University of New South Wales Law Journal
|
|
Home
| Databases
| WorldLII
| Search
| Feedback
University of New South Wales Law Journal |
|
TRADE MARK LAW’S IDENTITY CRISIS
(PART 2)
MICHAEL HANDLER[1]*
The concept of ‘substantial identity’ has not been the subject of sustained critical inquiry in Australian trade mark law, notwithstanding that it plays a crucial role in relation to trade mark ownership, non-use, amendments to representations, and the criminal offences. This second part of a two-part article provides a critique of the Federal Court’s recent move away from a strict approach to, and embrace of an expansive test of, substantial identity. It is argued that the Court’s new interpretation is inconsistent with earlier authority and pays insufficient regard to the historical and normative underpinnings of the provisions of the Trade Marks Act 1995 (Cth) in which substantial identity has vital work to do. It has the potential to destabilise core doctrines of Australian law and unjustifiably privilege the interests of particular parties in the trade marks system. The article concludes with options for reorienting Australia’s law of substantial identity.
In the first part of this two-part article[1] it was argued that the Full Court of the Federal Court in Accor Australia & New Zealand Hospitality Pty Ltd v Liv Pty Ltd (‘Accor’)[2] and Pham Global Pty Ltd v Insight Clinical Imaging Pty Ltd (‘Pham Global’)[3] disrupted the settled understanding of ‘substantial identity’ in Australian trade mark law. Substantial identity and its analogues in both the Trade Marks Act 1995 (Cth) (‘TMA’) and earlier legislation had been understood strictly for over a century. By reinterpreting Windeyer J’s test for substantial identity in The Shell Co of Australia Ltd v Esso Standard Oil (Australia) Ltd (‘Shell v Esso’)[4] so as to require greater weight to be given to the common presence of ‘essential features’ of the marks, and effectively to require the ‘inessential features’ of the marks to be disregarded, the Full Court of the Federal Court has significantly expanded the scope of the doctrine.
In the second part of this two-part article it is argued that this ‘new law of substantial identity’[5] represents a wrong turn in the law. The approach taken in Accor and Pham Global finds no support in, and is in fact inconsistent with, earlier authority. It has led to an unwarranted expansion of the scope of both the ownership ground of opposition in the TMA and other provisions in which substantial identity and its analogues have critical work to do, such as the non-use grounds of removal, the provisions allowing amendments to representations of marks, and the ‘false application’ criminal offences. It is suggested, by way of conclusion, that Accor and Pham Global are plainly wrong and should be overturned, but that irrespective of how the Full Court of the Federal Court might deal with this matter in the future, there are strong grounds for the legislature to intervene to address more fundamental issues with the operation of the concept of substantial identity throughout the TMA.
The most glaring problem with the Full Court of the Federal Court’s reformulated test for substantial identity in Accor and Pham Global is that it misconstrues Windeyer J’s test in Shell v Esso.[6] The Court has not only marginalised critical elements of the Shell v Esso test, but has failed to recognise that that test must itself be read in light of earlier, binding authority, and with an understanding of the role that substantial identity and its analogues have played and continue to play throughout the TMA.
As explained in Part II(A) of the first part of this article, the Shell v Esso test requires the decision-maker to assess the importance of the identified similarities and differences between the marks, having regard to both the essential features of the earlier mark and the total impression of resemblance that emerges from the comparison. This comparison is not an abstract one. The marks are to be assessed side by side as to the extent of visual similarity between them (with subsequent cases recognising that the assessment can include aural similarity).[7] Because it is a side-by-side comparison, the decision-maker is required to assess the extent to which the marks, as a whole, look and sound ‘practically the same’ (the phrase used by Isaacs J in Schweppes Ltd v E Rowlands Pty Ltd (‘Schweppes’))[8] or are ‘so alike that they can be described as one and the same mark’ (the phrase used by the unanimous High Court in Innes v Lincoln Motor Co (‘Innes’)).[9]
This is not to say that such an assessment is to be undertaken simply by tallying up the number and size of the visual and aural differences between the marks. Not all similarities and differences between the marks are to be given the same weight, and this is where the ‘essential features’ of the earlier mark become relevant. This factor in the Shell v Esso test will likely be decisive if the essential features of the earlier mark are not replicated in the later mark. In this situation, it is almost inconceivable that the marks will be substantially identical, even if the resulting visual and aural differences between the marks are relatively slight. To return to an example from Part II(B) of the first part of this article, the single letter difference between FORTEX and FOREX was rightly held to be enough to ensure that the marks were not substantially identical, given that the single, essential feature of the earlier mark is the six-letter word ‘fortex’.[10]
If the essential features of the earlier mark are, however, replicated in the later mark, the situation is more complex. The common presence of essential features is clearly a strong factor in favour of substantial identity and means that greater differences can exist between the marks’ other features without preventing the marks from being substantially identical. Thus, FORTEX AIRLINES would undoubtedly be substantially identical with FORTEX AIRWAYS,[11] even if the difference between ‘lines’ and ‘ways’ is larger than that between ‘fortex’ and ‘forex’. Similarly, the addition of a basic design feature, such as an oval border, to the word ‘fortex’ would not prevent it from being substantially identical to the word mark FORTEX.
The Full Court’s approach in Pham Global, and in Accor before it, overemphasises the common presence of essential features, effectively making that factor determinative of substantial identity, rather than being a factor to which regard must be had in undertaking the inquiry. In doing so, and in discounting or disregarding non-essential features, the Full Court has set up a test that marginalises the key part of the Shell v Esso test that requires the decision-maker to consider the total impression of resemblance that emerges from the side-by-side, visual and aural comparison.[12] The Shell v Esso test contemplates that even where the essential features of the earlier mark are contained in the later mark, there must come a point at which the differences between the other, non-essential elements are so visually and aurally striking that it cannot be said that there is a total impression of resemblance between the marks, taken as a whole. For example, the differences between the look and sound of FORTEX and FORTEX AIRWAYS are simply too prominent for the marks to be substantially identical. Equally, the addition of anything more than ‘the simplest of conventional embellishments’[13] to the word ‘fortex’, such as a sufficiently prominent device representing a bird, a wing, or a flower, would prevent that mark from being substantially identical with FORTEX.
Australian courts had appreciated what the Shell v Esso test requires in their pre-2017 decisions involving comparisons between word marks featuring descriptive elements, and comparisons between composite word and device marks. These decisions include those involving the ‘Opal Collection’ marks,[14] the ‘Bing’ marks,[15] the ‘Anchorage Capital’ marks[16] and the ‘Optical 88’ marks,[17] set out in Part II(A) and II(B) of the first part of this article. Most notably, in Optical 88 Ltd v Optical 88 Pty Ltd (No 2), in considering whether the device mark (below left) contained alterations that substantially affected the identity of the registered mark (below right), Yates J had regard to the essential features of the registered mark (the square shaped logo, ‘Optical 88’ and ‘眼鏡88’) but considered that the size and arrangement of these elements in the device mark meant that an overall impression of similarity did not emerge from the comparison with the registered mark.[18] This approach was supported by the Full Court of the Federal Court on appeal.[19]
 |
 |
These decisions all involve orthodox applications of the Shell v Esso test,[20] in contrast with the approach taken in Accor and Pham Global, and in the Trade Marks Office (‘Office’) decisions on substantial identity noted in Part III(B) of the first part of this article.
In response, it might be argued that the fact that Windeyer J incorporated the concept of ‘essential features’ (a term synonymous with ‘distinguishing features’)[21] from the case law on deceptive similarity into the test for substantial identity suggests that the two tests were intended to be much closer in scope than was ever appreciated by the courts, at least before Accor and Pham Global. In particular, the reference to ‘essential features’ in the substantial identity test might be taken to suggest that, as with deceptive similarity, the test is to be undertaken from the perspective of a consumer comparing the marks and seeking to determine if they represent different trade sources. Such a consumer might well be expected to ‘discount’ descriptive words,[22] or pay ‘minimal attention’[23] to devices next to distinctive words, or disregard aspects of stylisation or the precise placement of elements within a composite mark, in making source identification decisions, even in a side-by-side comparison. On this basis, could it be argued that the Full Court in Pham Global was right to reinterpret the Shell v Esso test as only ever requiring commonality between the essential, or distinguishing, features of the marks and that there is, in fact, nothing wrong with ‘substantial identity’ operating as a modified form of the ‘deceptive similarity’ inquiry?
The simple answer is no. The above interpretation is unsustainable, because it misunderstands the role that ‘essential features’ could possibly play in a test for substantial identity, given High Court authority prior to Shell v Esso that was not cited in Accor or Pham Global, and the role played by substantial identity throughout the Australian trade marks system.
Shell v Esso must be read in light of the High Court’s earlier decision in Schweppes. In that case, both Griffith CJ and Isaacs J, in considering the meaning of ‘substantially identical’ and ‘so nearly resembling ... as to be likely to deceive’ (which we would now term ‘deceptively similar’) in the infringement provision of the Trade Marks Act 1905 (Cth) (‘1905 Act’), recognised the two concepts to be conceptually distinct. Isaacs J held that to ask whether the defendant’s mark is ‘substantially identical’ with the registered mark effectively asks whether the former can be taken to be the registered mark – that is, it must be the ‘actual and identifiable property of the registered proprietor’.[24] His Honour held that this was not to be judged by reference to the sort of consumer responses to the marks as described two paragraphs above.[25] On the contrary, it is only if the defendant cannot be said to have used the owner’s ‘property’ that the analysis will turn to whether the marks are deceptively similar, which will depend on likely consumer reactions to the marks potentially co-existing in the marketplace.[26]
Shell v Esso and its reference to ‘essential features’ must therefore be read consistently with this prior understanding of substantial identity as not turning on consumer confusion, or how consumers would react on seeing the two marks in the marketplace, but rather turning on the different, narrower issue of whether the earlier mark owner’s property had been taken. To put it another way, even though the infringement provisions, and the provisions dealing with conflicts between applications for registration and earlier registered marks, might be said to be animated by a general concern to avoid consumer confusion if two marks were allowed to co-exist in the marketplace,[27] this does not provide support for the idea that the test for substantial identity is little more than a modified test for deceptive similarity. Such a view is inconsistent with Schweppes.
A related, and even more compelling, reason why substantial identity cannot be determined by the common presence of distinguishing features or by consumers’ willingness to disregard non-source identifying elements within marks is because of the role that substantial identity plays outside infringement cases. In disputes over ownership, consumer responses to the marks are entirely irrelevant. The earlier user X does not need to have any reputation in its mark to be the owner of applicant Y’s substantially identical mark. The test does not in any way turn on whether consumers would be confused if X and Y’s marks were to co-exist.[28] Instead, the issue is simply whether X’s mark is a substitute for Y’s mark. It is therefore erroneous to assess whether X and Y’s marks are substantially identical from the perspective of consumers ‘interpreting’ the marks and looking to make source identification decisions. It simply does not matter, in this context, that consumers would understand ‘choc’ in CHOC CHILL to represent flavour,[29] or ‘Cairns’ in CAIRNS HARBOUR LIGHTS to be a geographical reference,[30] or ‘Clinical Imaging’ and ‘radiology’ to be descriptive terms.[31] It is not to the point that consumers would pay little attention to devices adjacent to distinctive word components or aspects of stylisation in attempting to isolate the core, ‘distinctive’ elements of the mark. Such responses would clearly be decisive in a deceptive similarity inquiry. However, as the unanimous High Court recognised in Innes, a claim to proprietorship on the basis of use of a ‘nearly identical’ mark does ‘not extend to the case where confusion may arise through the use of trade marks which are not really the same’.[32]
The irrelevance of confusion to the ownership enquiry was reiterated by Gummow J in Carnival Cruise Lines Inc v Sitmar Cruises Ltd (‘Carnival’)[33] and by Allsop J in Colorado Group Ltd v Strandbags Group Pty Ltd (‘Colorado’).[34] In contrast, the current expansive interpretation of substantial identity, which places such weight on the ‘essential features’ aspect of the Shell v Esso test that it effectively looks to whether consumers would confuse the two marks based on their distinguishing features, rides roughshod over the long-standing and critical distinction that has been established between substantial identity and deceptive similarity.[35] It sets up a test that cannot be reconciled with the High Court’s unanimous decision on proprietorship in Innes.
The above problem also exists outside of ownership disputes. Whether a mark has been used with additions or alterations that do not substantially affect the mark’s identity,[36] or whether an amendment to the representation of a mark would not substantially affect its identity,[37] also turn on whether one mark can be said to be a substitute for the other. This is not something that can be judged from the perspective of consumers potentially encountering the two marks in a commercial setting – again, the relevant issue is not whether two marks can co-exist in the marketplace. If the Shell v Esso test is to be used as guidance in considering whether changes or amendments to a mark substantially affect the mark’s identity,[38] it would be problematic to apply the Pham Global approach, which downplays the question of whether a total impression of resemblance, assessed visually and aurally, emerges from the comparison.[39]
In summary, when the Shell v Esso test is read as being informed by Schweppes and Innes, and with a sensitivity to the role that substantial identity plays throughout the trade marks system, it is clear that it does not set up a modified deceptive similarity inquiry where the common presence of essential features is of overarching importance. Accor and Pham Global are thus inconsistent with earlier authorities. In this regard, it is encouraging to see that the Full Court of the Federal Court in Urban Alley Brewery Pty Ltd v La Sirène Pty Ltd has recently reaffirmed the different tests for and underlying rationales of the requirements of substantial identity and deceptive similarity, albeit without engaging with its earlier decisions in Accor and Pham Global.[40]
Can the approach adopted in Accor and Pham Global nevertheless be supported by High Court authority more recent than Shell v Esso? As was seen in Part III(B) of the first part of this article, a view has emerged that the High Court in E & J Gallo Winery v Lion Nathan Australia Pty Ltd (‘Gallo’)[41] interpreted an analogous concept to ‘substantial identity’ expansively. Specifically, the Office and Federal Court have interpreted Gallo as involving a finding that the registered owner Gallo’s use on its wine labels of a composite mark consisting of the word ‘Barefoot’ and a device that described that word (an image of a bare foot) constituted use of the word mark BAREFOOT ‘with additions or alterations not substantially affecting its identity’ for the purposes of section 100(3)(a) of the TMA, on the basis that consumers would give no weight to the device element.[42] Such a finding would seem to be consistent with the approach subsequently taken in Accor and Pham Global. However, a careful reading of Gallo shows that this interpretation is unsustainable.
The key passage from the judgment of French CJ, Gummow, Crennan and Bell JJ in Gallo, as well as Gallo’s wine label, are set out again for convenience:
The addition of the device [of the bare foot] to the registered trade mark is not a feature which separately distinguishes the goods or substantially affects the identity of the registered trade mark because consumers are likely to identify the products sold under the registered trade mark with the device by reference to the word BAREFOOT. The device is an illustration of the word. The monopoly given by a registration of the word BAREFOOT alone is wide enough to include the word together with a device which does not substantially affect the identity of the trade mark in the word alone. So much is recognised by the terms of s 7(1), which speak of additions or alterations which ‘do not substantially affect the identity of the trade mark’ ... The device is an addition to the registered trade mark that does not substantially affect its identity. Accordingly, the use of the registered trade mark with the device constitutes use of the registered trade mark in accordance with s 7(1).[43]
This passage must be understood in context. Two paragraphs earlier, their Honours accepted that the trial judge had ‘correctly rejected’ Lion Nathan’s submission that the ‘use of a trade mark consisting of the word BAREFOOT in combination with the device [of the bare foot] was not use of the registered trade mark, which was the word BAREFOOT simpliciter’.[44] As explained in Part III(B) of the first part of this article, Lion Nathan’s submission had two limbs. The first was that Gallo had, in fact, used a composite mark (the ‘BAREFOOT plus device’ mark); the second was that the use of this composite mark substantially affected the identity of the registered word mark. What the trial judge rejected – correctly, as held by the High Court – was the first limb only;[45] there was no need to consider the second limb. French CJ, Gummow, Crennan and Bell JJ then noted, in the next paragraph, that while Allsop J in Colorado had considered the use of ‘Colorado’ with a mountain peak device to be use of a composite mark,[46] this was ‘to be contrasted with the position in [the present] case’.[47] In other words, their Honours clearly accepted that Gallo had not used a composite mark. What then follows, in the passage from Gallo quoted above, can therefore be read only as a more detailed set of reasons in support of the trial judge’s finding that Gallo’s use of its word mark on the wine label met the requirements of section 100(3)(a) of the TMA.
When seen in this light, their Honours’ consideration of whether the ‘addition’ of the device of the bare foot substantially affected the identity of the word mark takes on an entirely new meaning. Their Honours were not considering whether the use of a composite (word plus device) mark substantially affected the identity of the word mark. Rather, they were considering whether the use of the word mark ‘with’ (that is, alongside, or in close proximity to) other elements somehow affected the identity of the word mark. The clear answer to this was no: the word mark had been used only in proximity to a decorative feature that merely illustrated the word and was thought to serve no trade mark purpose in that context. As such, that feature did not affect the identity of the word mark. In taking this approach, their Honours were drawing attention to the unusual scope of both section 7(1) of the TMA and its equivalent in the non-use context, section 100(3)(a). That is, although these provisions might in some cases look to whether two marks are, in effect, ‘substantially identical’, in other cases (such as Gallo) the provisions merely clarify that a word mark does not cease to be used as a word mark when the word is used in proximity to decorative or ‘non-trade mark’ matter (unless, of course, that matter changes the appearance and meaning of the word mark – for example, if the image of the bare foot had been placed in such a way that it obscured key letters in the word ‘Barefoot’).[48] Understood in this way, Gallo provides no support for the expansive approach adopted in Accor and Pham Global.
A notable feature of Accor, Pham Global and their antecedents is that they do not draw support for the expansive view of substantial identity from authorities older than Shell v Esso. Nonetheless, might it be possible to find such support in late nineteenth and early twentieth century cases and commentary, which might even cast doubt on the approach taken by the High Court in Schweppes and Innes? An analysis of how ‘less than identical’ marks were addressed in the first four decades of United Kingdom (‘UK’) trade mark law throws up a number of surprising statements that, at first glance, seem to suggest that ‘substantial identity’ was intended to be almost as broad as, and possibly even coterminous with, ‘deceptive similarity’. However, a careful assessment of these statements, and an examination of more relevant UK case law from that period, ends up providing further support for the argument that there is, in fact, no doctrinal basis for the expansive approach taken in Accor and Pham Global.
As noted in Part II(B) of the first part of this article, under Rules made under the UK’s Trade Marks Registration Act 1875, where multiple applicants for registration claimed to be registered as the proprietor of the same mark, or ‘nearly identical’ marks, the Registrar was given a discretion to manage such situations, or require the parties to have their rights determined by the court.[49] Although not the subject of litigation, these provisions, and those in subsequent Acts of 1883 and 1905, contemplated what might be called ‘traditional’ proprietorship disputes. That is, they recognised that disputes might arise between two applicants for registration of non-identical marks where one applicant claimed, by virtue of the earlier use of its mark, that it was the proprietor of the other applicant’s mark.[50]
However, the above provisions were drafted so as to go beyond ‘traditional’ proprietorship disputes. Rather, they allowed courts to consider whether multiple proprietors of the same mark or nearly identical marks should, in some cases, all be entered on the Register. For example, where two or three parties started trading in limited geographical areas, unaware of each other but using nearly identical marks, and such parties then applied for registration at much the same time, it was open to the court to accommodate the rights of these concurrent users on the Register by recognising each of them to be a proprietor. This is something courts were prepared to do under other, closely related legislative provisions. These provided that if a mark whose registration was being sought was either ‘identical’ to an already registered mark, or ‘so nearly resemble[ed]’ such a mark ‘as to be calculated to deceive’, it could not be registered, other than with the leave of the court (under the 1875 Act)[51] or if the court had determined that multiple parties were entitled to be registered as proprietors of the ‘same’ mark (under the 1883 Act).[52] A court approved practice called the ‘three-mark rule’ soon developed to flesh out the circumstances in which this prima facie bar to registration would be lifted.[53] In essence, if the already registered mark was in use before 13 August 1875 (known as an ‘old’ mark), one or two subsequent applications for registration of ‘duplicate’ old marks by other parties would be accepted[54] (the term ‘duplicate’ marks being adopted by treatise writer Duncan Kerly).[55] More specifically, the three-mark rule ‘extended to marks so similar, or so nearly resembling each other, as to be substantially the same marks, or calculated to deceive, and was not to be confined to identical marks’.[56] For example, in Re The Walkden Aërated Waters Co’s Application, each of the three marks in question contained an image of a cockerel, but in each case with a different company name surrounding the image.[57] What is most striking, however, is the shorthand terminology used to describe the ‘duplicate’ marks that were accepted for registration under the three-mark rule. Lewis Boyd Sebastian (writing in 1890) and Duncan Kerly (writing in 1894) both said that the rule extended to cover applications for ‘substantially identical’ marks. In support of this proposition, they cited cases in which the marks in question were not substantially the same, but rather so nearly resembled each other as to be calculated to deceive.[58]
Soon afterwards, the statutory language of ‘nearly identical’ came to be interpreted in an equally broad manner. By the 1890s, courts had shown themselves unwilling to lift the above-mentioned bar to registration for marks conflicting with already registered marks, in cases of honest concurrent use, where both marks were ‘new’ (that is, first used after 13 August 1875).[59] Recognising the ‘undoubted hardship’[60] caused by such decisions, the UK’s Trade Marks Act 1905 introduced a provision allowing the court to order the registration of the ‘same’ mark, or ‘nearly identical’ marks, in cases of honest concurrent use.[61] In Re Maeder’s Trade Mark Application (‘Maeder’) in 1916, Sargant J held that since the honest concurrent use proviso was intended to operate as a qualification to the general provision barring the registration of a mark that so nearly resembled an already registered mark as to be calculated to deceive, ‘the expression “nearly identical trade marks” corresponds with the expression ... “so nearly resembling such a trade mark as to be calculated to deceive”’.[62]
What traction might these statements have in interpreting Australian law? As was seen in Part II(A) of the first part of this article, in Australia’s 1905 Act the term ‘substantially identical’ was used in the core infringement provision in section 53, alongside the alternative term ‘so nearly resembling ... as to be likely to deceive’. Australia also adopted the UK’s general prohibition on the registration of marks ‘identical’ with an already-registered mark or ‘so nearly resembling such a trade mark as to be likely to deceive’, as well as the proviso, in section 28 of the 1905 Act, permitting the registration in cases of honest concurrent use of the ‘same’ mark or ‘nearly identical’ marks. In addition, Australia adopted the UK’s long-standing provision addressing disputes over proprietorship: section 27 of the 1905 Act entitled the Registrar, where multiple parties had applied to be registered as the proprietor of the same mark, or ‘nearly identical’ marks, to refuse to register the marks and instead refer the matter to the court. As has been demonstrated, the High Court in Schweppes and Innes interpreted ‘substantially identical’ in section 53 and ‘nearly identical’ in section 27 much more narrowly than how those terms had been used by Kerly and Sebastian, and by Sargant J in Maeder. Could it be said that the High Court should have been more alive to the historical origins of those terms, and interpreted them as having much the same scope as ‘deceptively similar’?[63] And, if so, could it be said that the Full Court of the Federal Court in Accor and Pham Global interpreted ‘substantially identical’ (albeit unwittingly) in manner that is consistent with its long forgotten, original meaning?
The answer to these questions is no. Both Kerly and Sebastian in discussing the ‘three-mark rule’, and Sargant J in Maeder, were considering situations where the starting point was that an application for registration was being blocked by the existence on the Register of a ‘deceptively similar’ mark. The question then turned to whether the application should be accepted, essentially because of the applicant’s honest concurrent use. The issue here was not one of competing claims to proprietorship of the one mark. Rather, the issue was whether multiple proprietors of the same, or deceptively similar marks, should be entitled to have their marks registered, notwithstanding the possibility of marketplace confusion. In this particular context, little turned on the breadth of the meaning given to terms such as ‘substantially identical’ or ‘nearly identical’, when the real issue was whether the marks were deceptively similar. Kerly, Sebastian and Sargant J in Maeder were not dealing with, or purporting to deal with, situations where it was important to know when the use of one mark could be taken to be the use of another, or be a substitute for another, on the basis of the marks being ‘nearly identical’. No court or treatise writer in this period ever suggested, for example, that a competing claim to proprietorship could be made out where the marks were anything less than ‘nearly identical’, strictly understood.
Critically, the High Court recognised this distinction in Innes. In interpreting the breadth of the term ‘nearly identical’ in section 27 of the 1905 Act, the Court indicated that it ‘ha[d] not overlooked the observations of Sargant J in [Maeder]’.[64] However, it recognised that a provision intended to resolve disputes over proprietorship could ‘not extend to the case where confusion may arise’ through the use of two trade marks,[65] as distinct from the honest concurrent use provision in issue in Maeder. For this reason, the Court held that section 27 only applied where the marks were ‘really the same’.[66] The effect of Innes was to create something of a tension within the 1905 Act, in that the term ‘nearly identical’ was to be interpreted narrowly in section 27 in cases on proprietorship (as not covering marks that were merely deceptively similar), but potentially more broadly in section 28 in cases of honest concurrent use (where it might cover deceptively similar marks, following Maeder). This bifurcation, however, reflected an awareness of the different roles played by proprietorship and deceptive similarity/honest concurrent use within the Australian trade marks system.
Even more significantly, to the extent that UK courts did address the issue of whether one mark could be taken to be a ‘substitute’ for another in this period, they applied a very strict test. The issue arose under a provision of the Trade Marks Registration Act 1875 that allowed the registration of ‘any special and distinctive word or words’ only if these had been in use before 13 August 1875.[67] When a number of these registrations were challenged on the basis that the marks had not been in use before that date, the registered owners sought to point to the use of marks containing, or similar to, the mark as registered. Most of these arguments were dismissed on the basis that the owners had not used the marks, as registered, alone. However, in two cases the English High Court considered whether the mark in use could be taken to be the mark as registered. In decisions that are entirely on fours with the narrow approach to ‘substantial identity’ that came to emerge in Australian law, it was held that the use of ‘Excelsior Spring Mattress’ preserved the registration of ‘The Excelsior Spring Mattress’,[68] but that the use of ‘Emollio Tablet’ was not sufficient to preserve the registration of ‘Emollio’ for soaps.[69] Both of these decisions were cited by Allsop J in Colorado as being consistent with the orthodox Australian approach to substantial identity, as reflected in Shell v Esso and Carnival.[70] In other words, the only nineteenth century UK case law directly on point ends up supporting the narrow, orthodox, pre-Accor approach to substantial identity.
In addition to being inconsistent with prior authority, there is another major problem with the approach to substantial identity developed by the Office from the mid-1990s and embraced in Accor and Pham Global. In these decisions and cases, tribunals have not paid sufficient attention to the normative bases of the grounds of the TMA under consideration – specifically, the ownership ground of opposition in section 58, and the non-use provisions in sections 92(4) and 100. An understanding of the rationales for these grounds and the roles they have historically played, and continue to play, within the registered trade marks system makes clear that they should be interpreted narrowly, involving strict readings of ‘substantial identity’ and analogous terminology. In addition, those tribunals, when interpreting the individual grounds of the TMA in the matters before them, have not considered the implications of taking a broad approach to substantial identity on other, related provisions within the trade marks system, such as those dealing with counterfeiting and forgery. These problems, which will be explored in detail in this Part, all coalesce around a common theme: that an expansive interpretation of substantial identity is normatively undesirable, because it results in certain parties within the trade marks system being unjustifiably privileged, to the detriment of the system as a whole.
When Australian courts have addressed the rationale for the ownership ground of opposition in section 58 of the TMA, they have usually explained that it reflects the ‘common law’ position where ‘rights to a trade mark are established by use’.[71] This, however, paints an incomplete picture. It is certainly the case that in the UK’s Trade Marks Registration Act 1875, the ‘proprietor’ of a mark was entitled to be registered, with this person being the first user of the mark or the purchaser of its business,[72] and that this model was adopted and has been retained in Australian registered trade mark law. As the Full Court of the Federal Court stated in Accor, the TMA does not set up ‘“a Torrens-title” system. It provides for registration of ownership not ownership by registration’.[73] What is, however, missing from the Australian courts’ engagement with the role of the ownership ground of opposition is an appreciation of how the nineteenth century ‘common law’ position changed fundamentally in the early twentieth century, such that rights to restrain another party’s conduct were no longer established by first use alone. Equally, courts have paid little attention to the fact that, over the course of the twentieth century, other provisions were introduced to ensure that those with actual ‘common law’ rights could prevent the registration of conflicting marks, and that prior users of unregistered marks could continue such use in the face of later registrations. These factors show that the current ownership provision has a highly unusual sphere of operation in the TMA, and ought to be interpreted in an exceptionally narrow manner.
To explain, by the 1860s a number of Chancery judges, most notably Lord Westbury LC, had started to justify equity’s intervention to restrain the use of a mark associated with the plaintiff, in the absence of any fraudulent conduct by the defendant, on the basis that what was being protected was an ‘invasion’ of the plaintiff’s ‘property’ in its mark.[74] This raised a question as to whether the plaintiff needed to show any particular level of use to establish ‘proprietorship’ of its mark. As Rob Batty has shown, Chancery judges soon held that any such level of use was very slight.[75] In Hall v Barrows, in 1863, Romilly MR stated that ‘although the mark may not have been adopted a week, and may not have acquired any reputation in the market, his neighbours cannot use that mark’.[76] An even broader approach was taken a year later in M’Andrew v Bassett.[77] Lord Westbury LC rejected an argument that a six week period of use, with no evidence of public reputation, was insufficient to establish a property interest in a mark, holding instead that a plaintiff in an infringement action need only show that it had applied its mark to a vendible article in the market, and that the defendant had imitated this mark.[78] Christopher Wadlow has described this decision as an ‘innovation’, noting that it was contrary to the approach taken by the Court of Chancery in the 1840s.[79] Nevertheless, based on decisions such as Hall v Barrows and M’Andrew v Bassett, it is unsurprising that the UK legislature set up a registration system in the mid-1870s that recognised the first user of the mark as its proprietor, without requiring that party to establish any reputation in its mark.
Crucially, however, outside the registered trade marks system the notion that there could be property in a trade mark, or that rights in such a mark could be established on the basis of first use alone, fell out of favour by the late nineteenth century. In 1896, in Reddaway v Banham, Lord Herschell doubted whether it was accurate to speak of ‘property’ in a trade mark at common law.[80] And in 1909, in Burberrys v JC Cording & Co Ltd, Parker J held that:
[i]f an injunction be granted restraining the use of a word or name, it is no doubt granted to protect property, but the property ... is not property in the word or name, but property in the trade or good-will which will be injured by its use ... [that] involves a misrepresentation.[81]
Lord Parker’s restatement of this view six years later in the groundbreaking passing off case of AG Spalding & Bros v AW Gamage Ltd received the unanimous support of the other members of the House of Lords.[82] It remains orthodox to this day.
Thus, by 1909 ‘proprietorship’ under registered trade mark law in the UK and Australia had fundamentally diverged from ‘proprietorship’ at common law. As a result, it remained possible for X to oppose, or seek the rectification of, the registration of Y’s mark on the basis of X’s first use of the mark, even if X could not prevent Y’s use of the mark at common law. This would have been the case if, by the time of Y’s application for registration, X had not made sufficient use of its mark such that it had not built up sufficient goodwill to enable it to bring a passing off action against Y, or because X had lost any residual goodwill after a period in which it had not used its mark.[83] This remains the case under section 58 of the TMA. It is perhaps surprising that over the course of the twentieth century courts in the UK and Australia did not seek to realign the ‘proprietorship’ provisions in trade mark statutes with the different position that had emerged at common law.[84] What Australian courts did at least recognise was that the scope of such proprietorship provisions should remain narrow, and not give X – a party that might have no common law rights in relation to its mark – the ability to prevent the registration of Y’s merely deceptively similar mark.[85]
The key point for present purposes is that we should not uncritically accept the idea that section 58 of the TMA has a valuable contemporary role to play as somehow giving effect to what the Full Court in Pham Global called ‘established common law principles’.[86] These principles have not been part of the common law for well over 100 years. This raises real questions about how we should go about interpreting section 58, consistent with the overall scheme of the TMA.
A further development in Australia is that, from the 1905 Act onwards, other provisions were introduced to ensure that a party with rights derived from its use of an ‘unregistered’ mark could prevent another trader from registering that mark or a sufficiently similar variant. Under the TMA, it is possible to point to a suite of at least five opposition grounds (including section 58) that attempt to regulate the relationship between applications for registration and use-based rights in potentially unregistered marks.[87] The three most important grounds are sections 42(b), 60 and 62A. Section 42(b) allows an opponent to argue that the applicant’s use of its mark would be ‘contrary to law’, which includes use that would amount to passing off and would contravene the prohibition on engaging in misleading or deceptive conduct in trade or commerce contained in section 18 of the Australian Consumer Law.[88] This ground ensures that the Register is an accurate reflection of the marketplace: if earlier user X has the ability to restrain Y’s use of its mark, then X should also be able to block Y’s attempt to register its mark. Section 60 allows an opponent to identify a mark that had acquired a ‘reputation’ in Australia before the applicant’s priority date and argue that because of that reputation the applicant’s use of its mark would be ‘likely to deceive or cause confusion’. Since this standard has been held to be lower than that needed to make out passing off or a contravention of the statutory prohibition on engaging in misleading or deceptive conduct,[89] section 60 can have a broader reach than section 42(b), and apply where X might not, in fact, be able to prevent the use of Y’s mark (that is, where Y’s use would cause ‘mere confusion’ that would not amount to passing off or misleading or deceptive conduct). Finally, section 62A sets up an opposition ground where the application for registration was made in bad faith, which might apply where Y’s application was made to inconvenience X’s intended use and its conduct ‘falls short of the standards of acceptable commercial behaviour’.[90] In addition to these opposition grounds, section 124 contains a defence to infringement of a registered mark where the defendant has continuously used its unregistered mark before the earlier of the registration date or the registered owner’s first use of its mark. The very presence of these and other provisions should impact on our understanding of section 58 and the role it ought to play as one of a complex set of provisions designed to manage the relationship between ‘use-based’ and ‘registration-based’ rights within the trade marks system.
Both the history and current context of the ‘ownership’ ground point towards it having a very narrow sphere of operation. Section 58 goes substantially further than sections 42(b) and 60 in allowing X, as the mere first user, to prevent the registration of Y’s mark without X needing to show any evidence of reputation, or even any ongoing use of its mark.[91] The only limiting conditions are that the marks are the same or substantially identical, and the goods or services offered under the marks are the same or the same kind of thing.[92] To the extent that it is thought that allowing Y to register its mark in circumstances not caught by sections 42(b) or 60 would be to reward Y’s bad faith conduct, or that X would somehow lose the ability to continue to use its mark in the face of Y’s registration, these situations are covered by sections 62A and 124 respectively. The best that can be said for section 58, therefore, is that it offers a ‘procedural and evidential shortcut’ to section 60. That is, if the opponent can show ‘double identity’ between the two marks and the two sets of goods/services, the opponent does not need to go to the expense of establishing reputation and consumer confusion under section 60.[93] However, this reading is only justifiable if courts and the Office adopt a very strict reading of what constitutes ‘substantially identical’ marks. Anything else would undermine the role played by the opposition grounds other than section 58, and would over-privilege X, a party that might well have no common law rights in relation to its mark. The consequences of the approach to substantial identity adopted by the Office and by the Court in Accor and Pham Global, which significantly expands the scope of section 58, are thus wholly undesirable.
Whereas the policy underpinnings of the ‘ownership’ ground of opposition in the TMA are shaky, the same cannot be said for the provisions of the TMA allowing registered marks to be removed from the Register on the grounds of ‘non-use’. It is trite to note that a trade mark confers social benefits, in reducing consumers’ search costs by enabling them to differentiate between traders’ goods or services, and thus warrants legal protection, only if the mark is in use.[94] Similarly, protecting unused trade marks creates barriers for other traders seeking to identify marks they can safely use.[95] These fundamental concerns are built in to the very definition of a ‘trade mark’ in section 17 of the TMA, which contemplates a sign that is ‘used, or intended to be used ... in the course of trade’ to distinguish goods or services,[96] and the TMA contains various pre-registration mechanisms to deal with applications that do not meet this criterion. Equally, it is uncontroversial to note that an efficient registration system requires the Register to convey accurate information about marks ‘which perform their statutory function’,[97] and that there is a strong public interest in ensuring that any ‘defects’ in the Register can be corrected.[98] This explains why the TMA permits third parties to apply for the removal of a trade mark from the Register if it has not been used for a period of time, post-registration.
However, despite the recognition that ‘use’ is central to why trade marks should be protected at all, and that the Register should reflect only those marks that are entitled to legal protection, the Australian system affords owners a high degree of latitude when questions arise as to whether they intend to use, or have used, their marks. To adapt the ‘Mindgym’ example from Part III(B) of the first part of this article, a trader providing online education services under the name ‘EduGym’ could potentially seek to register one or more word marks, such as EDUGYM, EDUGYM ONLINE COURSES and/or EDUGYM QUEENSLAND, as well as the stylised word ‘EduGym’ and/or a composite mark containing this stylised word and a logo. It would not be required at the time of filing to have anything more than an intention to use the mark in question.[99] Even though the Office is required in examination to consider whether such an intention exists, it presumes the existence of the intention by the applicant’s very act of filing for registration.[100] A third party could oppose registration on the basis that at the filing date the applicant lacked the intention to use the mark, but this can be difficult for an opponent to prove, especially since it has been held that evidence that a mark is not in use does not shift the onus to the applicant to justify its intentions.[101] From the date the mark is entered on the Register, the registered owner has a three year grace period within which a non-use challenge cannot be brought.[102] After this time, an application for removal on the basis that the registered mark was not used for a three year period can be successfully rebutted by the owner pointing to evidence of minimal, even inadvertent, use.[103] And even if no use can be established, the Office or court has the discretion not to order the removal of the mark from the Register if ‘reasonable’ to do so,[104] which does not require the registered owner to show the existence of ‘exceptional’ circumstances.[105]
The TMA thus contains numerous procedural and substantive provisions that are, in fact, highly favourable to owners. These ensure that a mark, even when unused, can be easily registered, stay registered for a long period of time, and not be vulnerable to removal. Therefore, to give effect to the underlying policy of the TMA and to ensure the accuracy of the Register, it would seem that a strict approach ought to be taken to the assessment of the evidence of the ‘use’ of the registered mark that is put forward by the owner in non-use proceedings. This raises the question of why any latitude should be afforded to an owner that cannot point to evidence of at least minimal use of its mark in the precise form in which it was registered. If the only use is of a variant of the mark as registered – for example, if the use is of EDUGYM when the registered mark is EDUGYM ONLINE COURSES, or the use is the stylised word ‘EduGym’ when the registered mark is the composite mark (‘EduGym’ plus logo) – why should any sympathy be extended to the registered owner, particularly when it had the opportunity to apply for registration of the variant in the first place?
Permitting the ‘preservative’ use of something different from what is registered can only be justified by commercial expedience. Owners may wish for commercial reasons to make minor variations to their registered marks, as used over time, and not have to file for new registrations (with later priority dates). Similar logic applies to the provisions of the TMA that allow for amendments to be made to representations of marks, either before or after registration. However, there are costs involved in recognising any form of preservative use that deviates from what is on the Register, or in permitting amendments to the representation of a mark. This is because the scope of an owner’s rights extends to prevent others from using not only the mark as registered, but also ‘deceptively similar’ marks.[106] If an owner is allowed to maintain a registration based on the use of a variant, or is allowed to amend its representation, there is a risk that the owner’s rights will be artificially, and unjustifiably, extended as a result.[107]
For this reason, only the use of a variant of a registered mark that does ‘not substantially affect the identity’ of the registered mark can be taken into account under section 100 in non-use proceedings. Similarly, only amendments that do ‘not substantially affect the identity’ of the representation of a mark are permissible under sections 65 and 83. The concern in these cases is to ensure that an owner cannot extend the scope of its monopoly without filing for a new registration. The idea is that if the mark as used, or the amended representation, is ‘practically the same’[108] as the registered mark or earlier representation, or the two are ‘so alike that they can be described as one and the same’,[109] there will be no increase in the scope of the owner’s rights, and thus no impact on the range of other marks that remain available for other traders to use. Since ‘even a minor [variation of or] alteration in a mark will, to some degree, affect the scope of its registration’,[110] the term ‘not substantially affect the [mark’s] identity’ must be interpreted strictly.
The above policy concerns do not appear to have been taken into account by the Office and Federal Court in the decisions outlined in Part III of the first part of this article adopting an expansive interpretation of ‘substantial identity’. As an illustration of the problems that might result from the expansive approach, if EduGym’s only registered mark is the composite mark (stylised ‘EduGym’ plus logo), but it is considered that its use of the stylised ‘EduGym’ alone does not substantially affect the identity of the composite mark,[111] this party will maintain rights to prevent other traders from using a mark that is deceptively similar to the unused composite mark. Other traders consulting the Register, concerned that their marks might potentially be deceptively similar to the composite mark (in particular, the unused logo element) will have to incur additional, unnecessary costs in selecting a sufficiently different mark or rebranding, even if their preferred marks are not deceptively similar to the stylised word ‘EduGym’. This concern is not far-fetched, since the scope of deceptive similarity is potentially broad, and often difficult to determine. It is possible to point to examples of visually dissimilar marks that have been held to have been deceptively similar under Australian law, applying the test of ‘imperfect recollection’,[112] and it is these far-reaching decisions that are likely to shape the practices of well-advised later market entrants looking for marks they can safely use.
The expansive approach to ‘substantial identity’ has the potential to over-privilege incumbent trade mark owners seeking to maintain rights in marks they have not used or seeking to amend their representations. These developments are unwelcome and are a further reason for rejecting the expansive approach. To the extent it is considered that a return to the strict reading of substantial identity might operate harshly against some owners, it must be remembered that such parties could always have filed for registration of the marks they actually used, and that in non-use proceedings they have the opportunity to persuade the Office or court that it is ‘reasonable’ not to order the removal of their unused marks from the Register.
As noted at the outset of the first part of this article, where the use of a trade mark is required to be shown under any provision of the TMA, the Registrar or court may find that such use can be established by use ‘with additions or alterations that do not substantially affect the identity’ of the mark.[113] The expansive interpretation of ‘substantial identity’ has the potential to disrupt the operation of these provisions, which go to fundamental issues such as prior continuous use of a mark under section 44(4), and the demonstration of acquired distinctiveness under sections 41(3)–(4).[114] However, there is one area where the expansive interpretation is likely to be of particular concern, which has not to date been given any attention by either the Office or the courts. This is in the context of section 146, which makes it an indictable or summary offence intentionally to apply a mark, without the owner’s permission, to goods or services that are to be dealt with in trade if that mark is ‘substantially identical’ with a registered mark.
Provisions making it an offence to ‘forge’ and to ‘falsely apply’ a registered trade mark were contained in the 1905 Act, applying where the defendant had made or applied either the registered mark or a ‘mark so nearly resembling it as to be likely to deceive’ (that is, a ‘deceptively similar’ mark).[115] Similar offence provisions were retained in the Trade Marks Act 1955 (Cth) (‘1955 Act’), with a crucial difference: they applied where the defendant had made or applied the mark, or a ‘substantially identical’ mark, but not a deceptively similar mark.[116] The precise reason for this change remains unclear,[117] but it would have been expected that ‘substantially identical’ would have been interpreted strictly in this context, and certainly more narrowly than deceptive similarity, consistent with the High Court’s reasoning in Schweppes.
When Australia became a member of the World Trade Organization it was required to ensure, under the Agreement on Trade-Related Aspects of Intellectual Property Rights (‘TRIPS Agreement’), that ‘criminal procedures and penalties [would] be applied ... in cases of wilful trademark counterfeiting’.[118] ‘Counterfeit trademark goods’ were defined in the TRIPS Agreement to be goods or packaging bearing a mark ‘identical to the trademark validly registered in respect of such goods, or which cannot be distinguished in its essential aspects from such a trademark’.[119] In enacting new trade marks legislation to ensure consistency with these TRIPS Agreement requirements,[120] the Commonwealth Parliament retained a version of the ‘false application’ offence provision from the 1955 Act that similarly extended to the application of a ‘substantially identical’ mark.[121] And even though the Explanatory Memorandum to the Trade Marks Bill 1995 (Cth) stated that the relevant offence provision ‘relate[d] to the counterfeiting and forgery of trade marks, thus serving to protect the commercial activities of the registered owner of a trade mark and preventing deception of consumers’,[122] Parliament did not see fit to extend the ‘false application’ provision’ in the TMA to cases of mere ‘deceptive similarity’.
Given how ‘substantial identity’ had come to be understood by 1995, section 146 of the TMA was clearly drafted on the basis that criminal liability should extend only to those situations where the mark falsely applied by the defendant was practically the same as the registered mark[123] – something that a defendant could easily ascertain from a side-by-side visual comparison between its mark and the mark on the Register. While section 146 has since been amended to create both indicatable and summary offences,[124] and to add fault elements to the element of the offences that the mark applied by the defendant be, or be substantially identical with, the registered mark,[125] the reach of these offences remains unchanged.[126]
The effect of the Full Court of the Federal Court’s adoption of a more expansive test for ‘substantial identity’ in Accor and Pham Global is to have potentially widened the reach of these offences, quite inadvertently.[127] By way of example, in Part II(A), reference was made to pre-Accor court decisions in which pairs of ‘Opal Collection’ marks, ‘Bing’ marks, ‘Anchorage Capital’ marks and ‘Optical 88’ marks were held not to be substantially identical with each other.[128] Assuming that the first mark in each of those pairs is registered and owned by A, it now entirely possible that B’s application of the other mark or marks in each case, without A’s consent, would not only constitute an infringement but might amount to a criminal offence, provided the fault elements are satisfied. Similarly, adapting the facts of Accor and Pham Global, the application of RIVER LIGHTS to lighting products, in the face of a registration for BRISBANE RIVER LIGHTS for such goods, might now give rise to criminal liability, as might the application to medical testing kits of ‘PhettleDiagnostics’ with a simple heart-shaped logo, in the face of a registration a composite mark consisting of ‘PHettle Healthcare Testing’ with a ‘similar but somewhat different’ logo[129] for the same goods. While prosecutions are likely to be rare, the key point to note is that these outcomes would have been unthinkable when section 146 of the TMA was enacted. It was never contemplated that marks whose degree of resemblance could be assessed only by reference to a modified deceptive similarity inquiry (where the common presence of ‘essential features’ becomes determinative, even if the presence of other elements mean that the marks have prominent visual and aural differences on a side-by-side comparison) would fall within the scope of section 146.
As a final point, an expansive interpretation of ‘substantial identity’ in section 146 runs contrary to established principles of statutory interpretation. It has long been recognised that particular care is needed when interpreting criminal provisions to ensure that they do not operate in an unduly harsh manner on defendants, given the potential penalties involved.[130] As Kiefel CJ and Keane J held in R v A2, reiterating statements made in earlier High Court decisions, although a ‘statutory offence provision is to be construed by reference to the ordinary rules of construction’,
[i]t is nevertheless accepted that offence provisions may have serious consequences. This suggests the need for caution in accepting any ‘loose’ construction of an offence provision. The language of a penal provision should not be unduly stretched or extended. Any real ambiguity as to meaning is to be resolved in favour of an accused.[131]
These concerns provide a further set of reasons for rejecting an ‘extended’ interpretation of ‘substantial identity’ in section 146 and, by extension, throughout the TMA.
Not all decisions since Accor and Pham Global have been characterised by an expansive approach to substantial identity, and it is important to consider whether these decisions show that there might, in fact, be space for the courts and the Office to return to something close to the strict position that existed before Accor.
In the Federal Court trial decision in Urban Alley Brewery Pty Ltd v La Sirène Pty Ltd,[132] in which URBAN BREWING COMPANY was held not to be substantially identical with URBAN ALE, Pham Global was read as not requiring the discounting of descriptive subject matter. After citing both Shell v Esso and Pham Global, O’Bryan J held that while there is an ‘obvious relationship’ between ‘brewing company’ and ‘ale’, and that those words ‘perform a descriptive function within the respective marks’, the different meaning of the words was decisive.[133] The fact that his Honour did not seek to discount the descriptive material (as the respondent had urged)[134] is significant, although it is arguable that if the descriptive content in both marks had conveyed a similar meaning, Pham Global might have dictated a different outcome. On appeal, in the course of considering the appellant’s argument that the marks were not deceptively similar, the Full Court sought to reinforce the distinction between the tests for substantial identity and deceptive similarity. In relation to the former, and without citing Accor or Pham Global, the Full Court noted that the side-by-side assessment involves a ‘studied comparison’ that ‘highlights the differences between the marks just as much as it shows their sameness’.[135] In relation to the latter, it held that the test is ‘fundamentally different’, involving not a studied comparison but rather a comparison between a mark and a hypothetically recalled impression of another mark,[136] and that the test is ‘directed to a different end’ – namely, to determine resemblance that would cause confusion.[137] This orthodox view is to be contrasted with Pham Global which, as explained in Part II(A) above, treats substantial identity as a modified form of deceptive similarity.
In some recent cases and decisions involving comparisons between word marks and composite marks, it is possible to see decision-makers giving significant weight to the device elements in the composite marks, classifying these as ‘essential features’, and thus enabling findings that the marks in question lacked substantially identity. Thus, in the Federal Circuit Court case of Chris & Dora Di Lorenzo Partnership v Denversian Pty Ltd, Judge Baird considered that the device element in the mark below was ‘not divisible or discountable from the word elements’[138] but rather ‘substantially inform[ed] and affect[ed] the trade mark’s identity’,[139] such that the composite mark was not substantially identical with the word mark BLACK SHEEP.[140]
At the Office, in Active Skin Pty Ltd v Yey Pty Ltd the Hearing Officer considered that visually prominent but conceptually generic elements of the composite mark below (for cosmetics) were essential features, and that the absence of these features in the word mark ACTIVE SKIN meant that the marks were not substantially identical.[141]
In other recent decisions, courts have taken an even stricter view of substantial identity and its analogues, although without engaging directly with Accor or Pham Global. For example, in the infringement case of Carshare Australia Pty Ltd v Singh,[142] Judge Baird held that ‘Go Get Cabs’ was not substantially identical with the registered mark GOGET for car hire services. Citing Shell v Esso, her Honour acknowledged that ‘Cabs’ was descriptive, and that for the purposes of assessing deceptive similarity it would not change the overall impression of the respondent’s mark as being ‘Go Get’,[143] but that its presence could not be ‘ignored or discounted’[144] in the test for substantial identity, where it played a ‘greater role’.[145] More recently, in the Federal Court case of Community First Credit Union Ltd v Bendigo & Adelaide Bank Ltd,[146] one issue was whether the applicant, in attempting to make a case that it had used the word marks COMMUNITY FIRST BANK and COMMUNITY FIRST MUTUAL BANK (for personal banking services), could rely on evidence of its use of COMMUNITY FIRST. Markovic J considered, in obiter dicta, that the use of the latter mark was with alterations that did substantially affect the identity of the first two marks.[147] Her Honour acknowledged that while ‘bank’ and ‘mutual bank’ were descriptive terms and ‘not the key cognitive cues in the marks’, their presence meant that the marks conveyed a different type of organisation, compared with ‘Community First’.[148] The outcomes in these cases are consistent with the High Court’s reasoning in Shell v Esso, since the pairs of marks in question are sufficiently visually dissimilar, and the decisions would have been unremarkable before Accor.[149] They are, however, in tension with the Full Court of the Federal Court’s reformulation of the Shell v Esso test in Pham Global.
Ultimately, there is a danger in hoping that courts and the Office will find ways of manoeuvring around Accor and Pham Global as a workable solution in this area of the law. Such a practice will only produce further uncertainty, since it is likely to leave those in the trade marks community in serious doubt as to how any tribunal might interpret the scope of key grounds in the TMA, such as those relating to ownership, non-use and amendments to representations. Therefore, to rectify the problems with ‘substantial identity’, two non-mutually exclusive options should be considered.
The first is that the Full Court of the Federal Court should revisit the issue of ‘substantial identity’ when the opportunity next arises. It should overturn both Accor and Pham Global on the basis that both decisions are plainly wrong.[150] For the reasons explained in Part II above, when read in light of the first part of this article, Accor and Pham Global involve misapplications of Shell v Esso, are inconsistent with the High Court’s earlier decisions in Schweppes and Innes, and cannot be supported by the High Court’s decision in Gallo. In addition, as demonstrated in Part III above, Accor and Pham Global involve an interpretation of section 58 that does not give effect to the purposes or objects of the TMA,[151] specifically the ownership, non-use, amendment of representations and ‘false application’ offence provisions.
A second way forward would involve legislative reform. If the goal of such reform is to ensure only that ‘substantial identity’ and its analogues in the TMA are interpreted narrowly, in line with the pre-Accor cases and decisions described in Part II of the first part of this article, careful drafting would be needed. It might, in fact, be preferable for the legislature to use this opportunity to consider more fundamental changes to the provisions of the TMA in which ‘substantial identity’ and its analogues play a role (which it could do even if the Full Court of the Federal Court were to overturn Accor and Pham Global).
Most obviously, little would be lost if ‘substantially identical’ were changed to ‘identical’ in sections 44 and 120, given the role played by ‘deceptive similarity’ in these sections.[152] In the context of the ‘false application’ offences in section 146, it would be worth considering using the precise language of the TRIPS Agreement[153] and amending the section so that it applies where the defendant intentionally applies a sign or mark that ‘cannot be distinguished in its essential aspects’ from the registered mark.[154] The legislature might consider whether an applicant for registration or a registered owner should be permitted to amend the representation of its mark at all, noting that New Zealand does not allow for such amendments, post-registration.[155] More controversially, the legislature might reconsider whether a registered owner should be able to point to the use of anything other than its mark as registered to rebut an allegation of non-use, given that the Registrar and court retain a discretion in section 101(3) not to order the removal of an unused mark ‘[i]f satisfied that it is reasonable to do so’.
The most far-reaching reform would be to consider abolishing the ‘ownership’ ground of opposition in section 58. This ground reflects ‘common law’ principles that only formed part of the common law from 1863 until 1909, and its retention in the TMA is difficult to justify. Removing the provision would cause little hardship to would-be opponents, who would still have the section 42(b) and section 60 grounds at their disposal if they are able to demonstrate reputation in their unregistered marks, and section 62A if the applicant has engaged in bad faith conduct.[156] At the very least, if it is considered desirable to maintain section 58 on the basis that it is a useful ‘double identity’ shortcut to section 60, thought should be given to restricting its operation to situations of absolute identity between the two marks and two sets of goods/services,[157] coupled with a reconsideration of the standard of ‘use’ required to establish ownership.[158]
Finally, and at a more general level, the problems identified in both parts of this article should not be seen to be isolated to the question of ‘substantial identity’. More attention needs to be paid to whether tribunals are interpreting other concepts in specific provisions of the TMA with sufficient regard to how those concepts operate within the Act as a whole, and to the normative underpinnings of the entire trade marks registration system,[159] to ensure that the competing interests of traders, consumers and the general public[160] are fully accommodated. Failing to do so runs the risk of creating incoherence within the trade marks system, as Accor and Pham Global have unfortunately shown.
* Professor, School of Private and Commercial Law, Faculty of Law & Justice, UNSW Sydney <https://orcid.org/ 0000-000[1]-8980-5099>. My thanks go to Rob Batty, Robert Burrell, Anna Harley, Megan Jones, Bill Ladas and the three anonymous referees.
[1] Michael Handler, ‘Trade Mark Law’s Identity Crisis’ (Pt 1) [2021] UNSWLawJl 14; (2021) 44(1) University of New South Wales Law Journal 394.
[2] [2017] FCAFC 56; (2017) 345 ALR 205 (‘Accor’).
[3] [2017] FCAFC 83; (2017) 251 FCR 379 (‘Pham Global’).
[4] [1963] HCA 66; (1961) 109 CLR 407 (‘Shell v Esso’).
[5] Warwick Rothnie, ‘Pham Global 2: The New Law of Substantial Identity’, IPWars.com (Blog Post, 19 June 2017) <http://ipwars.com/2017/06/19/pham-global-2-the-new-law-of-substantial-identity/> .
[6] [1963] HCA 66; (1961) 109 CLR 407, 414–15 (Windeyer J).
[7] See Torpedoes Sportswear Pty Ltd v Thorpedo Enterprises Pty Ltd [2003] FCA 901; (2003) 132 FCR 326, 335–6 [58]–[60] (Bennett J).
[8] [1910] HCA 36; (1910) 11 CLR 347, 357 (‘Schweppes’).
[9] [1921] HCA 16; (1921) 29 CLR 277, 282 (Knox CJ, Gavan Duffy, Rich and Starke JJ) (‘Innes’). The Court was considering the term ‘nearly identical’ in section 27 of the Trade Marks Act 1905 (Cth).
[10] Lonza Ltd v Kantfield Pty Ltd (1995) 33 IPR 396, 400 (Deputy Registrar Hardie) (Trade Marks Office). For a recent illustration, consistent with this approach, see Singtel Optus Pty Ltd v Optum Inc [2018] FCA 575; (2018) 140 IPR 1, 10 [13] (Davies J) (Federal Court) (OPTUM not substantially identical with OPTUS).
[11] See United Airlines Inc v United Airways Ltd [2011] EWHC 2411 (Ch), [40]–[41] (Vos J) (considering UNITED AIRLINES and UNITED AIRWAYS to be ‘identical’ under United Kingdom (‘UK’) law).
[12] A related problem, as noted by Davison and Horak, is that the approach taken in both Accor and Pham Global, perhaps stemming from the Full Court’s attempt to equate the ‘dominant cognitive cues’ within a mark to the ‘essential features’ of the mark, potentially results in some prominent, non-descriptive features of the mark being mischaracterised as ‘non-essential’ (such as the gold stars above the words ‘Harbour Lights’ in what the trial judge in the Accor litigation found to be a composite mark: Accor Australia & New Zealand Hospitality Pty Ltd v Liv Pty Ltd [2015] FCA 554; (2015) 112 IPR 494, 515 [108] (Rangiah J) (Federal Court)). See Mark Davison and Ian Horak, Thomson Reuters, Shanahan’s Australian Law of Trade Marks and Passing Off (online at 30 March 2021) [30.505]. See also Mark J Davison, Ann L Monotti and Leanne Wiseman, Australian Intellectual Property Law (Cambridge University Press, 4th ed, 2020) 648 (arguing that the Full Court ‘focused too greatly on the word signs rather than the totality of the two trade marks’).
[13] Re Lynson Australia Pty Ltd (1987) 9 IPR 350, 351 (Chief Assistant Registrar Farquhar) (Trade Marks Office).
[14] Cody Opal (Australia) Pty Ltd v Dimasi [2004] SASC 357; (2004) 64 IPR 378, 401 [122]–[124] (Mullighan J) (Supreme Court of South Australia).
[15] Bing! Software Pty Ltd v Bing Technologies Pty Ltd (No 1) [2008] FCA 1760; (2008) 79 IPR 454, 464–5 [32]–[37] (Collier J) (Federal Court).
[16] Anchorage Capital Partners Pty Ltd v ACPA Pty Ltd [2015] FCA 882; (2015) 115 IPR 67, 77 [39] (Perram J) (Federal Court).
[17] Optical 88 Ltd v Optical 88 Pty Ltd (No 2) [2010] FCA 1380; (2010) 275 ALR 526, 576 [256]–[258], 577 [261] (Yates J) (Federal Court) (‘Optical 88 Trial’).
[18] Ibid 577 [261]. The square device containing four circles contained within both marks was held to be an original artistic work, of which Joseph CC Wong is the author: at 531 [13], 582 [285]–[289].
[19] Optical 88 Ltd v Optical 88 Pty Ltd [2011] FCAFC 130; (2011) 197 FCR 67, 80 [39] (Cowdroy, Middleton and Jagot JJ).
[20] See also Insight Radiology Pty Ltd v Insight Clinical Imaging Pty Ltd [2016] FCA 1406; (2016) 122 IPR 232, 239–40 [18] (Davies J) (Federal Court).
[21] See Saville Perfumery Ltd v June Perfect Ltd (1941) 58 RPC 147, 162 (Sir Wilfrid Greene MR).
[22] Southcorp Brands Pty Ltd v Australia Rush Rich Winery Pty Ltd [2019] FCA 720; (2019) 369 ALR 299, 308 [60] (Beach J) (Federal Court).
[23] Warner-Lambert Co v Harel (1995) 32 IPR 189, 192 (Hearing Officer Homann) (Trade Marks Office).
[24] Schweppes [1910] HCA 36; (1910) 11 CLR 347, 357 (Isaacs J).
[25] Ibid (noting that the registered owner ‘is not called upon in such a case ... to prove probability of deception’).
[26] Ibid 358. Griffith CJ also considered whether the marks were substantially identical based on a visual comparison, and separately considered whether the marks were deceptively similar, applying a broader test: at 352–4.
[27] The High Court has recognised this to be an object of the Trade Marks Act 1995 (Cth) (‘TMA’): Cantarella Bros Pty Ltd v Modena Trading Pty Ltd [2014] HCA 48; (2014) 254 CLR 337, 345 [17] (French CJ, Hayne, Crennan and Kiefel JJ).
[28] See Robert Burrell and Michael Handler, Australian Trade Mark Law (Oxford University Press, 2nd ed, 2016) 266.
[29] Cf PB Foods Ltd v Malanda Dairyfoods Ltd [1999] FCA 1602; (1999) 47 IPR 47, 54–5 [38] (Carr J) (Federal Court) (‘PB Foods’).
[30] Cf Accor [2017] FCAFC 56; (2017) 345 ALR 205, 251 [233] (Greenwood, Besanko and Katzmann JJ).
[31] Cf Pham Global [2017] FCAFC 83; (2017) 251 FCR 379, 393 [56] (Greenwood, Jagot and Beach JJ).
[32] Innes [1921] HCA 16; (1921) 29 CLR 277, 281 (Knox CJ, Gavan Duffy, Rich and Starke JJ) (emphasis added).
[33] [1994] FCA 936; (1994) 120 ALR 495, 513 (Gummow J) (Federal Court) (‘Carnival’).
[34] [2007] FCAFC 184; (2007) 164 FCR 506, 529 [86] (Allsop J) (‘Colorado’).
[35] A further reason why it is problematic to have confusion play any role in section 58 is that it disrupts the logic of permitting a mark to be registered on the basis of honest concurrent use. Assume, for example, that X has registered FORTEX AIRWAYS for aviation services and Y applies to register FORTEX for the same services. Y’s application will be blocked by section 44(2) (since the marks are, at least, deceptively similar), but Y might be able to demonstrate that there has been ‘honest concurrent use’ of FORTEX AIRWAYS and FORTEX, such that its application should be accepted for registration under section 44(3)(a). In accepting the application on this basis the decision-maker will effectively be required to accept that there is a ‘prospect of deception and confusion’, albeit a tolerable one, between the two marks: Campomar Sociedad Limitada v Nike International Ltd [2000] HCA 12; (2000) 202 CLR 45, 68 [50] (Gleeson CJ, Gaudron, McHugh, Gummow, Kirby, Hayne and Callinan JJ) (‘Campomar’). If the marks are considered to be deceptively similar but not substantially identical (as would have been the case, pre-Accor), then X would not be able to oppose Y’s registration of FORTEX under section 58. However, applying the Pham Global test for substantial identity – which is in effect a modified deceptive similarity inquiry that looks to how consumers would respond to the two marks and whether they would see the marks as indicating a common trade source – X would be successful under section 58 if its first use of FORTEX AIRWAYS predates Y’s first use of FORTEX. This completely undermines the rationale for allowing Y’s mark to have been accepted for registration under section 44(3)(a) in the first place. See further the discussion in Part III(B) of the first part of this article, including options for revitalising the role of honest concurrent use in the TMA 1995 (Cth).
[36] TMA 1995 (Cth) ss 7(1), 100(2)(a)–(b)(i), (3)(a)–(b)(i).
[38] Optical 88 Trial [2010] FCA 1380; (2010) 275 ALR 526, 576 [256] (Yates J) (Federal Court).
[39] See Planet Plumbing SW Works Pty Ltd v Green Planet Maintenance Pty Ltd [2021] ATMO 32, [18] (Hearing Officer Barbey) (not applying a test focusing on the ‘perception of consumers and whether they are likely to identify the trade origin of the services’ in interpreting the scope of section 100(3)(a)).
[40] (2020) 384 ALR 230, 244–7 [90]–[108] (Middleton, Yates and Lee JJ).
[41] [2010] HCA 15; (2010) 241 CLR 144 (‘Gallo’).
[42] For Trade Marks Office (‘Office’) decisions, see, eg, Buckley v South Cone Inc [2013] ATMO 29, [25]–[26] (Hearing Officer Thompson); American Airlines Inc v Advantagecard Pty Ltd (2013) 103 IPR 110, 122 [36]–[37] (Hearing Officer Thompson); MCD Asia Pacific LLC v Dogon [2014] ATMO 52, [19] (Hearing Officer Thompson); Insight Clinical Imaging v Insight Radiology Pty Ltd [2014] ATMO 85; (2014) 109 IPR 86, 91 [22] (Hearing Officer Thompson). In the Federal Court, see Anchorage Capital Partners Pty Ltd v ACPA Pty Ltd [2018] FCAFC 6; (2018) 259 FCR 514, 530–1 [59] (Nicholas, Yates and Beach JJ); Dunlop Aircraft Tyres Ltd v Goodyear Tire & Rubber Co [2018] FCA 1014; (2018) 262 FCR 76, 104–5 [131] (Nicholas J).
[43] Gallo [2010] HCA 15; (2010) 241 CLR 144, 170 [69]. The reproduction of the label is from E & J Gallo Winery v Lion Nathan Australia Pty Ltd [2008] FCA 934; (2008) 77 IPR 69, 107 [184] (Flick J) (Federal Court). If the ‘bare foot’ image within this label is an original artistic work, it is anonymous (although it may be that this image is a reproduction of a painting by Bonnie Harvey, one of the founders of Barefoot Wine: Barefoot Wines, ‘Barefoot Pinot Grigio: A Balance of Charm and Boldness’, Just Wines (Blog Post, 15 June 2015) <https://justwines.com.au/blog/barefoot-pinot-grigio-a-balance-of-charm-and-boldness/>).
[44] Gallo [2010] HCA 15; (2010) 241 CLR 144, 169 [67] (French CJ, Gummow, Crennan and Bell JJ).
[45] See E & J Gallo Winery v Lion Nathan Australia Pty Ltd [2008] FCA 934; (2008) 77 IPR 69, 108 [186] (Flick J) (Federal Court).
[46] Gallo [2010] HCA 15; (2010) 241 CLR 144, 170 [68].
[47] Ibid 170 [69].
[48] See Burrell and Handler, Australian Trade Mark Law (n 28) 349.
[49] Rules under the Trade Marks Registration Act 1875, 38 & 39 Vict, c 91, r 17.
[50] This might have been the case where one applicant was the first user of the mark, and the other applicant was a former employee of the first applicant who subsequently adopted a nearly identical version of the mark, and both applied for registration of their marks at much the same time.
[51] Trade Marks Registration Act 1875, 38 & 39 Vict, c 91, s 6.
[52] Patents, Designs, and Trade Marks Act 1883, 46 & 47 Vict, c 57, ss 72(1)–(2) (as amended by the Patents, Designs, and Trade Marks Act 1888, 51 & 52 Vict, c 50).
[53] See generally Phillip Johnson, ‘The Rise and Fall of Honest Concurrent Use’ in Ilanah Simon Fhima (ed), Trade Mark Law and Sharing Names: Exploring Use of the Same Mark by Multiple Undertakings (Edward Elgar, 2009) 31, 35–40.
[54] See, eg, Re The Walkden Aërated Waters Co’s Application (1877) 54 LJ Ch 394 (‘Walkden’); Re Jelley, Son and Jones’s Application (1878) 51 LJ Ch 639.
[55] See, eg, DM Kerly, The Law of Trade-Marks, Trade-Names and Merchandise Marks (Sweet & Maxwell, 1894) 176.
[56] Re Brook’s Trade Marks (1878) 26 WR 791, 792 (emphasis added). See also Re Trade-Mark ‘White Rose’ [1885] UKLawRpCh 159; (1885) 30 Ch D 505 (allowing a merely similar mark on the Register). See generally ‘Registration of Trade-Marks: The Three-Mark Rule’ (1885) 29 Solicitors’ Journal and Reporter 414.
[57] Walkden (1877) 54 LJ Ch 394.
[58] Lewis Boyd Sebastian, The Law of Trade Marks and Their Registration (Stevens & Sons, 3rd ed, 1890) 58 (referring to Walkden (1877) 54 LJ Ch 394); Kerly (n 55) 178 (referring to Re Trade-Mark ‘White Rose’ [1885] UKLawRpCh 159; (1885) 30 Ch D 505).
[59] See, eg, Re Trade-Mark of John Dewhurst & Sons Ltd [1896] UKLawRpCh 71; [1896] 2 Ch 137, 148 (Lindley LJ); Re Ehrmann’s Applications [1897] UKLawRpCh 110; [1897] 2 Ch 495, 501 (Stirling J).
[60] DM Kerly and FG Underhay, The Law of Trade Marks and Trade Name (Sweet & Maxwell, 3rd ed, 1908) 264–5.
[61] Trade Marks Act 1905, 5 Edw 7, c 15, s 21.
[62] [1916] 1 Ch 304, 310 (emphasis added).
[63] Admittedly, the effect of this interpretation is that section 53 of the Trade Marks Act 1905 (Cth) would have contained a tautology (carried through to the Trade Marks Act 1955 (Cth) and TMA 1995 (Cth)). However, such tautologies are not unknown in Australian trade mark law. As noted above, under section 25 of the Trade Marks Act 1905 (Cth) an application for registration was to be rejected if it so nearly resembled an already registered mark ‘as to be likely to deceive’. In the Trade Marks Act 1955 (Cth), the equivalent provision (section 33(1), as interpreted by section 6(3)) applied if the resemblance was such ‘as to be likely to deceive or cause confusion’ (emphasis added). However, a year before this Act was passed, in Southern Cross Refrigerating Co v Toowoomba Foundry Pty Ltd (1954) 91 CLR 592, the High Court had held that the term ‘likely to deceive’ was broad enough to catch mere confusion: at 595 (Kitto J) (supported on appeal: at 608 (Dixon CJ, McTiernan, Webb, Fullagar and Taylor JJ)). This meant that the addition of the words ‘or cause confusion’ in the Trade Marks Act 1955 (Cth) did ‘no more than confirm the existing law’: DR Shanahan, Australian Trade Mark Law and Practice (Lawbook, 1982) 140. For other possible redundancies in the TMA 1995 (Cth), see Burrell and Handler, Australian Trade Mark Law (n 28) 247–8.
[64] Innes [1921] HCA 16; (1921) 29 CLR 277, 281 (Knox CJ, Gavan Duffy, Rich and Starke JJ).
[65] Ibid.
[66] Ibid.
[67] Trade Marks Registration Act 1875, 38 & 39 Vict, c 91, s 10.
[68] Re Chorlton & Dugdale’s Trade Mark (1885) 53 LT (NS) 337, 339 (Pearson J).
[69] Re Grossmith’s Trade Mark ‘Emollio’ (1889) 60 LT (NS) 612, 613 (Kay J).
[70] Colorado [2007] FCAFC 184; (2007) 164 FCR 506, 535–6 [108].
[71] Food Channel Network Pty Ltd v Television Food Network GP [2010] FCAFC 58; (2010) 185 FCR 9, 24 [49] (Keane CJ, Stone and Jagot JJ).
[72] Trade Marks Registration Act 1875, 38 & 39 Vict, c 91, s 3; Kerly (n 55) 60.
[73] Accor [2017] FCAFC 56; (2017) 345 ALR 205, 240 [170] (Greenwood, Besanko and Katzmann JJ), citing PB Foods [1999] FCA 1602; (1999) 47 IPR 47, 60–1 [78]–[80] (Carr J) (Federal Court)).
[74] Edelsten v Edelsten [1872] UKLawRpCh 62; (1863) 1 De G J & S 185; 46 ER 72, 78 (Lord Westbury LC). See also Clement v Maddick [1859] EngR 15; (1859) 5(1) The Jurist 592; 65 ER 841, 842 (Stuart V-C); Cartier v Carlile [1862] EngR 514; (1862) 31 Beav 292; 54 ER 1151, 1153 (Romilly MR); Hall v Barrows [1863] EngR 1099; (1863) 4 De G J & S 150; 46 ER 873, 876 (Lord Westbury LC); M’Andrew v Bassett [1864] EngR 448; (1864) 4 De G J & S 380; 46 ER 965, 967 (Lord Westbury LC); Leather Cloth Co Ltd v American Leather Cloth Co Ltd [1863] EngR 1050; (1863) 4 De G J & S 137; 46 ER 868, 870 (Lord Westbury LC), affd [1865] EngR 199; (1865) 11 HLC 523; 11 ER 1435, 1440–1 (Lord Cranworth), 1444–5 (Lord Kingsdown). On the foundations of this development, see Lionel Bently, ‘From Communication to Thing: Historical Aspects of the Conceptualisation of Trade Marks as Property’ in Graeme B Dinwoodie and Mark D Janis (eds), Trademark Law and Theory: A Handbook of Contemporary Research (Edward Elgar, 2008) 3, 13–15 (but see also at 8–13, tracing how the common law courts and many Chancery judges in the mid-nineteenth century still maintained a requirement of fraudulent conduct).
[75] See Rob Batty, ‘Loss of Property Ownership and Registered Trade Mark Law’ (2018) 28(4) Australian Intellectual Property Journal 118, 126 (‘Loss of Property Ownership’).
[76] (1863) 32 LJ Ch 548, 551.
[77] [1864] EngR 448; (1864) 4 De G J & S 380; 46 ER 965.
[78] Ibid 967–8.
[79] Christopher Wadlow, The Law of Passing-Off: Unfair Competition by Misrepresentation (Sweet & Maxwell, 4th ed, 2011) [1-047], referring to London & Provincial Law Assurance Society v London & Provincial Joint-Stock Life Insurance Co (1847) 11(1) The Jurist 938 (Ch), where Shadwell V-C considered that ‘the Court will always have regard to ... whether there has been such a length of exclusive usage of the name under which the plaintiff carries on his trade or business’: at 938.
[80] [1896] UKLawRpAC 18; [1896] AC 199, 209–10.
[81] (1909) 26 RPC 693, 701 (Ch).
[82] (1915) 32 RPC 273, 284 (Lord Atkinson agreeing at 288, Lord Sumner agreeing at 288, Lord Parmoor agreeing at 289).
[83] X might also have lost its claim to proprietorship if it had abandoned its mark, but this required an intention to abandon rather merely non-use. See Mouson & Co v Boehm [1884] UKLawRpCh 76; (1884) 26 Ch D 398; Batty, ‘Loss of Property Ownership’ (n 75) 126–9.
[84] A relevant factor that might explain the courts’ reluctance to develop a new approach to proprietorship in registered trade mark law is that, in the absence of a defence to infringement for prior continuous use, it was thought that the first user of the mark would be prevented from using its mark in Australia by the other party’s registration: Blackadder v Good Roads Machinery Co Inc [1926] HCA 57; (1926) 38 CLR 332, 337 (Starke J) (a view also taken on appeal: at 340 (Isaacs J)). Such a defence was only introduced in section 64(1)(c) of the Trade Marks Act 1955 (Cth). In addition, in the absence of a ‘bad faith’ ground of opposition to registration, courts showed themselves to be comfortable, in cases involving local traders ‘appropriating’ foreign marks, in finding that only a ‘very small amount of use’ of the mark in Australia by the foreign mark owner was needed to make it the proprietor: The Seven Up Co v OT Ltd [1947] HCA 59; (1947) 75 CLR 203, 211 (Williams J). See also Re The Registered Trade Mark ‘Yanx’; Ex parte Amalgamated Tobacco Corporation Ltd [1951] HCA 28; (1951) 82 CLR 199, 205 (Williams J); Thunderbird Products Corporation v Thunderbird Marine Products Pty Ltd [1974] HCA 51; (1974) 131 CLR 592, 600–2 (Jacobs J).
[85] Cases such as Innes [1921] HCA 16; (1921) 29 CLR 277 and Carnival [1994] FCA 936; (1994) 120 ALR 495 are clear illustrations of this circumspect approach.
[86] Pham Global [2017] FCAFC 83; (2017) 251 FCR 379, 384 [16] (Greenwood, Jagot and Beach JJ).
[87] See Michael Handler and Robert Burrell, ‘Reconciling Use-Based and Registration-Based Rights within the Trade Mark System: What the Problems with Section 58A of the Trade Marks Act Tell Us’ (2014) 42(1) Federal Law Review 91, 113–18 (‘Reconciling Use-Based and Registration-Based Rights’).
[88] Competition and Consumer Act 2010 (Cth) sch 2.
[89] See, eg, Parkdale Custom Built Furniture Pty Ltd v Puxu Pty Ltd [1982] HCA 44; (1982) 149 CLR 191, 198 (Gibbs CJ), 210 (Mason J).
[90] DC Comics v Cheqout Pty Ltd [2013] FCA 478; (2013) 212 FCR 194, 206 [62] (Bennett J).
[91] Subject to the doctrine of abandonment, discussed above n 83. See also Riv-Oland Marble Co (Vic) Pty Ltd v Settef SpA [1988] FCA 344; (1988) 19 FCR 569, 571–2 (Bowen CJ), 602 (Lockhart J). For consideration of whether a period of non-use, resulting in a registered mark being removed from the Register, will be considered to have the same effect as abandonment of the mark, see International Consolidated Business Pty Ltd v SC Johnson & Son Inc [2019] NZCA 61; [2019] 3 NZLR 318, 343 [90]–[91] (Brown J for the Court) (New Zealand Court of Appeal).
[92] See Carnival [1994] FCA 936; (1994) 120 ALR 495, 513 (Gummow J) (Federal Court); Colorado [2007] FCAFC 184; (2007) 164 FCR 506, 512–14 [14]–[19] (Kenny J), 529–30 [87]–[89] (Allsop J).
[93] See Handler and Burrell, ‘Reconciling Use-Based and Registration-Based Rights’ (n 87) 114. See generally Robert Burrell, ‘Rethinking the Relationship between Registered and Unregistered Trade Marks’ in Graeme W Austin et al (eds), Across Intellectual Property: Essays in Honour of Sam Ricketson (Cambridge University Press, 2020) 38, 43–6 (on the logics of ‘double identity’ provisions in registered trade marks systems).
[94] See, eg, William M Landes and Richard A Posner, ‘The Economics of Trademark Law’ (1988) 78(3) Trademark Reporter 267, 280.
[95] This is likely to be a particular concern given that many ‘commonplace’ words that traders might wish to use have already been registered. On this phenomenon and its consequences, see Barton Beebe and Jeanne C Fromer, ‘Are We Running Out of Trademarks? An Empirical Study of Trademark Depletion and Congestion’ (2018) 131(4) Harvard Law Review 945.
[96] See Gallo [2010] HCA 15; (2010) 241 CLR 144, 163 [44] (French CJ, Gummow, Crennan and Bell JJ, Heydon J agreeing at 175 [87]).
[97] Health World Ltd v Shin-Sun Australia Pty Ltd [2010] HCA 13; (2010) 240 CLR 590, 597 [22] (French CJ, Gummow, Heydon and Bell JJ).
[98] For empirical research into the extent to which UK and European Union (‘EU’) trade mark registers might be ‘cluttered’ with unused marks, see Georg von Graevenitz et al, ‘Trade Mark Cluttering: An Exploratory Report’ (Report No 2012/11, Intellectual Property Office (UK), April 2012); Georg von Graevenitz, Richard Ashmead and Christine Greenhalgh, ‘Cluttering and Non-Use of Trade Marks in Europe’ (Report No 2015/48, Intellectual Property Office (UK), August 2015). For preliminary research in Australia, see Haiyang Zhang, ‘Does Trade Mark Cluttering Exist in Australia?’ (Research Paper No 07, IP Australia, June 2019).
[99] TMA 1995 (Cth) s 27(1)(b).
[100] Ibid s 31(a). For criticism of the Office’s practice, see Burrell and Handler, Australian Trade Mark Law (n 28) 35–6.
[101] TMA 1995 (Cth) s 59. See Suyen Corporation v Americana International Ltd [2010] FCA 638; (2010) 187 FCR 169, 200 [207] (Dodds-Streeton J).
[102] TMA 1995 (Cth) s 93(2) (where the filing date of the mark was on or after 25 February 2019: see Intellectual Property Laws Amendment (Productivity Commission Response Part 1 and Other Measures) Act 2018 (Cth) s 2(1), sch 1 item 46).
[103] See, eg, Gallo [2010] HCA 15; (2010) 241 CLR 144, 165 [51] (French CJ, Gummow, Crennan and Bell JJ, Heydon J agreeing at 175 [87]) (adventitious presence in Australia of goods bearing the mark enough to establish ‘use’); Dick Smith Investments Pty Ltd v Ramsey [2016] FCA 939; (2016) 120 IPR 270, 284–5 [98]–[104], 294 [156]–[157] (Katzmann J) (Federal Court) (pre-trade publicity for goods not yet in existence, and still only at a development stage, enough to establish ‘use’). For criticism, see Anna Harley, ‘Keeping the Faith: Is There a Standard of Sufficient “Use” of a Trade Mark in Australia?’ (2018) 111 Intellectual Property Forum 25.
[104] TMA 1995 (Cth) s 101(3).
[105] See Austin Nichols & Co Inc v Lodestar Anstalt (No 1) [2012] FCAFC 8; (2012) 202 FCR 490, 496 [28], 504 [69] (Jacobson, Yates and Katzmann JJ).
[106] TMA 1995 (Cth) ss 120(1)–(3).
[107] This concern was recognised by the Supreme Court of New Zealand in Crocodile International Pte Ltd v Lacoste [2017] NZSC 14; [2017] 1 NZLR 679, 701 [62] (Glazebrook J for the Court).
[108] Schweppes [1910] HCA 36; (1910) 11 CLR 347, 357 (Isaacs J).
[109] Innes [1921] HCA 16; (1921) 29 CLR 277, 282 (Knox CJ, Gavan Duffy, Rich and Starke JJ).
[110] Re Lawson (2001) AIPC 91-693, 39360 (Hearing Officer Zars) (Trade Marks Office). See also OTRIVIN Trade Mark [1967] RPC 613, 615 (Tookey QC) (Board of Trade) (noting that an overly broad amendment might result in the mark conflicting with other parties’ rights).
[111] Cf Matzka v The Mind Gym Ltd (2006) 68 IPR 339, 344 [36] (Hearing Officer McDonagh) (Trade Marks Office).
[112] See, eg, Jafferjee v Scarlett [1937] HCA 36; (1937) 57 CLR 115, 122 (Latham CJ) (finding that the ‘general idea of men in athletic costumes engaged in an athletic contest with a line across their fronts’ contained in the two marks was enough to make them deceptively similar, notwithstanding the ‘many differences’ between them (these differences being described at 120, including that the opponent’s mark contained the words ‘Double Javan’)). For more recent examples, see Effem Foods Pty Ltd v Wandella Pet Foods Pty Ltd [2006] FCA 767; (2006) 69 IPR 243, 251 [36] (Moore J) (Federal Court) (DOGS GO WACKO FOR SCHMACKOS deceptively similar to WHACKOS); Kimberly-Clark Worldwide Inc v Goulimis [2008] FCA 1415; (2008) 253 ALR 76, 85–6 [41]–[43] (Jagot J) (Federal Court) (‘“HUGGIE” MUMMY’ deceptively similar to HUGGIES).
[113] TMA 1995 (Cth) s 7(1).
[114] There is, perhaps, only a potential for disruption because the Registrar and courts retain a discretion as to whether to accept the use of a ‘substantially identical’ mark in such cases. This follows from the use of the word ‘may’ in section 7(1) of the TMA 1995 (Cth). See Acts Interpretation Act 1901 (Cth) s 33(2A); Anchorage Capital Partners Pty Ltd v ACPA Pty Ltd [2018] FCAFC 6; (2018) 259 FCR 514, 547 [141] (Nicholas, Yates and Beach JJ).
[115] Trade Marks Act 1905 (Cth) s 86(a) (as defined by s 91(a)) (forgery), s 86(b) (as defined by s 92(4)) (false application).
[116] Trade Marks Act 1955 (Cth) s 98(1)(a) (as defined by 106(1)(a)) (forgery), s 98(1)(b) (as defined by s 107(4)) (false application).
[117] In its 1938 report, the Knowles Committee referred to two clauses of a draft Trade Marks Bill that corresponded with sections 91 and 92 of the Trade Marks Act 1905 (Cth), noting only that these clauses contained a new sub-clause placing the burden on the defendant to prove the assent of the proprietor of the mark to the application to avoid liability: Report of the Committee Appointed by the Attorney-General of the Commonwealth to Consider What Alterations Are Desirable in the Trade Marks Law of the Commonwealth Together with Copy of the Report of the Committee Previously Appointed to Consider the Same Matter (Report, 1954) 55. The Dean Committee, in its 1954 report, did not comment on this part of the Knowles Committee report or the clauses that would become sections 106 and 107 of the Trade Marks Act 1955 (Cth): at 28–9. There was no discussion of this issue in Parliamentary debates over the Trade Marks Bill 1955 (Cth). In contrast, by 1955 the equivalent provision in the UK, section 2 of the Merchandise Marks Act 1887, 50 & 51 Vict, c 28, still extended to deceptively similar marks.
[118] 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 61.
[119] Ibid art 51 n 14.
[120] Explanatory Memorandum, Trade Marks Bill 1994 (Cth) 1.
[121] Trade Marks Act 1994 (Cth) s 155(2). This Act was passed by both Houses of Parliament but never commenced and was repealed by section 5 of the TMA 1995 (Cth).
[122] Explanatory Memorandum, Trade Marks Bill 1995 (Cth) 25.
[123] See TMA 1995 (Cth) s 146(2) (as originally enacted).
[124] TMA 1995 (Cth) ss 146(1)–(2) (as introduced by the Intellectual Property Laws Amendment (Raising the Bar) Act 2012 (Cth) sch 5 item 27, with effect from 15 April 2013: s 2(1)).
[125] Recklessness is the fault element for this physical element of the indictable offence in the TMA 1995 (Cth) s 146(1)(c): see Criminal Code Act 1995 (Cth) sch 1 (‘Criminal Code’) s 5.6. For a defendant to be reckless, it must be ‘aware of a substantial risk’ that the mark it is applying is, or is substantially identical with, the registered mark and ‘having regard to the circumstances known to him or her, it is unjustifiable to take the risk’: Criminal Code s 5.4(1). This can be satisfied by proof of knowledge, in the sense of an awareness that the mark the defendant is applying is substantially identical with the registered mark: Criminal Code ss 5.3, 5.4(4). Negligence is the fault element of the same physical element of the summary offence in the TMA 1995 (Cth): at s 146(2)(c); see s 146(3). For a defendant to be negligent, its conduct must involve ‘such a great falling short of the standard of care that a reasonable person would exercise in the circumstances’ and ‘such a high risk’ that the mark it is applying is, or is substantially identical with, a registered mark ‘that the conduct merits criminal punishment for the offence’: Criminal Code s 5.5.
[126] See Seiko Epson Corporation v Calidad Pty Ltd [2017] FCA 1403; (2017) 133 IPR 1, 74 [373] (Burley J) (Federal Court) (stating that ‘it is apparent that the overall objectives of ss 145 to 148 were not intended to be materially altered by the amendments to the provisions’).
[127] See also Mark P McKenna, ‘Criminal Trademark Enforcement and the Problem of Inevitable Creep’ (2017) 51(4) Akron Law Review 989, observing a somewhat similar trend in the United States. Under the Trademark Counterfeiting Act of 1984, 18 USC § 2320(f)(1)(B) (1984), a ‘counterfeit mark’ is defined as ‘a spurious designation that is identical with, or substantially indistinguishable from, a designation as to which the remedies of the Lanham Act are made available’ (emphasis added). McKenna has argued that courts have at times ‘watered down’ the requirement of mark similarity to find defendants guilty of counterfeiting in inappropriate circumstances: at 1008–13.
[128] See above nn 14–17.
[129] This is how the ‘Insight Clinical Imaging’ and ‘Insight Radiology’ composite marks were described in Urban Alley Brewery Pty Ltd v La Sirène Pty Ltd [2020] FCA 82; (2020) 150 IPR 11, 42 [149] (O’Bryan J) (Federal Court) (‘Urban Alley Trial’).
[130] Under section 146 of the TMA 1995 (Cth), the penalty for the indictable offence is 5 years imprisonment and/or 550 penalty units, while the penalty for the summary offence is 12 months imprisonment and/or 60 penalty units.
[131] [2019] HCA 35; (2019) 373 ALR 214, 227 [52] (footnotes omitted) (referring to Beckwith v The Queen [1976] HCA 55; (1976) 135 CLR 569, 576 (Gibbs J), Stevens v Kabushiki Kaisha Sony Computer Entertainment [2005] HCA 58; (2005) 224 CLR 193, 211 [45] (Gleeson CJ, Gummow, Hayne and Heydon JJ) and Milne v The Queen [2014] HCA 4; (2014) 252 CLR 149, 164 [38] (French CJ, Hayne, Bell, Gageler and Keane JJ)).
[132] [2020] FCA 82; (2020) 150 IPR 11. For consideration of this case, see Amanda Scardamaglia, ‘When Your Trade Mark Is Not ALE It’s Cracked Up to Be’ (2021) 43(2) European Intellectual Property Review 129.
[133] Urban Alley Trial [2020] FCA 82; (2020) 150 IPR 11, 44 [159].
[134] Ibid 43 [153].
[135] Urban Alley Brewery Pty Ltd v La Sirène Pty Ltd (2020) 384 ALR 230, 245 [98] (Middleton, Yates and Lee JJ).
[136] Ibid 245 [99].
[137] Ibid 245 [100].
[138] [2020] FCCA 1718, [104]. If the composite mark, or the device element within the mark, are original artistic works, the author appears to be Chris Di Lorenzo: at [39].
[139] Ibid [106].
[140] Ibid [109]. Her Honour’s decision is entirely consistent with Colorado [2007] FCAFC 184; (2007) 164 FCR 506.
[141] (2018) 145 IPR 499, 515 [60] (Hearing Officer Cooper). If the ‘Active Skin’ device mark, or the flower devices, are original artistic works, they are anonymous.
[143] Ibid [75].
[144] Ibid [73].
[145] Ibid [75].
[146] [2019] FCA 1553; (2019) 146 IPR 185.
[147] Ibid 319 [472]–[473].
[148] Ibid 319 [472]. This issue was not considered on appeal: Bendigo & Adelaide Bank Ltd v Community First Credit Union Ltd [2021] FCAFC 31.
[149] See also Urban Alley Trial [2020] FCA 82; (2020) 150 IPR 11, where O’Bryan J held, in obiter dicta, that any use of ‘Urban’ was not a use with alterations that did not substantially affect the identity of ‘Urban Pale’, stating that ‘[w]hile the word “pale” is descriptive of beer, the removal of that word from the mark “Urban Pale” substantially alters the mark’: at 57 [229].
[150] See, eg, Minister for Immigration and Multicultural Affairs v Singh [2000] FCA 845; (2000) 98 FCR 469, 475 [17] (Black CJ, Sundberg, Katz and Hely JJ).
[151] See Acts Interpretation Act 1901 (Cth) s 15AA.
[152] Cf Designs Act 2003 (Cth) ss 16(1)–(2) (on the requirement for registrable designs to be ‘new’ (that is, not ‘identical to a design that forms part of the prior art base’) and ‘distinctive’ (that is, not ‘substantially similar in overall impression to a design that forms part of the prior art base’)).
[153] Noting also that this language has been adopted in article 17.11.19 n 17-[33] of the Australia–United States Free Trade Agreement, signed 18 February 2004, [2005] ATS 1 (entered into force 1 January 2005).
[154] See above n 119 and accompanying text. In any event, section 146 ought to be amended to make clear that the defendant commits an offence only where the sign or mark it applies is being used as a trade mark: see Burrell and Handler, Australian Trade Mark Law (n 28) 646.
[155] Cf Trade Marks Act 2002 (NZ) s 77.
[156] See further Robert Burrell and Michael Handler, ‘The Intersection between Registered and Unregistered Trade Marks’ [2007] FedLawRw 15; (2007) 35(3) Federal Law Review 375, 392–3, 396; Rob Batty, ‘The Challenges of Prior Use to New Zealand Registered Trade Mark Law’ (2014) 45(2) Victoria University of Wellington Law Review 257, 279–87 (‘The Challenges of Prior Use’). For discussion of how prior rights are managed in the EU’s ‘registration-based’ system, see Annette Kur, ‘Not Prior in Time, but Superior in Right: How Trademark Registrations Can Be Affected by Third-Party Interests in a Sign’ (2013) 44(7) International Review of Intellectual Property and Competition Law 790.
[157] Some support for this idea can be found in The North Face Apparel Corporation v Sanyang Industry Co Ltd [2014] NZCA 398, [50] (Harrison J for the Court) (‘North Face’) (holding that only the first use of an ‘identical mark’ can give rise to ownership of the applicant’s mark under New Zealand law). It should, however, be noted that the Intellectual Property Office of New Zealand has interpreted North Face as not having overturned the previous position that ownership can be established on the basis of use of a substantially identical mark: see, eg, Berryman v Fresco Nutrition Ltd [2015] NZIPOTM 21, [85]–[94] (Assistant Commissioner Alley). For critique, see Earl Gray and Rob Batty, ‘Ownership Jurisprudence under the Trade Marks Act 2002: Past and Future Directions’ (2019) 9(1) New Zealand Intellectual Property Journal 1, 8. Moving to a requirement of ‘identity’ might still allow courts to discount ‘insignificant differences’: see, under EU law, LTJ Diffusion SA v Sadas Vertbaudet SA (C-291/00) [2004] ECR I-2799, I-2844 [53]. However, following the approach to ‘identity’ taken in the UK, this would still result in much narrower outcomes than those flowing from Accor and Pham Global: see, eg, Reed Executive Plc v Reed Business Information Ltd [2004] EWCA Civ 47; [2004] RPC 40, [38]–[41] (Jacob LJ) (REED not identical with ‘Reed Business Information’); Compass Publishing BV v Compass Logistics Ltd [2004] RPC 41, [21] (Laddie J) (‘Compass Logistics’ not identical with ‘Compass’); Och-Ziff Management Europe Ltd v Och Capital LLP [2011] ETMR 1, [71] (Arnold J) (OCH not identical with OCH CAPITAL).
[158] See Batty, ‘The Challenges of Prior Use’ (n 156) 287–90. If section 58 is to be retained, thought should also be given to amending the TMA 1995 (Cth) to ensure that if an application for registration is filed in the name of a party other than the ‘owner’ (eg, a licensee of the first user, or a company in a closely held group where another company within the group was the first user), the name of the true ‘owner’ can be substituted for the name of the applicant or registered owner, instead of the application for registration being able to be opposed or the ensuing registration being able to be cancelled. Cf Patents Act 1990 (Cth) ss 22A, 138(4) and 191A. This is a more efficient way of dealing with this problem than that taken in PDP Capital Pty Ltd v Grasshopper Ventures Pty Ltd [2020] FCA 1078; (2020) 154 IPR 68, where Markovic J held that the ground of cancellation contained in sections 88(2)(a) and 58 of the TMA 1995 (Cth) was established where the applicant for registration was only an authorised user of the true owner of the mark: at 154 [334], but then exercised the court’s discretion under section 88(1) not to order the cancellation of the registration: at 158–9 [355], [357].
[159] See, eg, Michael Handler and Robert Burrell, ‘Zombie Marks? Ceased Registrations, Failed Applications and Citation Objections under Section 44 of the Trade Marks Act’ (2013) 23(4) Australian Intellectual Property Journal 206 (criticising an argument that found support in Chia Khim Lee Food Industries Pte Ltd v Red Bull GmbH (No 1) [2012] FCA 1184, namely that conflicts under section 44 should be assessed only by reference to the state of the Register at the applicant’s priority date, given the negative impact this would have on the operation of a host of other provisions of the TMA 1995 (Cth)); Harley (n 103) (raising concerns about the implications of the low standard of ‘use’ adopted in the non-use case of Dick Smith Investments Pty Ltd v Ramsey [2016] FCA 939; (2016) 120 IPR 270 for the infringement test and the ‘no intent to use’ opposition ground); Michael Handler, ‘Disentangling Functionality, Distinctiveness and Use in Australian Trade Mark Law’ (2018) 42(1) Melbourne University Law Review 55 (arguing that attempting to read ‘functionality’ considerations into the tests for distinctiveness and use ‘as a trade mark’ has resulted in the latter doctrines becoming distorted in cases of non-functional, non-traditional marks).
[160] These being key interests served by registered trade mark law, as identified in Campomar [2000] HCA 12; (2000) 202 CLR 45, 65–8 [40]–[49] (Gleeson CJ, Gaudron, McHugh, Gummow, Kirby, Hayne and Callinan JJ). See also JT International SA v Commonwealth [2012] HCA 43; (2012) 250 CLR 1, 27 [30] (French CJ), 39 [68] (Gummow J); Patricia Loughlan, ‘The Campomar Model of Competing Interests in Australian Trade Mark Law’ (2005) 27(8) European Intellectual Property Review 289.
AustLII:
Copyright Policy
|
Disclaimers
|
Privacy Policy
|
Feedback
URL: http://www.austlii.edu.au/au/journals/UNSWLawJl/2021/17.html