for the Court).

(222) TrafFix (n 219) 33 (Kennedy J for the Court). It is worth noting, however, that courts in some circuits have interpreted the Supreme Court's statements on aesthetic functionality as obiter dicta, and have been ambivalent about, or not recognised, the doctrine: see generally McKenna (n 82) 849-50; Hughes (n 84) 1244-6.

(223) See generally Frederick Schauer, Playing by the Rules: A Philosophical Examination of RuleBased Decision-Making in Law and in Life (Clarendon Press, 1991); Louis Kaplow, 'Rules versus Standards: An Economic Analysis' (1992) 42(3) Duke Law Journal 557; Kathleen M Sullivan, 'Foreword: The Justices of Rules and Standards' (1992) 106(1) Harvard Law Review 22.

(224) See Burrell, Smith and Coleman (n 3) 162-3.

(225) Koninklijke Philips Electronics NV v Remington Consumer Products Ltd (n 89) I-5517 [79]. This was reaffirmed in Lego Juris A/S v Office for Harmonisation in the Internal Market (Trade Marks and Designs) (C-48/09 P) [2010] ECR I-8403, I-8465 [72] ('Lego Juris').

(226) Koninklijke Philips Electronics NV v Remington Consumer Products Ltd (n 89) I-5517 [81], I-5518 [83].

(227) See Bone, 'Trademark Functionality Reexamined' (n 82) 231-5.

(228) Societe des Produits Nestle SA v Cadbury UK Ltd [2014] FSR 28.

(229) Hauck (n 10) [39]-[43]; Societe des Produits Nestle SA v Cadbury UK Ltd (Court of Justice of the European Union, C-215/14, ECLI:EU:C:2015:604, 16 September 2015) [46]-[48].

(230) L Bently and B Sherman, Intellectual Property Law (Oxford University Press, 4th ed, 2014) 916.

(231) That is, a utilitarian functionality bar could be worded to make clear that the availability of alternate means of achieving the same technical result will not prevent the bar from applying, and the various types of functionality bars could be structured so that they potentially apply in a 'cumulative' manner.

(232) See above nn 93, 107, on the short-lived role played by 'aesthetic functionality' in Office decisions before the Full Federal Court's decision in Kenman Kandy (n 73).

(233) 2015 Trade Marks Directive (n 19) art 4(1)(e)(iii).

(234) Hauck (n 10) [32]. These need not be the only values: such features can, at the same time, add value by serving a utilitarian function, such as providing safety or comfort: at [29]-[30].

(235) Ibid [35].

(236) See Hughes (n 84) 1251-5.

(237) See above n 100 and accompanying text.

(238) Brunswick Corporation v British Seagull Ltd, 35 F 3d 1527 (Fed Cir, 1994).

(239) Dippin' Dots Inc v Frosty Bites Distribution LLC, 369 F 3d 1197 (11th Cir, 2004).

(240) Re Florists' Transworld Delivery Inc, 106 USPQ 2d 1784 (TTAB, 2013).

(241) See Hughes (n 84) 1248-51.

(242) Keene Corporation v Paraflex Industries Inc, 653 F 2d 822 (3rd Cir, 1981); The LD Kichler Co v Davoil Inc, 192 F 3d 1349 (Fed Cir, 1999).

(243) M-5 Steel Mfg Inc v O'Hagin's Inc, 61 USPQ 2d 1086, 1091 (Administrative Trademark Judge Simms) (TTAB, 2001).

(244) For a recent case where compatibility arguments fell on the other side of the line, see Deere & Co v FIMCO Inc, 239 F Supp 3d 964, 997, 1001-3 (Russell J) (WD Ky, 2017) (John Deere's green and yellow colour combination for tractors not 'aesthetically functional').

(245) See Kur (n 8) 8-9.

(246) As noted in Hauck (n 10) [19].

(247) 198 F 2d 339 (9th Cir, 1952).

(248) Ibid 343 (Orr J).

(249) Famolare Inc v Melville Corporation, 472 F Supp 738, 744-5 (Schwarzer J) (D Haw, 1979).

(250) TrafFix (n 219) 32-3 (Kennedy J for the Court), citing Qualitex (n 18).

(251) McKenna (n 82) 859. See generally Michael S Mireles Jr, 'Aesthetic Functionality' (2013) 21(2) Texas Intellectual Property Law Journal 155.

(252) See, eg, Wallace International Silversmiths Inc v Godinger Silver Art Co Inc, 916 F 2d 76 (2nd Cir, 1990); Abercrombie & Fitch Stores Inc v American Eagle Outfitters Inc, 280 F 3d 619 (6th Cir, 2002).

(253) Benelux Convention on Trade Marks, Belgium-Netherlands-Luxembourg, signed 19 March 1962, 704 UNTS 301 (entered into force 1 July 1969) annex ('Benelux Uniform Law on Trade Marks'). Article 13 of the Benelux Convention on Trade Marks ('Benelux Convention') provided that the Benelux Uniform Law on Trade Marks would come into force 18 months after the Benelux Convention entered into force. On the background to the Benelux Uniform Law on Trade Marks, see Tobias Cohen Jehoram, Constant van Nispen and Tony Huydecoper, European Trademark Law: Community Trademark Law and Harmonized National Trademark Law (Kluwer Law International, 2010) 21-2.

(254) Benelux Uniform Law on Trade Marks (n 253) art 1.

(255) Kur (n 8) 8.

(256) (T-508/08) [2011] ECR II-6975 ('Bang & Olufsen').

(257) (OHIM Fifth Board of Appeal, R 664/2011-5, 18 March 2015) ('Consorzio Origini').

(258) As summarised by Arnold J in The London Taxi Corporation Ltd v Frazer-Nash Research Ltd [2016] EWHC 52 (Ch), 'it is the shape itself which must add substantial value to the goods and that goodwill derived from sales and advertising is not relevant': at [209]. See also Christopher Buccafusco and Mark A Lemley, 'Functionality Screens' (2017) 103(7) Virginia Law Review 1293, 1347.

(259) Bang & Olufsen (n 256). The shape in question had, in fact, been found to be prima facie distinctive: at II-7001 [74]. As to why the decision remains problematic, see below n 263.

(260) See Kur (n 8) 15-16. See also The London Taxi Corporation Ltd v Frazer-Nash Research Ltd [2018] FSR 7, where Floyd LJ wondered 'whether, in addressing substantial value, one should take into account or ignore the fact that consumers will recognise the shape' as indicating source: at 178 [76].

(261) Hauck (n 10) [35].

(262) Ibid.

(263) In Consorzio Origini (n 257), the Board considered that '[a] marketing campaign in which [design excellence] is dominant may be revelatory of inherent value', without explaining how this value could somehow be extricated from trade mark value at the time trade mark protection was being sought: at [21]. See also Bang & Olufsen (n 256). As noted above n 259, the speaker shape had been found to be prima facie distinctive. However, what is absent from the decision is any consideration of the impact of the fact that consumers would have immediately recognised the shape to be operating as a trade mark (meaning that the shape had instant 'trade mark value') on the 'substantial value' said to have been provided by the shape.

(264) See McKenna (n 82) 846-8.

(265) My thanks go to Dev Gangjee and Mark McKenna for raising this idea.

(266) For a recent, albeit controversial, illustration from the General Court of the European Union, see Flamagas SA v European Union Intellectual Property Office (General Court of the European Union, T-580/15, ECLI:EU:T:2017:433, 27 June 2017) [80]-[81].

(267) See generally Burrell and Handler, Australian Trade Mark Law (n 54) 141-3. See also, under EU law, Lidl Stiftung & Co KG v Crocs Inc (OHIM Cancellation Division, 8577 C, 28 October 2014) (noting that the functionality bar 'cannot apply where the trade mark relates to a shape of goods in which a non-functional element, such as the Crocs Logo on the rivet of the strap in this case, plays an important role': at [32]). This finding was upheld on appeal in Lidl Stiftung & Co KG v Crocs Inc (OHIM Fifth Board of Appeal, R 3021/2014-5, 29 January 2016) [35]-[37] and is consistent with Court of Justice authority: Lego Juris (n 225) (the functionality bar does not apply where 'a non-functional element, such as a decorative or imaginative element, plays an important role': at I-8465 [72]).

(268) See Gangjee (n 209).
COPYRIGHT 2018 Melbourne University Law Review
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2018 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Handler, Michael
Publication:Melbourne University Law Review
Date:Aug 1, 2018
Words:28804
Previous Article:IS A CAUSE OF ACTION A CASTLE? STATUTORY CHOSES IN ACTION AS PROPERTY AND S 51 (XXXI) OF THE CONSTITUTION.
Next Article:TRICKED INTO MARRIAGE.
Topics:

Terms of use | Privacy policy | Copyright © 2021 Farlex, Inc. | Feedback | For webmasters |