← All cases

[2026] HCA 5

Taylor v Killer Queen LLC

Appeal allowed with costs. Set aside orders 1, 3, 4, 5, 6, 7 and 8 of the Full Court of the Federal Court of Australia on 22 November 2024. Remit the matter to the Full Court of the Federal Court of Australia in respect of grounds 5, 6 and 7 of the appeal to that Court and any other outstanding issue including the costs of that appeal and of the proceeding before the primary judge.

Catchwords

Trade marks – Rectification of Register of Trade Marks under Trade Marks Act 1995 (Cth) – Where appellant registered owner of trade mark "Katie Perry" in class 25 (clothes) from 29 September 2008 ("priority date") – Where second respondent known as Katy Perry is famous pop star – Where respondents sought rectification of Register by cancelling "Katie Perry" trade mark – Whether use of "Katie Perry" trade mark at priority date likely to deceive or cause confusion due to reputation of pop star's trade mark – Whether use of "Katie Perry" trade mark at time of application for rectification likely to deceive or cause confusion because of circumstances at that time – Whether discretion not to rectify Register under Act enlivened. Words and phrases – "acquired a reputation", "act or fault", "actual use", "assiduous infringer", "blameworthy conduct", "cancelling the registration of a trade mark", "deceptively similar", "fashion designer", "lack of evidence", "likely to deceive or cause confusion", "material error", "merchandise", "normal and fair use", "notional use", "pop star", "priority date", "real and tangible danger", "rectification", "reputation", "trade mark", "undue weight", "wrongful conduct". Trade Marks Act 1995 (Cth), ss 60, 88(1), 88(2)(a), 88(2)(c), 89.

Legislation cited

  • Trade Marks Act 1995
  • 88
  • 89
  • TM Act
  • Trade Marks Regulations 1995
  • Pt 5
  • (ss 57
  • 62A
  • Part 8
  • Pt 8
  • Trade Marks Act 1955
  • 44
  • 60
  • Trade Marks Act
  • National Security Act 1939
  • (s 60(a))
  • (s 60(b))
  • (s 17)
  • (s 19)
  • (s 7(4)
  • (s 20(1)).
  • (s 41(1)).
  • 120
  • (s 89(1))
  • 41
  • 207
  • Trade Marks Regulations
  • Acts Interpretation Act 1901
  • Acts Interpretation Act
  • Trade Marks Amendment Act 2006
  • Pt 1
  • Trade Marks Amendment Bill 2006
  • 92
  • Trade Marks Bill 1995
  • Regulation 3.1(1)(a)
  • 6

Judgment

  1. [1]

    GORDON A-CJ AND BEECH-JONES J. The appellant, Katie Jane Taylor (born Katie Jane Perry), is an Australian fashion designer. On 29 September 2008, Ms Taylor applied under the Trade Marks Act 1995 (Cth) ("the TM Act") to be the registered owner of Australian trade mark no 1264761, the word mark "KATIE PERRY", in respect of goods in class 25 for "[c]lothing" ("the Designer's Mark").

  2. [2]

    The second respondent, Katheryn Hudson, is a music artist based in the United States who has performed under the stage name "Katy Perry" since 2002. Ms Hudson is the sole director, shareholder and controlling mind of the first and third respondents, Killer Queen LLC and Kitty Purry Inc. On 26 June 2009, Ms Hudson sought registration of the word mark "KATY PERRY" ("the Singer's Mark") (Australian trade mark no 1306481), which is registered in classes 9 and 41 but not for clothes.

  3. [3]

    This appeal raises three important issues concerning the continued registration of the Designer's Mark on the Register of Trade Marks kept under s 207 of the TM Act ("the Register"). The first issue concerns an application by Ms Hudson and the first respondent under s 88(2)(a) of the TM Act for rectification of the Register by cancelling the registration of the Designer's Mark alleging that, because of the reputation of the Singer's Mark in Australia before the priority date for the registration of the Designer's Mark, the use of the Designer's Mark would be likely to deceive or cause confusion. That issue is addressed based on the circumstances applying as at the priority date, being the date of the Designer's Mark's registration. The registration is taken to have had effect from the filing date of the application, being 29 September 2008 ("the Priority Date").

  4. [4]

    The second issue concerns an application by Ms Hudson and the first respondent under s 88(2)(c) of the TM Act for rectification of the Register by cancelling the registration of the Designer's Mark alleging that, because of the circumstances applying when they filed their application for rectification, the use of the Designer's Mark was likely to deceive or cause confusion. That issue is addressed based on the circumstances applying on 20 December 2019, a date more than ten years after registration of the Designer's Mark.

  5. [5]

    The third issue arises if one or both of those applications for rectification of the Register is successful. It concerns the court's discretion not to grant an application for rectification of the Register by cancelling the registration of a trade mark where the registered owner of the trade mark – here, Ms Taylor – satisfies the court that the ground for rectification has not arisen through any act or fault of hers. In this appeal, that issue is addressed as at the Priority Date in relation to the ground in s 88(2)(a) and as at 20 December 2019 in relation to the ground in s 88(2)(c).

  6. [6]

    Although we differ on the result, we agree with Jagot J on the questions of statutory construction about the provisions of the TM Act raised by this appeal. The difference between us concerns the application of those provisions.

  7. [7]

    For the reasons that follow, the Full Court of the Federal Court of Australia was correct to find that the grounds for rectification in s 88(2)(a) and (c) were established. As at the Priority Date, the use of the Designer's Mark would be likely to deceive or cause confusion because of the reputation of the Singer's Mark in relation to music and entertainment, having regard to the similarity of the marks and the commonly accepted practice of pop stars selling merchandise including clothing. As at the date of the respondents' application for rectification, the use of the Designer's Mark was even more likely to deceive or cause confusion for the same reasons and also because of the greater international fame of Ms Hudson by that time, with associated merchandising including clothing. The discretion not to rectify the Register under s 89 was not enlivened. The ground in s 88(2)(a) arose through Ms Taylor's "act" of applying for registration with knowledge of Ms Hudson, the reputation of the Singer's Mark and the commonly accepted practice of popular music stars lending their names to be used in connection with clothes and selling merchandise. The ground in s 88(2)(c) arose through both that "act" and Ms Taylor's encouragement of the association between herself and Ms Hudson as well as her failure to take any action to enforce her statutory monopoly. TM Act

  8. [8]

    The TM Act was considered in Self Care IP Holdings Pty Ltd v Allergan Australia Pty Ltd. As the Court explained, the TM Act provides for the registration of trade marks and sets out and protects the rights deriving from registration. The structure and policy of the TM Act may be relevantly summarised as follows: (1) the Register will protect distinctive trade marks; (2) a trade mark is registered in respect of particular goods or services; (3) the registered trade mark is treated as property itself and the registered owner of a trade mark has rights including the exclusive right to use the trade mark in relation to the goods or services in respect of which it is registered ("the monopoly"); (4) the monopoly following registration is a sufficient basis upon which to seek relief from infringement; and (5) the likelihood of deception and confusion between trade marks should be avoided.

  9. [9]

    The Register, which is publicly available, must be kept by the Registrar of Trade Marks ("the Registrar"). All particulars of registered trade marks must be entered on the Register (including the identified limits on the monopoly in the form of the mark, the class, and the particular goods), and the Register is prima facie evidence of any particular entered in it. It is inspection of the Register that enables a person to identify what marks are registered, the extent of the monopoly of each mark and what actions the person might take under the TM Act.

  10. [10]

    A person may apply for registration of a trade mark in respect of specific goods in one or more of the classes of goods specified in Sch 1 to the Trade Marks Regulations 1995 (Cth) if the person claims to be the owner of the trade mark and the person is using or intends to use the trade mark in relation to those goods. The Registrar must examine, and report on, whether the application is made in accordance with the TM Act and whether there are grounds for rejecting it. If the application is accepted by the Registrar, the decision is advertised and registration of the trade mark may then be opposed by another person under Pt 5 of the TM Act. Division 2 of Pt 5 of the TM Act (ss 57 to 62A) sets out the grounds for opposing registration. Those grounds include s 60 which provides a ground for opposition where, by reason of another trade mark having acquired a reputation in Australia before the priority date for the trade mark in respect of which the application has been filed, the use of the trade mark that is the subject of the application would be likely to deceive or cause confusion.

  11. [11]

    Part 8 of the TM Act deals with amendment, cancellation and revocation of registration of trade marks. Division 2 of Pt 8 addresses action by a court. For example, a prescribed court: may, on the application of an aggrieved person, order that the Register be rectified to correct an error in or omissions from an entry in the Register; and may, on the application of an aggrieved person or the Registrar, order that the Register be rectified on the ground that a condition or limitation entered in the Register in relation to the trade mark has been contravened.

  12. [12]

    This appeal is concerned with s 88, headed "Amendment or cancellation – other specified grounds". Section 88(1) relevantly provides that, subject to ss 88(2) and 89, "a prescribed court may, on the application of an aggrieved person or the Registrar, order that the Register be rectified by ... cancelling the registration of a trade mark". Section 88(2) then provides that an application may only be made on the following grounds: "(a) any of the grounds on which the registration of the trade mark could have been opposed under this Act; (b) an amendment of the application for the registration of the trade mark was obtained as a result of fraud, false suggestion or misrepresentation; (c) because of the circumstances applying at the time when the application for rectification is filed, the use of the trade mark is likely to deceive or cause confusion; (e) if the application is in respect of an entry in the Register – the entry was made, or has been previously amended, as a result of fraud, false suggestion or misrepresentation."

  13. [13]

    It is by s 88(2)(a) that s 60, headed "Trade mark similar to trade mark that has acquired a reputation in Australia", is directly relevant. Section 60 provides: "The registration of a trade mark in respect of particular goods or services may be opposed on the ground that: (a) another trade mark had, before the priority date for the registration of the first-mentioned trade mark in respect of those goods or services, acquired a reputation in Australia; and (b) because of the reputation of that other trade mark, the use of the first‑mentioned trade mark would be likely to deceive or cause confusion."

  14. [14]

    If an aggrieved person succeeds under s 88(2)(a) or (c), then s 89(1) relevantly provides that a court may decide not to grant an application for rectification made: "(b) on the ground that the trade mark is liable to deceive or confuse (a ground on which its registration could have been opposed, see paragraph 88(2)(a)); or (c) on the ground referred to in paragraph 88(2)(c); if the registered owner of the trade mark satisfies the court that the ground relied on by the applicant has not arisen through any act or fault of the registered owner." In making a decision under s 89(1), the court must take into account any matter that is prescribed and it may take into account "any other matter that the court considers relevant".

  15. [15]

    The following matters are prescribed matters that the court must take into account, so far as they are relevant, in making a decision under s 89(1): "(a) the extent to which the public interest will be affected if registration of the trade mark is not cancelled; (b) whether any circumstances that gave rise to the application have ceased to exist; (c) the extent to which the trade mark distinguished the relevant goods and/or services before the circumstances giving rise to the application arose; (d) whether there is any order or other remedy, other than an order for rectification, that would be adequate in the circumstances".

  16. [16]

    The scope of the discretion conferred by s 89(1) is relevant to the construction of the power to rectify the Register under s 88(1). Section 88(1) provides that a prescribed court "may", in the specified circumstances, order that the Register be rectified. Although the word "may" generally denotes a discretion, that rule of construction is subject to a contrary intention. Here, the terms of s 89(1) supply a contrary intention. If s 88(1) entailed a discretion to refuse an application for rectification, the limitations on the exercise of the discretion in s 89(1) could be avoided. Consequently, s 89 is the only source of power not to rectify the Register. Legislative history of ss 88(2)(c) and 89

  17. [17]

    Sections 88(2)(c) and 89 had no direct counterparts in the Trade Marks Act 1955 (Cth) ("the 1955 Act"). In the 1955 Act, the Register could be rectified "by the expunging or amendment of an entry wrongly made in or remaining in the Register". One of the grounds upon which a mark could be expunged in s 28(a) of the 1955 Act was that a mark "the use of which would be likely to deceive or cause confusion ... shall not be registered as a trade mark". Registration of a trade mark was also conclusive unless it was shown, relevantly, "that the trade mark offend[ed] against the provisions of [s] 28". Until this Court's decision in Campomar Sociedad, Limitada v Nike International Ltd, it was unsettled whether s 28(a) of the 1955 Act permitted a validly registered mark to be expunged due to circumstances arising after the date of the application for registration and whether those circumstances had to involve blameworthy conduct on the part of the registered owner of the mark.

  18. [18]

    In the context of the 1955 Act, that issue had been addressed in 1990 in New South Wales Dairy Corp v Murray Goulburn Co‑op Co Ltd, where four members of the Court held that s 28(a) had a continuing operation after registration and that it was necessary to show blameworthy conduct by the registered owner before a mark could be expunged due to circumstances arising after the date of registration. However, Dawson and Toohey JJ held that a "blameworthy act" was "no more than a shorthand method of referring to all those circumstances which might disentitle an applicant to relief in a court of equity". And none of Deane, Gaudron and McHugh JJ considered that s 28(a) operated by reference to events which occurred after registration. After the TM Act was enacted, this Court in Campomar ultimately overruled the decision in Murray Goulburn, concluding that s 28(a) did not have a continuing operation, so that there was no occasion to consider the doctrine of "blameworthy conduct".

  19. [19]

    However, before the TM Act was enacted, in 1992 the Working Party to Review the Trade Marks Legislation published its report titled "Recommended Changes to the Australian Trade Marks Legislation". The report, released after the decision in Murray Goulburn, referred to recent "decisions of the Federal and High Courts", stating that there was "[d]oubt ... as to the continuing relevance of pre‑acceptance tests for registrability to the situation ... post‑registration", such that it was "clear ... that action need[ed] to be taken at the legislative level to remove the perceived ambiguities from the [1955] Act". The report recommended that the new trade marks legislation should include a section that a person aggrieved may apply to the court for rectification of the Register relevantly on the ground that: "the use of the trade mark as at the date of commencement of the rectification proceedings is liable to deceive or confuse; provided that the mark shall not be removed on [this ground] if the circumstances have not arisen through fault on the part of the proprietor, the onus being on the proprietor to show this; this proviso should also apply to the ground that the registration or use of the mark by the applicant would contravene a right of the applicant for rectification, the onus being on the applicant to demonstrate its right. For the purpose of deciding [this ground], the court will have a discretion which should be exercised in accordance with but not restricted to the following: (i) how the public interest will be affected by the registration remaining on the Register; (ii) has any external cause ceased to be a cause; (iii) how distinctive was the trade mark before these circumstances arose; or (iv) is there an adequate remedy or order short of the requested rectification." Sections 88(2)(c) and 89, as originally enacted, largely adopted this recommendation.

  20. [20]

    When s 88(2)(c) was originally enacted in 1995 it provided that an application for rectification may be made on the ground that: "because of the circumstances applying at the time when the application for rectification is filed, the use of the trade mark is likely to deceive or cause confusion for a reason other than one for which: (i) the application for the registration of the trade mark could have been rejected under section 43 or 44; or (ii) the registration of the trade mark could have been opposed under section 60." (emphasis added) In 2006, s 88(2)(c) was amended. The italicised words were deleted. The Explanatory Memorandum stated that it was not clear that "trade marks registered under a previous Act can be removed from the Register due to grounds that have arisen since they were registered" and the amendment to s 88(2)(c) was "to make it clear that any trade mark may be removed from the register, irrespective of the Act under which it was registered, if it is likely to deceive or cause confusion to consumers at the time an application to remove it from the Register is made".

  21. [21]

    As was just explained, it was after the enactment of the TM Act that this Court in Campomar held that s 28(a) of the 1955 Act did not have a continuing operation and thus that there was no occasion to consider "blameworthy conduct". Commentators recognised that that decision "cut through most of the difficulties" to which ss 88(2)(c) and 89 were directed. That legislative history, as well as this Court's decision in Campomar, are important when construing ss 88(2)(c) and 89. In particular, that history assists in understanding the resolution of the apparent inconsistencies or tensions within the TM Act of expressly providing, for example, for the registration of a substantially identical or deceptively similar trade mark where there has been honest concurrent use of two trade marks, the assignment of a registered trade mark, in whole or in part, and with or without goodwill, and the power to authorise a person to use a trade mark in relation to goods or services under the control of the owner of the trade mark, while, at the same time, prohibiting the registration of and continued registration of trade marks where the use of that mark would be likely to deceive or cause confusion.

  22. [22]

    To that extent, the TM Act "establish[es] and sanction[s] a system the operation of which involve[s] a measure of likely deception or confusion". The TM Act strikes a balance between, on the one hand, the interest of consumers in "recognising a trade mark as a badge of origin of goods or services and in avoiding deception or confusion as to that origin" and, on the other hand, "the interest of traders, both in protecting their goodwill through the creation of a statutory species of property protected by the action against infringement, and in turning this property to valuable account by licensing or assignment". Sections 88(2)(c) and 89 – and, indeed, ss 88(2)(a) and 89 – are examples of the balance that is struck. Whereas s 88(2)(c) operates to protect consumers from deception and confusion, the favourable exercise of the discretion under s 89 admits the potential for some degree of deception or confusion in recognition of other interests recognised by the TM Act, including those of the registered proprietor of the trade mark. Facts and background

  23. [23]

    The summary of the facts and background below is drawn from the reasons of the primary judge and the Full Court. Ms Taylor

  24. [24]

    Early in 2007, Ms Taylor started her own fashion label, initially using a different name before settling on "KATIE PERRY". On 5 April 2007, Ms Taylor applied to register the business name, "Katie Perry", which was subsequently registered on 12 April 2007. On 2 May 2007, Ms Taylor registered a domain name for her business at www.katieperry.com.au. Ms Taylor has designed and sold clothes under the brand name "KATIE PERRY" since about 2007.

  25. [25]

    On 13 September 2007, Ms Taylor applied to register a trade mark for a "fancy logo" where the prominent feature of the logo was the words "KATIE PERRY" in cursive font ("the First KP Trade Mark"). She applied for registration of the First KP Trade Mark in class 42 for "clothing and fashion designing". At that time, Ms Taylor did not appreciate that goods and services belong to different classes. Ms Hudson

  26. [26]

    It is unnecessary to chart Ms Hudson's career as a solo artist in detail. It is sufficient to record that in November 2007, in advance of the planned release of an album, Ms Hudson's single "Ur So Gay" and music video were digitally released on the youth-orientated social media website, MySpace. The song was immediately popular and "went viral". An extended play including the single was subsequently released and, from January 2008, was available for purchase in a compact disc and in a seven-inch vinyl version.

  27. [27]

    On 17 June 2008, Ms Hudson released her second studio album, "One of the Boys", which included the singles "Ur So Gay", "I Kissed a Girl", "Hot n Cold", "Thinking of You" and "Waking up in Vegas". The album and the singles ranked on international music charts. During the northern hemisphere summer of 2008, Ms Hudson's commercial and career momentum increased. Her singles were popular on radio and received extensive airplay. In July 2008, Mr Steven Jensen, a partner of Direct Management Group Inc, a talent agency which managed Ms Hudson's career, first started considering branded merchandise to sell in countries where she was popular including Australia. The first merchandise design concept was ready on 6 September 2008.

  28. [28]

    On 9 September 2008, Ms Hudson commenced her promotional tour in the United Kingdom. The other countries in the tour were France, Germany, Italy, Spain, Japan, Mexico and Australia. In September 2008, before Ms Hudson's arrival in Australia, there was publicity about her intended visit and the fact that she would be performing concerts in Sydney and Melbourne, with tickets going on sale on 12 September 2008. On 11 September 2008, the first of Ms Hudson's t‑shirts was approved internally for retail sale. On 16 September 2008, another of Ms Hudson's t-shirts was approved, with Mr Jensen stating he wanted to distribute the t-shirt for retail sale in certain countries, including Australia, quickly. Ms Hudson's online store, katyperry.shop.bravadousa.com, went live in early October 2008. Members of the public from any country around the world, including Australia, could access Ms Hudson's website and purchase "KATY PERRY" branded merchandise through it. The Australian leg of the promotional tour ran from 10 to 14 October 2008 and received extensive press coverage.

  29. [29]

    During 2008, Ms Hudson's popularity spread internationally as a result of international coverage of her performances on the "Warped Tour" and the popularity of the singles she had released, including "I Kissed a Girl", on radio in countries such as the United Kingdom, Japan, Germany, France, Australia, New Zealand, Italy, Spain, and Russia, as well as in South America. In Australia, the "End of year chart – top 100 singles in 2008" published by the Australian Recording Industry Association showed that, for 2008, Ms Hudson's single "I Kissed a Girl" was at number six and her single "Hot n Cold" was at number 16. Subsequent events

  30. [30]

    Ms Taylor first heard about Ms Hudson in July 2008 when she was listening to the radio and heard Ms Hudson's stage name and her song "I Kissed a Girl". Ms Taylor bought the song on iTunes because she wanted to support an artist who had the same name as her.

  31. [31]

    In about August 2008, Ms Taylor realised that she had not paid the registration fee for the First KP Trade Mark. After contacting IP Australia, she was told that she could still register that mark but that she would have to pay late fees as well as the registration fee. At about the same time or shortly thereafter, Ms Taylor came to understand that, to obtain better protection, she should register a word mark in relation to clothes, which were the type of goods she intended to sell. Accordingly, she decided to let the First KP Trade Mark application lapse (it lapsed on 17 July 2008).

  32. [32]

    On 29 September 2008, Ms Taylor applied to be the registered owner of the Designer's Mark. On 18 December 2008, Ms Taylor received an early notice of acceptance from IP Australia for the Designer's Mark. IP Australia noted that the application had passed the examination stage and that the next step was advertising in order to give other people the opportunity to oppose registration of the trade mark.

  33. [33]

    At some point in the period between 5 and 8 June 2009, shortly after the launch of her showroom, Ms Taylor received a letter dated 21 May 2009 marked "without prejudice" from a lawyer acting for Ms Hudson demanding, among other things, that Ms Taylor immediately withdraw the trade mark application for the Designer's Mark. Ms Taylor was aware that a notice of opposition to the registration of the Designer's Mark had been sent to her post office box and that Ms Hudson had filed an application for an extension of time to file the notice of opposition when she filed the notice of opposition.

  34. [34]

    Then, on 15 June 2009, another firm of lawyers acting for Ms Hudson sent a further letter to Ms Taylor asserting that she had wrongfully applied for the registration of the Designer's Mark and that the application was "opposed by Ms [Hudson]". The letter described what it referred to as Ms Taylor's unlawful conduct and set out a demand that, by 4 pm on Friday 19 June 2009, she immediately withdraw the trade mark application for the Designer's Mark. A draft originating application to be filed in the Supreme Court of Queensland was enclosed. Ms Taylor was subsequently contacted by Ms Hudson's lawyer by telephone on 16 June 2009, was sent correspondence his law firm on 17 June 2009 and received an email requesting that she call him on 19 June 2009.

  35. [35]

    In the meantime, on 26 June 2009, Ms Hudson sought registration of the Singer's Mark in classes 9, 25 and 41. In relation to class 25, Ms Hudson's application was in respect of "apparel", including shirts, sweat pants, and scarves.

  36. [36]

    On 29 June 2009, Ms Taylor posted a video message on her YouTube channel for Ms Hudson, describing herself to Ms Hudson and asking Ms Hudson to leave her to carry on her dream. Ms Taylor knew that Ms Hudson was planning to tour Australia in August 2009 and that it was likely that clothing featuring the Singer's Mark would be offered for sale at her concerts. On 1 July 2009, Ms Hudson's lawyers sent Ms Taylor a letter stating that both parties had legitimate rights and that those rights might co-exist. Drafts of a deed of agreement were exchanged but the terms ultimately proposed by Ms Hudson were unacceptable to Ms Taylor.

  37. [37]

    On 21 July 2009, the Designer's Mark was entered on the Register and on 6 August 2009 the registration of the Designer's Mark was advertised.

  38. [38]

    In September 2009, Ms Hudson amended her application, seeking to register the Singer's Mark only in classes 9 and 41. The Singer's Mark did not include any registration for clothes.

  39. [39]

    On 24 October 2019, Ms Taylor commenced proceedings in the Federal Court alleging, relevantly, that the respondents had imported for sale, distributed, advertised, promoted, marketed, offered for sale, supplied, sold and/or manufactured in Australia, or to people in Australia, clothes in class 25 bearing the Singer's Mark and that by that conduct the respondents had infringed Ms Taylor's Designer's Mark pursuant to s 120 of the TM Act or were liable as joint tortfeasors in the direct infringement of some of the respondents or identified third parties.

  40. [40]

    On 20 December 2019, Ms Hudson and the first respondent filed a cross‑claim seeking rectification of the Register by cancelling the Designer's Mark pursuant to s 88(1)(a) of the TM Act on the grounds set out in s 88(2)(a) and (c) of the TM Act. The date – 20 December 2019 – is the relevant date for considering s 88(2)(c) of the TM Act. Decisions below

  41. [41]

    The primary judge found that Kitty Purry Inc infringed the Designer's Mark as a joint tortfeasor and ordered it to pay additional damages as its infringement was flagrant. The primary judge also found that Ms Hudson would have infringed the Designer's Mark by certain social media posts and tweets except that Ms Hudson was entitled to rely on the "own name" defence in s 122(1)(a)(i) of the TM Act.

  42. [42]

    The respondents were unsuccessful in seeking rectification of the Register by cancelling the Designer's Mark under s 88(2)(a) (including by reference to s 60) and s 88(2)(c) of the TM Act. As to ss 60 and 88(2)(a), the primary judge found that, as at the Priority Date, although Ms Hudson had established a reputation in the Singer's Mark in Australia in relation to entertainment and music, that reputation did not extend to clothes and that the respondents had not established the requisite likelihood of deception or confusion. As to s 88(2)(c), which was considered as at 20 December 2019, the primary judge held that the ground was not established because of the absence of any confusion arising from Ms Taylor's use of the Designer's Mark and the strength of the reputation of Ms Hudson. The primary judge did not need to consider whether to exercise the discretion in s 89 and did not consider how she would have exercised it.

  43. [43]

    The respondents appealed the primary judge's findings on infringement, and on their cross-claim seeking rectification of the Register by cancelling the Designer's Mark, to the Full Court. Ms Taylor cross-appealed in relation to the primary judge's findings that Killer Queen LLC and the fourth respondent, Purrfect Ventures LLC, were not liable as joint tortfeasors and on the scope of the registered goods and the respondents' defences to trade mark infringement.

  44. [44]

    The Full Court overturned the primary judge's findings in relation to ss 88(2)(a) and 88(2)(c), concluding that both grounds for rectification were made out. In relation to s 88(2)(a), the Full Court held that the primary judge had "unduly confined" the reputation of the Singer's Mark. First, the Court considered that the reputation in Ms Hudson's stage name was of necessity a reputation in her name as a mark. Second, the Court held that, by confining the reputation in the Singer's Mark to music and entertainment, the primary judge overlooked relevant evidence of the common practice of pop stars to sell merchandise including clothing at concerts and to launch their own clothing labels. The Court then found that the primary judge had erred by overlooking the close similarity of the marks, placing undue weight on the absence of evidence of actual confusion and taking into account post-Priority Date use of the Singer's Mark. The Full Court held that, because of the reputation of the Singer's Mark, and taking into account the common practice identified, the use of the Designer's Mark would be likely to deceive or cause confusion.

  45. [45]

    In relation to s 88(2)(c), the Full Court held that, as at the date of the respondents' cross-claim, consumers were likely to be deceived or confused for the same reasons as it reached that conclusion in relation to s 88(2)(a), whether the comparison was the actual or notional use of the Designer's Mark. The Full Court observed that, by the time of the cross-claim, the reputation of the Singer's Mark was stronger than at the Priority Date. Further, where there is little difference between the marks, the Court considered that the reputation in the Singer's Mark was more likely to enhance, rather than reduce, the likelihood of confusion.

  46. [46]

    The Full Court considered the discretion in s 89 of the TM Act, finding that it was not enlivened in relation to either ground of rectification because Ms Taylor's act of applying to register the Designer's Mark, with knowledge of Ms Hudson, her reputation, her mark and the practice of pop stars lending their name to be used in connection with clothes and selling merchandise at concerts, "contribute[d]" to the grounds being established. The Full Court also stated that, even if the discretion had been enlivened, the Full Court would not have exercised it in Ms Taylor's favour.

  47. [47]

    Ms Taylor now appeals each of those findings. As to the ground of rectification under s 88(2)(a), Ms Taylor contended that the Full Court wrongly conflated Ms Hudson's reputation with the reputation of the Singer's Mark and that the Singer's Mark had not acquired a reputation in respect of clothing before the Priority Date. Ms Taylor also submitted that the Full Court was wrong to adopt an analysis of whether the marks were deceptively similar rather than focusing on whether the reputation of the Singer's Mark would be likely to cause deception or confusion. Further, she contended that the primary judge was correct to place significant weight on the absence of evidence of actual confusion arising from her use of the Designer's Mark.

  48. [48]

    As to the ground of rectification under s 88(2)(c), Ms Taylor submitted that the ground is to be assessed having regard to the registered owner's actual rather than notional use of the challenged trade mark. On that construction, the absence of evidence of actual confusion was of central importance. Even if the inquiry is as to notional use, Ms Taylor said that the primary judge was correct to find that the ground was not made out due to the lack of evidence of actual confusion and because the strength of the reputation of the Singer's Mark reduced the likelihood of confusion.

  49. [49]

    In the alternative, Ms Taylor submitted that the Full Court was wrong to conclude that the discretion in s 89 was not enlivened and that the mere act of registering a trade mark is an "act" that will prevent the discretion from being enlivened. She contended that "something more" is required but was not present in this case. Ms Taylor also contended that the ground relied on for rectification will not have "arisen through" an act or fault of the registered owner if the act or fault merely contributed to the ground arising. She said that this Court should exercise the discretion in her favour having regard to all the circumstances of the case.

  50. [50]

    For the reasons that follow, each of Ms Taylor's grounds of appeal to this Court must be dismissed. Sections 60 and 88(2)(a) of the TM Act

  51. [51]

    As we have seen, s 88(2)(a) of the TM Act provides that an application that the Register be rectified by cancelling a trade mark may be made on any of the grounds on which the registration of the trade mark could have been opposed. One of those grounds is s 60. Principles

  52. [52]

    The purpose of s 60 is to provide protection for marks that are well known, whether they are registered or unregistered, by preventing the registration of a trade mark if there is a likelihood of deception or confusion because of the reputation of the well-known mark in Australia as at the priority date. In determining whether the s 60 ground is made out, it is necessary to answer two questions in turn: first, whether another trade mark ("the prior mark") had, before the priority date for the registration of a trade mark in respect of particular goods or services ("the opposed mark"), acquired a reputation in Australia; and second, whether, because of the reputation of that prior mark, the use of the opposed mark would be likely to deceive or cause confusion. Reputation

  53. [53]

    "Reputation" refers to the "recognition of [the mark] by the public generally" and includes the credit, image and values projected by the trade mark. In the context of s 60, "reputation" has been considered to refer to recognition of a trade mark by the public generally. Section 60 does not expressly refer to the prior mark having acquired a reputation in respect of particular goods or services. However, given that a trade mark is a sign used, or intended to be used, to distinguish goods or services, it follows that the prior mark must have acquired a reputation in Australia in respect of particular goods or services. On the other hand, it is not necessary for the prior mark to have acquired a reputation in respect of the same goods or services in respect of which the opposed mark is proposed to be registered, although whether it has done so will be relevant to the likelihood of deception or confusion which is the subject of the second inquiry. It has also been held to be "commonplace to infer reputation from a high volume of sales, together with substantial advertising expenditures and other promotions, without any direct evidence of consumer appreciation of the mark, as opposed to the product". Deceive or cause confusion

  54. [54]

    As this Court observed in Self Care, the likelihood of deception or confusion from the use of the opposed mark is required to arise because of the reputation in Australia of the prior mark, not reputation alone. Whether the opposed mark is likely to deceive or cause confusion depends on whether or not a reasonable number of people may be caused to wonder if a trade connection exists between the marks. "Deceive" implies the creation of an incorrect belief or mental impression; "cause confusion" may merely involve "perplexing or mixing up the minds" of potential customers. There must be a real, tangible danger of deception or confusion – a mere possibility of deception or confusion is insufficient. On the other hand, the question is whether the use of the opposed mark "would be likely" to deceive or cause confusion; evidence of actual confusion is not required although it may be persuasive if present. With that said, the absence of evidence of past deception or confusion may be explicable by the small extent to which the opposed mark has been used or special circumstances affecting the past use of the mark that may not continue to prevent deception or confusion in the future.

  55. [55]

    The inquiry required by s 60(b) involves comparing the reputation of the prior mark based on actual use of that mark against a notional normal and fair use of the opposed mark. "Notional normal and fair use" refers to how a registered mark could be used in a hypothetical, typical market, rather than actual use. The concept of notional normal and fair use requires consideration of "all legitimate uses to which [the] mark might be put by its owner". As Mason J explained in Berlei Hestia Industries Ltd v Bali Co Inc, the question whether there is a likelihood of deception or confusion is to be answered "not by reference to the manner in which [the registered owner] has used its mark in the past, but by reference to the use to which it can properly put the mark".

  56. [56]

    The inquiry under s 60 about "notional normal and fair use" of the opposed mark is necessarily hypothetical. It asks whether the opposed mark would be likely to deceive or cause confusion among the public because of the reputation of the prior mark. The inquiry should be conducted by reference to the notional buyer, who is understood by reference to the nature and kind of customer who would be likely to buy the goods or services proposed to be covered by the registration. The notional buyer is aware of the prior mark to an extent that is consistent with the content and extent of its reputation. Where the goods in question are sold to the general public, it is appropriate for the court to consider not only any evidence of other members of the public that has been adduced but also to use its own common sense in answering the question of whether there would be a likelihood of deception or confusion. The hypothetical inquiry is undertaken at the priority date.

  57. [57]

    As has been explained, the prior mark need not have had a reputation in respect of the same goods or services in respect of which the opposed mark is proposed to be registered. That is important. It reflects that there might be a risk of deception or confusion because of the reputation of a prior mark in relation to goods or services of a different kind as a result of some connection which the notional buyer is likely to make between the relevant trade marks.

  58. [58]

    Against that background, it is necessary to address each of the questions in turn. Before the Priority Date, had the Singer's Mark acquired a reputation in Australia as a trade mark?

  59. [59]

    The first question under s 60(a) is whether the Singer's Mark had, before the Priority Date, acquired a reputation in Australia and, if so, what the nature and extent of that reputation in Australia was at that time.

  60. [60]

    The primary judge found that the Singer's Mark had acquired a reputation in Australia before the Priority Date in relation to music and entertainment. That finding was based on a number of facts and matters. First, there was no dispute that by the Priority Date Ms Hudson had released several singles and her "One of the Boys" album. The primary judge also accepted that, by that time, at least the singles "I Kissed a Girl" and "Hot n Cold" and the album "One of the Boys" had achieved significant exposure in Australian media, the singles had been played on Australian radio and each had achieved a level of popularity given its respective ranking on the Australian charts. In particular, the "One of the Boys" album received publicity in many Australian newspapers. The primary judge inferred that, in order for the "One of the Boys" album to have been reported to be "gold" by 15 October 2008, there must have been significant sales and recognition of the album prior to that date, including before the Priority Date. By 4 September 2008, "I Kissed a Girl" had been at number one in the Australian charts for six weeks and, in the week of 15 September 2008, both the singles "I Kissed a Girl" and "Hot n Cold" were in the Australian top ten. Those singles were also reported as numbers four and five respectively on iTunes in 2008 and were both in the Australian Recording Industry Association's "End of Year Charts – Top 100 Singles in 2008" in Australia.

  61. [61]

    The primary judge then inferred, based on those singles' and that album's popularity and wide reporting in the media, that the singles "I Kissed a Girl" and "Hot n Cold", and the album "One of the Boys", had been widely marketed and offered for sale. While the primary judge correctly found that the media coverage did not constitute trade mark use, it included numerous references to the "Katy Perry" name under which Ms Hudson was selling her singles and album and to that extent referred to the Singer's Mark and evidenced the extent of Ms Hudson's reputation before the Priority Date in Australia. The primary judge found that the reference to "Katy Perry" in relation to the singles, album and associated marketing material was not merely to describe the performer of the singles and the songs in the album but rather was use of the name as a trade mark.

  62. [62]

    The Full Court concluded that recognition of the name "Katy Perry" in Australia was of necessity evidence of a reputation in that name as a mark. So much must be accepted in light of the unchallenged findings of the primary judge. As the Full Court observed, the primary judge found that, by the Priority Date, Ms Hudson, under her stage name "Katy Perry", was an internationally famous pop star and, by reason of her fame and success as an international entertainer, there would be recognition of the Singer's Mark by the public generally.

  63. [63]

    However, the Full Court was wrong to characterise the reputation that the Singer's Mark had acquired before the Priority Date by reference to evidence of the "common practice" of pop stars to sell merchandise including clothing at concerts and to launch their own clothing labels. The relevant evidence and findings were that Ms Taylor accepted that, as at September 2008, she likely knew that music artists and pop stars sold tour merchandise like t-shirts bearing their names and likenesses at music concerts. Ms Taylor also accepted that, in September 2008, she knew that it was commonplace for musical artists to put out items of clothing bearing their name and likenesses, referring to the example of Gwen Stefani. As counsel for the respondents submitted, that was evidence from a member of the public of a practice which was visible to members of the public who shopped for clothes. In addition, Mr Jensen's unchallenged evidence was that, in addition to merchandise associated with concerts and performances, artists license their name, or another name or brand they own, as a brand for merchandise, including clothing, giving examples like Rihanna, Jennifer Lopez, Beyoncé and Gwen Stefani.

  64. [64]

    The Full Court also referred to an article published in the 16 June 2009 issue of The Australian which stated that "[f]ashion is a common area into which celebrities launch brands, with singers including Jennifer Lopez and Gwen Stefani launching their own fashion labels". The article quoted a trade mark lawyer who said, "[a] lot of US popstars are moving into all sorts of merchandising" and "[c]lothing is an obvious one for popstars". While the article was published after the Priority Date, as the respondents submitted it was contextual evidence in support of the common practice referred to as at and around the Priority Date.

  65. [65]

    However, that common practice did not mean that the Singer's Mark had acquired a reputation in Australia in clothing before the Priority Date. "Reputation" in s 60 is concerned with the recognition of a mark – here, the Singer's Mark – by the public generally. As at the Priority Date, the recognition of a mark – here, the Singer's Mark – by the public generally was in relation to music and entertainment. As at the Priority Date, there were no findings that the Singer's Mark had been used in relation to sales of merchandise. That is unsurprising given the findings of the primary judge that Ms Hudson's website store, katyperry.shop.bravadousa.com, did not go live until after the Priority Date and that members of the public, including from Australia, could not access Ms Hudson's website and purchase "KATY PERRY" branded merchandise through it before the Priority Date. That is not to deny the fact that whether the Singer's Mark had a reputation in Australia in respect of clothing as at the Priority Date was relevant to, but not determinative of, the likelihood of deception or confusion, the subject of the next question. As at the Priority Date, would the Designer's Mark be likely to deceive or cause confusion because of the reputation in Australia of the Singer's Mark?

  66. [66]

    The second question under s 60(b) is whether as at the Priority Date the notional normal and fair use of the Designer's Mark across all legitimate uses to which the Designer's Mark might be put by Ms Taylor – in relation to clothing – would create a real, tangible danger of confusion because of the reputation in Australia of the Singer's Mark based on its actual use? Put in different terms, the hypothetical question is whether a reasonable number of people may be caused to wonder if a trade connection exists between the Designer's Mark based on its notional normal and fair use on clothing and the Singer's Mark.

  67. [67]

    The notional normal and fair use of the Designer's Mark across all legitimate uses to which the Designer's Mark might be put by Ms Taylor as at the Priority Date would extend to include the use of the Designer's Mark on labels of clothes as well as on the front of a t‑shirt. The question whether the Designer's Mark is likely to deceive or cause confusion depends on whether as at the Priority Date it could be concluded that a reasonable number of people may be caused to wonder if a trade connection exists between the Designer's Mark based on its use on that clothing and the Singer's Mark. Or, put in different terms, as at the Priority Date would a reasonable number of people wonder whether a t-shirt bearing the Designer's Mark was connected with the Singer's Mark given the reputation that the Singer's Mark had in Australia at that time? The answer is "yes". Similarity of the marks

  68. [68]

    The similarity of the marks may be relevant to but is not determinative of the question just posed. Although the cause of the likelihood of deception or confusion must be the reputation of the prior mark, that does not preclude a court from considering the extent to which the two trade marks are similar. Such similarities might, depending on the circumstances, increase or decrease the likelihood of deception or confusion by reason of the prior mark's reputation. For example, the strength of the reputation of a prior mark may mean that consumers notice small differences between trade marks so that, "giving the matter a moment's reflection", they "would readily conclude" that the trade marks were not related.

  69. [69]

    It was common ground that the Designer's Mark and the Singer's Mark were "deceptively similar", within the meaning of s 10 of the TM Act. Both marks are word marks consisting of two words, they are aurally identical and, as the Full Court found, the difference in spelling is not "striking". Adapting the words of the Full Court, while some die‑hard fans of Ms Hudson might have recognised the incorrect spelling, the ordinary consumer with an awareness of the Singer's Mark that corresponded to its reputation as at the Priority Date who was confronted with a garment with the Designer's Mark on it would have been likely to be confused as to the source of the item and to have wondered whether it was associated with Ms Hudson. Common practice of pop stars selling merchandise including clothing

  70. [70]

    Next, the Full Court relied on evidence and findings of the common practice of pop stars to sell merchandise including clothing at concerts and to launch their own clothing labels.

  71. [71]

    To the extent that the primary judge said that "[i]t may be that popular music stars such as Ms Hudson lend their names (and trade marks) to be used in connection with clothes and sell merchandise at their concerts but that does not make a reputation in one thing, ie entertainment and music, related to another, in this case clothes", that analysis was misplaced. The question is not whether the Singer's reputation in entertainment and music is related to or extends to clothing. The question is that posed by s 60 of the TM Act – whether an ordinary consumer with an awareness of the Singer's Mark consistent with its reputation as at the Priority Date confronted with a garment with the Designer's Mark on it would be likely to be confused as to the source of the item and wonder whether it was associated with Ms Hudson. That there was a real risk or danger of that kind of deception or confusion principally arises from the fact that there was a commonly accepted practice of pop stars selling merchandise including clothing at concerts.

  72. [72]

    It may be accepted that the evidence that there was a common practice as at the Priority Date of celebrities launching their own fashion brands was less strong. Even putting the evidence of celebrities launching their own brands to one side, the Full Court was correct to conclude that the use of the Designer's Mark would be likely to deceive or cause confusion because of the reputation of the Singer's Mark as at the Priority Date. The reputation of the Singer's Mark in music and entertainment, in combination with evidence of the practice of pop stars selling merchandise including clothing and the close similarity of the marks, meant that there was a real, tangible danger that, as at the Priority Date, a notional buyer, when faced with the use of the Designer's Mark on clothing like "hoodies, t-shirts, tank tops and booty shorts", would be caused to wonder if a trade connection existed between the Designer's Mark and the Singer's Mark. It may be that a savvy consumer might have reflected on, for example, the absence of an image of Ms Hudson or details of a recent tour date on a t‑shirt, or the fact that the t-shirt was not being sold at a concert, and then concluded that there was no trade connection between the marks. That does not preclude the conclusion that there was a real, tangible danger that a notional buyer would be deceived or confused because of the reputation of the Singer's Mark. Evidence of actual confusion and actual use of Designer's Mark

  73. [73]

    The Full Court was correct to conclude that the primary judge had placed "undue weight" on the absence of evidence of actual confusion at and since the Priority Date by describing it as a "powerful factor". It was wrong to give the absence of evidence of actual confusion such weight in circumstances where evidence of actual confusion is not required and the absence of such evidence was explicable by the small extent to which the Designer's Mark had been used.

  74. [74]

    That is important because the consideration for s 60 is the notional normal and fair use of the Designer's Mark across all relevant goods or services – clothing – not just the actual use of the mark by Ms Taylor. It may be accepted that there were differences between the styles of Ms Taylor's clothing (described as "luxury loungewear") and the merchandise of pop stars, such as "hoodies, t-shirts, tank tops and booty shorts", as at the Priority Date. However, the inquiry under s 60 is concerned with the notional normal and fair use of the Designer's Mark, not its actual use. The prominent display of the Designer's Mark on the front of a hoodie, t‑shirt or tank top was a normal and fair use of the trade mark as at the Priority Date. It was likely that the use of the Designer's Mark in that manner would have given rise to deception or confusion by reason of the reputation that the Singer's Mark had acquired in Australia before the Priority Date. In circumstances where the actual use of the Designer's Mark was limited but the inquiry as to deception or confusion considered the notional normal and fair use of that mark, the absence of actual evidence of confusion in the past was of little significance. Post-Priority Date use of Singer's Mark

  75. [75]

    The Full Court also held that the primary judge wrongly considered post‑Priority Date use of the Singer's Mark on clothes that also bore images of Ms Hudson. One of the reasons the primary judge considered that there was no real, tangible danger of confusion was because "unlike the clothes to which the [Designer's Mark] is affixed, including items such as t-shirts, the clothes offered for sale by the respondents rarely bear only the [Singer's Mark] on them". The quoted passage suggests that the primary judge did not appropriately confine the question to whether the notional normal and fair use of the Designer's Mark at the Priority Date was likely to deceive or cause confusion because of the reputation of the Singer's Mark acquired before the Priority Date. Conclusion

  76. [76]

    For those reasons, the registration of the Designer's Mark in respect of clothing could be opposed on the ground that the Singer's Mark had, before the Priority Date for the Designer's Mark in respect of clothing, acquired a reputation in Australia (in music and entertainment) and, because of that reputation, the use of the Designer's Mark would be likely to deceive or cause confusion. The Full Court was correct to find that the respondents had established the ground in s 88(2)(a) of the TM Act.

  77. [77]

    This conclusion does not mean that all prominent celebrities and entertainers with a reputation that amounts to a mark necessarily acquire some de facto protection against allegations of infringement across the entire range of possible uses of that mark. In this case, even though the Singer's Mark had not acquired a reputation in relation to clothes as at the Priority Date, the Designer's Mark engaged s 60 because it was commonly accepted at the relevant time that persons whose trade marks had the reputation enjoyed by the Singer's Mark would sell clothing of a particular kind. That conclusion does not necessarily apply to other categories of goods. Section 88(2)(c) of the TM Act

  78. [78]

    Under s 88(2)(c), an application that the Register be rectified by cancelling a trade mark may be made on the ground that "because of the circumstances applying at the time when the application for rectification is filed, the use of the trade mark is likely to deceive or cause confusion". The question under s 88(2)(c) is broader than that posed by s 88(2)(a) read with s 60 of the TM Act because there is no requirement in s 88(2)(c) that the source of the confusion be the reputation of another mark.

  79. [79]

    Given how Ms Taylor's case was argued in this Court and in the Courts below, it is unnecessary to determine whether s 88(2)(c) is to be construed as excluding any circumstance which has arisen through wrongful conduct of an applicant for rectification. What is the relevant inquiry?

  80. [80]

    The primary judge approached the likelihood of deception or confusion for the purposes of s 88(2)(c) by reference to any normal and fair use of the Designer's Mark. At trial, no party contended for a different approach. In the Full Court, Ms Taylor contended that this ground was to be assessed having regard to the registered owner's actual use of the trade mark. While the Full Court did not expressly deal with the issue as a question of statutory construction, the Full Court regarded the s 88(2)(c) ground as established even if Ms Taylor's construction were correct.

  81. [81]

    Thus, as a threshold issue, it is necessary to address Ms Taylor's submission that this ground of rectification was to be assessed having regard to the registered owner's actual use of the challenged mark as at the relevant date and not the notional normal and fair use of the mark. Ms Taylor's submissions are contrary to the legislative history of s 88(2)(c), as well as the text, context and purpose of s 88(2)(c), all of which compel the conclusion that the ground is to be assessed by reference to the notional normal and fair use of the registered mark, not its actual use, although its actual (normal and fair) use may be relevant as one of "the circumstances applying at the time when the application for rectification is filed". Text, context and purpose

  82. [82]

    The text of s 88(2)(c) necessarily dictates the nature of the inquiry. The text provides a ground for rectification of the Register where "because of the circumstances applying at the time when the application for rectification is filed, the use of the trade mark is likely to deceive or cause confusion". It concerns the registration of a trade mark on the Register, which registration, consistent with the structure of the TM Act, must be in relation to specified goods or services. As has been explained, the registration of a trade mark on the Register under the TM Act grants exclusive rights to the owner of the mark across the breadth of the specification (and, in the context of infringement, more broadly to goods or services of the same description or that are closely related). Those monopoly rights are not constrained by any actual use the registered owner makes of the mark.

  83. [83]

    Separately and relevantly, s 88(2)(c) addresses the fact that a registered mark may be rectified by reason of circumstances which have arisen after registration. The reference in s 88(2)(c) to "the circumstances applying at the time when the application for rectification is filed" (emphasis added) is not limited to actual use of the mark by its registered owner. To limit the inquiry to actual use of the mark by its registered owner at that date would be inconsistent with the statutory monopoly of the registered owner. The rights attached to that monopoly do not change from the date of registration. Upon registration, the registered owner is entitled to use that mark exclusively in relation to specified goods or services. The fact that, as at the date that the application for rectification was filed, the registered owner has not exploited the monopoly to its fullest extent cannot and does not detract from the fact that the registered owner holds those monopolistic rights or the nature of the inquiry under s 88(2)(c). Put in different terms, one of the circumstances applying at the time when the application for rectification is filed is that the registered owner is entitled to use the registered mark exclusively in relation to specified goods or services.

  84. [84]

    Section 88(2)(c) uses similar language – "the use of the trade mark is likely to deceive or cause confusion" – as that found in s 88(2)(a) read with s 60. Both provisions refer to the "likely" effect of the "use" of the trade mark at the relevant point in time. The question is necessarily hypothetical. Neither provision expressly confines the inquiry to how the trade mark is or was in fact used at that time. In that sense, the provisions may be distinguished from other provisions of the TM Act that direct attention to "any use that has been made of the trade mark", how a trade mark "has been used" or how a party "has not used" the trade mark.

  85. [85]

    In sum, the relevant "use" of the trade mark is necessarily the notional normal and fair use which has been explained in the context of s 60. By requiring consideration of the "circumstances applying at the time when the application for rectification is filed", s 88(2)(c) necessarily permits consideration of how any actual (normal and fair) use of the trade mark at that time might be likely to deceive or cause confusion. However, that expression does not confine the inquiry to the actual use of the trade mark at the time of the application for rectification. It reflects that the time at which the application is assessed is as at the date of the application for rectification and not some earlier time. The proper inquiry is whether, at the time of the rectification application, any notional normal and fair use of the trade mark by its registered owner across the full range of goods or services to which the registration extends is likely to deceive or cause confusion. Framing the inquiry in this manner properly reflects that the proprietor enjoys a statutory monopoly across the full range of goods or services to which the registration extends.

  86. [86]

    To confine the inquiry to the actual use of the trade mark would frustrate the purpose of s 88(2)(c) – that purpose being to maintain the integrity of the Register by ensuring that no registration of a mark is to continue if it is not valid. It would permit the registered owner of a trade mark to resist rectification on the basis of its narrow use of the trade mark with a broad specification, even while there would be a real, tangible danger of deception or confusion if the mark were used beyond that narrow use. Put another way, the trade mark would not function as an accurate badge of origin for the full range of goods and services in respect of which it was registered because, if it were used in a different but normal and fair manner, it would be likely to deceive or cause confusion. Was the use of the Designer's Mark likely to deceive or cause confusion?

  87. [87]

    The inquiry required by s 88(2)(c) involves considering the notional normal and fair use of the registered trade mark – how the mark could be used in a hypothetical, typical market, involving "all legitimate uses to which [the] mark might be put by its owner" – and then asking whether that use is likely to deceive or cause confusion. The inquiry as to the cause of the likely deception or confusion is necessarily broader than that required under s 60. The cause does not need to be the reputation in Australia of another mark. The cause is limited not by reference to any particular sections of the TM Act, but rather by reference to the circumstances that apply as at the date on which the application for rectification is filed. So, for example, it has been suggested that, if a trade mark is allowed to become so descriptive that other traders will be able to use it in good faith descriptions of their own products as permitted by s 122(1)(b) of the TM Act, the registered owner's use of the trade mark for the purpose of indicating trade origin might then lead the public to believe that the products of those other traders are those of the registered owner, which again may cause confusion or deception as contemplated by s 88(2)(c).

  88. [88]

    In this appeal, the inquiry is whether as at 20 December 2019 the notional normal and fair use of the Designer's Mark is likely to deceive or cause confusion. The answer to that question depends on whether or not a reasonable number of people may be caused to wonder if a trade connection exists between the Designer's Mark and the Singer's Mark. Although the inquiry under s 88(2)(c) is different to that posed by s 60, the applicable principles are analogous. There must be a real, tangible danger of deception or confusion – a mere possibility of deception or confusion is insufficient. On the other hand, evidence of actual confusion is not required.

  89. [89]

    As will be explained, by 20 December 2019 the likelihood of the notional normal and fair use of the Designer's Mark deceiving or causing confusion in the mind of a consumer – in the sense of causing them to wonder whether there was a trade connection between the Designer's Mark and the Singer's Mark – was significant and, on any view, much greater than as at the Priority Date.

  90. [90]

    The matters relevant to the question of likely deception or confusion under s 60 as at the Priority Date still applied and were relevant to the inquiry under s 88(2)(c) as at the date on which the application for rectification was filed. Those matters were the similarity of the marks and the common practice of pop stars selling merchandise. But there were other matters relevant to the inquiry under s 88(2)(c) as at the date on which the application for rectification was filed. It was not disputed that, by that time, Ms Hudson had achieved international fame and the Singer's Mark had a significant reputation. The primary judge found that that was clear from the extensive evidence chronicling Ms Hudson's career trajectory, her international reputation and recognition, and the increasing range of merchandise marketed and sold to the public under the Singer's Mark, including, by that time, her shoe collection as well as other items of clothing. Put in different terms, by 20 December 2019, Ms Hudson had achieved international fame as a superstar and with that fame came substantial activity directed to exposing and exploiting the Singer's Mark on merchandise, as well as by brand extension to footwear, and the use of the name "Katy Perry", when used not as a mark, in media and by other means.

  91. [91]

    On the other hand, it must be accepted that, in relation to the actual use of the Designer's Mark, the primary judge found that, although Ms Taylor's label had online presence by that time and was being showcased from a showroom, her business was "very modest" and she and the Designer's Mark had "a much lesser reputation and little public presence". However, that does not detract from the conclusion that the likelihood of deception or confusion as at 20 December 2019 was even greater than at the Priority Date. Even despite the heightened reputation of the Singer's Mark as at 20 December 2019, there was no evidence that at that time the Singer's Mark was so notorious or well-known that consumers would have been aware of the correct spelling of the Singer's Mark.

  92. [92]

    Again, the absence of any evidence of actual confusion does not detract from that conclusion in circumstances where, despite the passage of ten years since the Designer's Mark was registered, it had achieved "little public presence". As explained, such evidence could only relate to confusion in relation to the limited use of the Designer's Mark. The absence of evidence of confusion based on such limited actual use of the trade mark does not preclude the conclusion that a broader normal and fair use of the trade mark was likely to deceive or cause confusion.

  93. [93]

    As the Full Court correctly observed, where, as here, the marks are deceptively similar and there is in fact little difference between them, the strong reputation in the Singer's Mark was more likely to enhance, rather than reduce, confusion. The Full Court was correct to conclude that Delfi Chocolate Manufacturing SA v Mars Australia Pty Ltd and Singtel Optus Pty Ltd v Optum Inc may be distinguished. In those cases, the Court reasoned that the strong reputation of the other mark meant that a consumer was likely to recognise differences between the marks and, after a moment's reflection, readily conclude that there was no trade connection between them. That conclusion is readily understandable where the two marks are "Maltesers" and "Maltitos", or even "Optus" and "Optum". By contrast, it is not evident that, even by 2019, the notional buyer would have been so familiar with the Singer's Mark as to recall its correct spelling and would not have been caused to wonder whether a hoodie, t-shirt or tank top bearing the words "Katie Perry" was connected to Ms Hudson. Conclusion

  94. [94]

    For those reasons, because of the circumstances applying at the time when the respondents filed the application for rectification of the Register, the use of the Designer's Mark in a hypothetical, typical market, across all legitimate uses to which that mark might be put by Ms Taylor, was likely to deceive or cause confusion in the mind of a consumer, in the sense that they would be caused to wonder whether there was a trade connection between the Designer's Mark and the Singer's Mark. Again, the fact that as at 20 December 2019 there were differences between the style of Ms Taylor's clothing (described as "luxury loungewear") and the merchandise of pop stars, such as "hoodies, t-shirts, tank tops and booty shorts", does not alter the conclusion reached. The inquiry under s 88(2)(c) is concerned with the notional normal and fair use of the Designer's Mark. For the reasons explained, the inquiry cannot be limited to the actual use of the Designer's Mark. As at 20 December 2019, the prominent display of the Designer's Mark on the front of a hoodie, t-shirt or tank top was a normal and fair use of the trade mark and it was likely that the use of the Designer's Mark in that manner would give rise to deception or confusion by reason of the international fame of Ms Hudson, the similarity of the marks and the common practice of pop stars selling merchandise. Section 89 of the TM Act – discretion not to grant application for rectification

  95. [95]

    Where a ground for rectification of the Register has been established, including the ground in s 88(2)(a) where the ground is that the trade mark is liable to deceive or confuse and the ground in s 88(2)(c), s 89 confers a discretion on the court not to grant the application for rectification. In either case, the discretion is only enlivened if the registered owner of the trade mark "satisfies the court that the ground relied on by the applicant has not arisen through any act or fault of the registered owner". Several aspects of s 89 require explanation.

  96. [96]

    First, the registered owner of the trade mark the subject of the application for rectification of the Register bears the onus of satisfying the court that the ground relied on has not arisen through any act or fault of the registered owner. Once they have discharged that onus, the court has a discretion to decide not to grant the application for rectification despite the ground being satisfied. Section 89(2) provides that, in exercising that discretion, the court must also take into account the prescribed matters and may take into account any other matter that the court considers relevant. Put another way, s 89(1) confers a broad discretion, constrained only by the general scope and objects of the TM Act and the requirement to consider the prescribed matters.

  97. [97]

    Second, the words "any act or fault" in s 89 form a composite phrase. Construing "a composite phrase simply by combining the dictionary meanings of its component parts" may be misleading. Here, construed as a whole, the expression "any act or fault" extends to a range of conduct – both acts and omissions – in which a registered owner of a trade mark might have engaged that gave rise to the ground relied on by the applicant for rectification. As the Full Court observed, s 89 refers to "act or fault", not just "fault". The question is whether the liability of the trade mark to removal on the ground of deception or confusion arose through the conduct of the registered owner of the trade mark.

  98. [98]

    The kinds of "act or fault" which might have given rise to the ground relied on will necessarily depend on the nature of the ground relied on. The word "fault" indicates that an inadvertent omission that gave rise to the ground relied on will not prevent the discretion from being enlivened. It may be accepted that an "act" may give rise to the ground relied on without any accompanying state of mind. In that sense, the inquiry may be distinguished from the concept of "blameworthy conduct", which developed as a concept specifically in connection with applications to rectify the Register by cancelling a trade mark due to circumstances that arose after registration. As has been explained, in Campomar, this Court held the concept of "blameworthy conduct" to be a "gloss" on the statutory text of s 28 of the 1955 Act. In any event, such a concept is inconsistent with the use of the word "act", which does not necessarily require fault on the part of the registered owner.

  99. [99]

    On the other hand, given the collocation of the word "act" with the word "fault", the registered owner's state of mind is not irrelevant to the inquiry as to whether the ground has arisen through their "act". It may readily be accepted that something more than the mere act of registering the trade mark is required to prevent the discretion from being enlivened. If it were otherwise, s 89(1)(b) would have no work to do where ss 60 and 88(2)(a) were relied on because the discretion would never be enlivened.

  100. [100]

    Third, for the discretion to be enlivened, the ground relied on must not have "arisen through" any act or fault of the registered owner. The words "arisen through" indicate a less direct or proximate causal relationship between the act or fault and the ground relied on than the words "caused by". The act or fault need not be the sole cause. It is sufficient if the relevant act or fault caused or substantially contributed to the ground relied on.

  101. [101]

    In this appeal, given the conclusion reached in relation to s 88(2)(a) read with s 60 and then s 88(2)(c), it is necessary to consider s 89 at two separate points in time – as at the Priority Date and as at the date when the respondents filed the application for rectification. Priority Date – 29 September 2008

  102. [102]

    Ms Taylor's complaint was that the Full Court did not rely on the fault limb of s 89(1) and, it having expressly accepted that her conduct as at the Priority Date was not "blameworthy", this meant that the discretion in s 89 was enlivened because the mere application to register a trade mark cannot be an act giving rise to a likelihood of deception or confusion for the purposes of s 88(2)(a) (or s 88(2)(c)) because otherwise the availability of the discretion under s 89 would inevitably be denied on each and every occasion.

  103. [103]

    In fact, the Full Court found that the ground for rectification of the Register by cancelling the Designer's Mark under ss 60 and 88(2)(a) arose through Ms Taylor's act of filing the trade mark application for the Designer's Mark with knowledge of Ms Hudson, her reputation and her mark, and in circumstances where Ms Taylor knew that popular music stars lend their names to be used in connection with clothes and sell merchandise at their concerts and accepted such conduct as a common practice. There was no error in that finding.

  104. [104]

    The Full Court correctly concluded that Ms Taylor's "act" substantially contributed to the ground under s 88(2)(a) arising. Ms Taylor applied to register the Designer's Mark with knowledge of Ms Hudson, the Singer's Mark and its reputation. She became aware of Ms Hudson in July 2008, before the Priority Date, when she first heard the song "I Kissed a Girl". By the Priority Date, she knew that Ms Hudson was a nationally and internationally famous pop star. Ms Taylor also accepted that she likely knew that musical artists sold clothing bearing their names and likenesses at music concerts and accepted that it was commonplace for musical artists to put out items of clothing bearing their names and likenesses. Knowing those facts, Ms Taylor decided to apply for registration of the Designer's Mark in relation to clothes and thereby substantially contributed to the likelihood of deception or confusion because of the reputation of the Singer's Mark. It was those matters combined that constituted the "act" that caused or contributed to the Designer's Mark being liable to rectification under s 88(2)(a) of the TM Act.

  105. [105]

    In those circumstances, Ms Taylor had not discharged her onus of satisfying the Court that the ground relied on and established by the respondents had not arisen through any act or fault of Ms Taylor. The discretion was not enlivened and there was no occasion to consider whether it should have been exercised. Date of application for rectification – 20 December 2019

  106. [106]

    The Full Court addressed separately the question whether, in relation to the respondents' success under s 88(2)(c), Ms Taylor had satisfied the threshold to enliven the discretion in s 89(1). That is, the Court considered whether Ms Taylor as the registered owner of the Designer's Mark had satisfied the Court that the ground relied on by the respondents had not arisen through any act or fault of Ms Taylor. The Full Court again identified that Ms Taylor's act of applying for registration of the Designer's Mark with knowledge of the matters identified gave rise to the likelihood of deception or confusion as at 20 December 2019 and for that reason the Court did not consider that the s 89 discretion was enlivened.

  107. [107]

    Ms Taylor's complaint again was that the Full Court did not rely on the fault limb of s 89 and, it having expressly accepted that her conduct as at the Priority Date was not "blameworthy", this meant that the discretion in s 89 was enlivened because the mere application to register a trade mark cannot be an act giving rise to the likelihood of deception or confusion for the purposes of s 88(2)(c) because otherwise s 89 would inevitably be denied on each and every occasion. That submission fails.

  108. [108]

    It may be accepted that Ms Taylor's act of applying for registration of the Designer's Mark with knowledge of the matters identified at the Priority Date was a necessary precursor to the circumstances applying on 20 December 2019, a date more than ten years after registration of the Designer's Mark. But those two matters combined that caused or contributed to the Designer's Mark being liable to removal from the Register under s 88(2)(a) of the TM Act were not the only acts or faults that caused or contributed to the likelihood of deception or confusion for the purposes of s 88(2)(c).

  109. [109]

    First, Ms Taylor encouraged the association between herself and Ms Hudson in the minds of consumers. On 9 February 2011, she posted an audio clip of a radio host mentioning her "Adelaide pants" when introducing Ms Hudson's single "Teenage Dream". On 17 May 2013, she posted an audio clip of an interview on a radio program in which the presenters of the program created an advertisement for Ms Taylor's label which included puns based on the titles of Ms Hudson's songs. On 3 July 2017, she posted on Facebook, "Heard my namesake Katy Perry is in town and wore an Alex Perry dress. Imagine if Katy Perry wore Katie Perry !!" In relation to the latter occasion, the primary judge found that "Ms Taylor once again took the opportunity to obtain publicity for her label by reference to Ms Hudson" and that that was "the only conceivable reason why she would create such a post on her Facebook page on which she promoted her own label". Such conduct had the potential to increase the likelihood of consumers being deceived or confused.

  110. [110]

    Second, the TM Act proceeds "on the fundamental assumption that the registered proprietor of a mark will protect the integrity of [their] mark by exercising [their] statutory right to restrain infringement of it". Given that fact, a failure to take advantage of the right to obtain relief against infringement, or to take some other action to enforce the statutory monopoly conferred on the registered proprietor, may in certain circumstances constitute, or at least contribute to, an "act or fault" that prevents the discretion from being enlivened.

  111. [111]

    Despite refusing to enter into a co-existence agreement with Ms Hudson, Ms Taylor did not instruct her advisers that she had any concerns about Ms Hudson using or licensing the Singer's Mark in relation to clothes in Australia. Moreover, although she knew in 2009 that Ms Hudson was proposing to use or license the Singer's Mark in relation to clothes in Australia, Ms Taylor had no concerns about that occurring and never conveyed any different position to any of the respondents until her letter of demand sent just prior to the commencement of these proceedings on 18 September 2019. In sum, Ms Taylor failed to commence proceedings for infringement or otherwise raise any objection to the respondents' conduct for over ten years, during which time the reputation of the Singer's Mark grew and the likelihood of confusion increased. That is, Ms Taylor's failure to enforce her statutory monopoly substantially contributed to the likelihood of deception or confusion.

  112. [112]

    Again, there is no occasion to consider whether the discretion should have been exercised since it was not enlivened. Conclusion and orders

  113. [113]

    The appeal must be dismissed with costs.

  114. [114]

    STEWARD J. I generally agree with the reasons of Jagot J that ss 88(2)(a) and 60 of the Trade Marks Act 1995 (Cth) are not satisfied in respect of the appellant's mark. It follows that it was validly entered in the Register of Trade Marks ("the Register") on 21 July 2009. To the extent they are consistent with those set out below, I also adopt gratefully the facts described in those reasons. I also generally agree with the reasons of Jagot J concerning the application of s 88(2)(c) of the Trade Marks Act.

  115. [115]

    However, even if the reasoning of Jagot J concerning the application here of s 88(2)(c) were thought to be mistaken, there is, in any event, a further reason for concluding that s 88(2)(c) is not satisfied in respect of the appellant's mark. These reasons describe why that is so.

  116. [116]

    In Re Bali Brassiere Co Inc's Registered Trade Mark and Berlei Ltd's Application, Windeyer J asked rhetorically: "Is a mark which was valid when registered to be removed because later events, perhaps the assiduous efforts of an infringer, have created a likelihood of confusion?"

  117. [117]

    This is a case where "the assiduous efforts of an infringer" or infringers, namely the respondents who sold clothing bearing the "Katy Perry" mark in Australia, led the Full Court of the Federal Court of Australia to order the removal of the appellant's mark from the Register. For the reasons which follow, that should not have happened. The effect of that order was to reward the wrongdoers – the "assiduous ... infringer[s]" – which were the respondents. The orders made by the Full Court below for the rectification of the Register should be set aside. A short chronology of events

  118. [118]

    The appellant, who was named "Katie Jane Perry" at birth, and who used "Katie Perry" as her name from about 2006 until she married Matthew Taylor in 2015, is a fashion designer based in Sydney. In April 2007, she applied to register "Katie Perry" as a business name for her new fashion and clothing business ("First Application"). She initially had planned to trade under the name "Palms", but was persuaded not to do so by a friend. In May 2007, she registered a domain name for her business at www.katieperry.com.au. In August 2007, she engaged a third party to design business cards, a letterhead and swing tags for her garments featuring a logo comprising her name "Katie Perry" in cursive script. In September 2007, she applied to register a trade mark for a logo featuring her name in respect of "clothing and fashion designing" for the first time. She commenced the manufacturing and sale of her clothes using her logo in 2008. The appellant had also, since 2007, promoted her "Katie Perry" brand on her personal Facebook page. In January 2009, she established a specific Facebook page for her business at www.facebook.com/katieperryclothing. Until July 2008, the appellant had never heard of the American artist known by her stage name as "Katy Perry" (the second respondent, her actual name being "Katheryn Hudson") who is hereinafter referred to as "the American artist".

  119. [119]

    The appellant's First Application lapsed in July 2008 because she had failed to pay the necessary registration fee. On 29 September 2008, she applied again to register "Katie Perry", on this occasion as a word mark, in class 25 for clothes ("Second Application"). In December 2008, the appellant received notice that her Second Application had received early acceptance. She accepted, in cross‑examination, that at that time of the Second Application "she likely knew that music artists sold clothing bearing their names at concerts". That generalised subjective understanding is not, of course, evidence of the reputation of the "Katy Perry" mark in Australia at that time.

  120. [120]

    When the appellant made the Second Application in 2008, the respondents had not sold any clothing in Australia, although a webstore, accessible worldwide, was established in October 2008 – just after the appellant's Second Application – for the sale of "Katy Perry" branded merchandise, including clothing such as T‑shirts. The American artist first toured Australia in October 2008. Her manager, Steven Jensen, became aware of the appellant's mark in May 2009. He informed the American artist by email on 16 June 2009 as follows: "We have learned that Katie Perry's real name (the Australian fashion designer) is Katie Howell. She has 'traded' under the name of Katie Perry since 2007. Of course, this issue has been blown way out of proportion as we (you) have not tried to keep her from trading under her name, and have certainly not sued her for trademark infringement. This popped up as the lawyers were submitting applications for trademarking your name around the world, including Australia, and the system automatically submits an 'objection' if anyone is trading under your trademark or any trademark that is similar to Katy Perry ... and they do not have a registered trademark. Apparently, Katie Perry is not trademarked in Australia. This is a common practice, and just makes the lawyers (and us/you) aware of any potential conflicts so we can deal with them if necessary. As is common in Australia, the tabloids have picked this up and made it into a 'story'. We are arranging for some publicity (probably a press release) to clarify the situation, and want to have that in place by end of business tomorrow. We'll keep you posted as we will probably want to include a statement from you. We're on it!"

  121. [121]

    The American artist wrote on 17 June 2009: "I say keep me outta it entirely ... Make it less important but release something from management, pretty much stating the facts. Like from the opening of this email, beside the Katy, to the 'story' part. Those exact words are great. Don't soften it up, don't apologize, nothing. Let me know what’s going out before it does pls ... Stupid bitches. I wouldn’t have even bothered with this mtv hadn’t picked up this silliness. Dumb bitch! Rawr!"

  122. [122]

    As it happens, Mr Jensen incorrectly stated that the respondents had done nothing to stop the registration of the appellant's mark. First, in May 2009, the American artist filed a notice of opposition to the registration of the appellant's mark out of time, with an application for an extension of time to file that notice. Second, also in May 2009, the appellant received a cease and desist letter from Australian trade mark attorneys acting for the American artist.

  123. [123]

    In June 2009, the American artist applied to register the "Katy Perry" mark in Australia in respect of several classes of goods, including class 25 for clothes. It would appear that the American artist was advised that her application for an extension of time to oppose the appellant's mark would fail, so that application was withdrawn on 16 July 2009. Later, in November 2011, following an amended application by the American artist, the "Katy Perry" mark was entered in the Register, but with respect to classes of goods and services other than class 25.

  124. [124]

    On 21 July 2009, the appellant's mark "Katie Perry" was entered in the Register and she applied the mark on her clothing range. She also promoted the "Katie Perry" brand on Facebook and on her YouTube channel. Her enterprise remained at all times a self-funded small business.

  125. [125]

    In contrast, the American artist's fame grew rapidly. So did her business as a singer. In August 2009, she toured Australia again and sold "Katy Perry" branded apparel and merchandise at her concerts. This was done in full knowledge of the fact of the appellant's registered trade mark. At the time, the American artist's manager wrote to her and expressed the view that "the public would not confuse your Katy Perry branded tour merchandise with this lady's [ie, the appellant's] clothes". Indeed, he gave evidence at trial that since then he had not become aware of any cases of confusion. Nonetheless, he accepted that selling "Katy Perry" branded clothing in Australia was a "calculated risk". The primary judge found that the respondents, knowing that their conduct was infringing conduct, engaged in a "calculated disregard of [the appellant's] rights" and, on account of this, awarded additional damages.

  126. [126]

    In 2014, the American artist again toured Australia and "Katy Perry" merchandise including clothing was sold at her shows. In the same year, "pop-up" stores opened in Melbourne and Sydney selling "Katy Perry" merchandise. Between 2014 and 2016, "Katy Perry" merchandise was sold by Target Australia, both in stores and online through its website. Between 2014 and 2017, "Katy Perry" merchandise including clothing was sold by Australian retailer Harris Scarfe, and between 2015 and 2019 such merchandise was sold by another Australian retailer, Best & Less. Merchandise was sold during this time by a group of companies (which included Bravado International Group Inc and Bravado International Group Merchandising Services, Inc) ("Bravado") under licence from a company owned and controlled by the American artist called Kitty Purry Inc ("Kitty Purry") (the third respondent).

  127. [127]

    I would characterise the foregoing conduct as that of a persistent or assiduous infringer of the appellant's validly registered "Katie Perry" mark. The Full Court below thought otherwise. That was primarily because of the following considerations: "Ms Hudson honestly adopted her name as a mark in the United States in 2002. After learning of Ms Taylor and her mark, Ms Hudson proposed a co-existence agreement which was rejected. At all times Mr Jensen did not consider there was any likelihood of any consumers being deceived or confused. There was no deliberate 'swamping' of Ms Taylor's Mark."

  128. [128]

    With great respect, I do not agree. There are four pertinent matters to which the reasoning of the Full Court fails to give sufficient weight. First, much, but not all, of the merchandise sold in Australia was "Katy Perry" branded clothing, which was an infringement of the appellant's mark. Second, that infringing conduct was consistently pursued over a period spanning multiple years. Third, the infringing conduct occurred with knowledge, and in the face, of the appellant's mark, in circumstances where, in September 2009, the American artist had voluntarily amended her June 2009 application to register the "Katy Perry" mark in Australia to exclude goods in class 25. All of this conduct was very much deliberate. Fourth, this conclusion is not avoided because Bravado, under licence from Kitty Purry, sold the merchandise in question. Both Bravado and Kitty Purry were assiduous infringers.

  129. [129]

    At some point after 2009 and before 2018, the appellant became aware that "Katy Perry" branded clothing was being sold in Australia. But she did not take steps to prevent the infringement of her mark until 2019. The primary judge found, and it is not now disputed, that the appellant "did not consider that she was in a financial position to engage lawyers and initiate legal proceedings" nor "that anyone would act for her for free in litigation against" the American artist. It was not until 2018 that the appellant became aware of the concept of litigation funding and thought that this would make it possible for her to commence proceedings against the respondents. In 2019, the appellant commenced her infringement proceeding in the Federal Court of Australia. A defence of laches, relied upon by the respondents, was rejected by the primary judge. The decisions of the primary judge and of the Full Court

  130. [130]

    The primary judge found that Kitty Purry infringed the appellant's mark as a joint tortfeasor with Bravado. It was not in dispute that Bravado had infringed her mark. It is not necessary to detail the actual specific infringements. The primary judge would have found that the American artist had also infringed the appellant's mark, but for the availability of a defence for the use of one's "own name" in s 122(1)(a)(i) of the Trade Marks Act. The primary judge rejected the first and second respondents' cross-claim for rectification of the Register by the cancellation of the appellant's mark pursuant to s 88(2)(a) and (2)(c), and their defences on the grounds of laches (as already mentioned) and acquiescence.

  131. [131]

    On appeal, the Full Court allowed the respondents' cross-claim. Their Honours' reasons for doing so are set out in the reasons of Gordon A-CJ and Beech-Jones J. Section 88(2)(c) of the Trade Marks Act

  132. [132]

    Section 88 of the Trade Marks Act relevantly provides: "(1) Subject to subsection (2) and section 89, a prescribed court may, on the application of an aggrieved person or the Registrar [of Trade Marks], order that the Register be rectified by: (a) cancelling the registration of a trade mark; or (b) removing or amending an entry wrongly made or remaining on the Register; or (c) entering any condition or limitation affecting the registration of a trade mark that ought to be entered. (2) An application may be made on any of the following grounds, and on no other grounds: ... (c) because of the circumstances applying at the time when the application for rectification is filed, the use of the trade mark is likely to deceive or cause confusion; ..."

  133. [133]

    At trial, as already mentioned, the respondents cross-claimed seeking rectification of the Register by the cancellation of the appellant's mark, pursuant to s 88(2)(c). They claimed that as at the date of their application for rectification, which was 20 December 2019, the use by the appellant of her "Katie Perry" mark was "likely to deceive or cause confusion" due to the extensive use in Australia by that time of the "Katy Perry" mark. They argued that, by 2019, the American artist's reputation as a singer, as well as in relation to clothes and related goods sold using the "Katy Perry" mark, was very significant. Given that reputation, it was said, a purchaser of the appellant's clothing might wonder whether there was an affiliation between those goods and the American artist. In that respect, any notional normal and fair use of the appellant's mark would need to include its possible use to sell all types of clothing, including such clothing as sold by the respondents, whether under licence, or otherwise. It followed, it was contended, that the appellant's mark was likely to deceive or cause confusion if it were to remain on the Register and be used to sell her clothing. This submission was upheld in the Full Court.

  134. [134]

    It should not be accepted, for the reasons given by Jagot J, that the use of the "Katie Perry" mark on clothing in Australia was likely, at least by 2019, to deceive or cause confusion amongst consumers. The respondents failed to demonstrate that outcome on the evidence before the Court. However, even if that conclusion were mistaken, and the evidence had made out a likelihood of deception or confusion, no order for rectification in accordance with s 88(2)(c) would have been justified. The reasons which follow assume a likelihood of deception or confusion, contrary to the conclusion reached by Jagot J, which I otherwise respectfully agree with. They demonstrate why the likelihood of deception or confusion – even if any such likelihood had existed – would not have been that with which s 88(2)(c) is concerned. Taking advantage of wrongful conduct

  135. [135]

    An interpretation of an Act that permits a person to take advantage of their own wrongdoing should not be preferred. That is an application of the maxim "nullus commodum capere potest de injuria sua propria" – no man can take advantage of his own wrong.

  136. [136]

    The decision of Holden v Nuttall illustrates that proposition. In that case, a defendant, in possession of leased premises, by his wrongful conduct, sought to take advantage of certain regulations made under the National Security Act 1939 (Cth), by inflicting upon himself a state of "hardship" to sustain his possession of those premises. Herring CJ decided that the word "hardship", as used in the regulations, should "be limited as a matter of construction so as to avoid attributing to the regulation-maker the intention of bringing about an injustice or allowing a man to benefit from his own wrong". Herring CJ went on to observe: "It is certainly most undesirable that people should be encouraged to make use of the regulations for the purpose of acquiring benefits for themselves at the expense of the legitimate rights of others."

  137. [137]

    The maxim is not confined to the construction of contractual terms. As Holden illustrates, it can be a tool of statutory construction. Indeed, as Giles JA observed in Ruthol Pty Ltd v Tricon (Australia) Pty Ltd, it "founds a like rule of construction of statutes". Giles JA went on to explain as follows: "In Broom’s Legal Maxims ... it is said (at 191) that the maxim that no man shall take advantage of his own wrong is 'based on elementary principles' and 'admits of illustration from every branch of legal procedure'."

  138. [138]

    That is not to say that the principle lacks boundaries. In Ruthol, Giles JA cited the following explanation from the reasons of Baron Bramwell in Hooper v Lane: "[I]t seems to me that that rule only applies to the extent of undoing the advantage gained, where that can be done, and not to the extent of taking away a right previously possessed. ... [T]he maxim ... means that no one shall gain a right by his own wrong; and not that if he has a right, he shall lose it, or the power of exercising it, by a wrong done in connection with it". The predecessor legislation to the Trade Marks Act and taking advantage of wrongful conduct

  139. [139]

    The authorities on the proper construction of the predecessor legislation to the Trade Marks Act, namely the Trade Marks Act 1955 (Cth) ("the 1955 Act"), engaged with and often deployed the principle that that nobody can take advantage of their wrongful conduct, expressed by the maxim as set out above.

  140. [140]

    In Re Bali Brassiere, the applicant was a manufacturer of brassieres which were sold using the mark "Berlei". The applicant sought to have the mark, "Bali-Bra", owned by another company and also registered in respect of brassieres, expunged from the Register. Section 61(1)(b) of the 1955 Act relevantly provided that a registered mark was to "be taken to be valid in all respects, unless ... the trade mark offends against the provisions of [s 28] of this Act". Section 28 relevantly provided: "A mark – (a) the use of which would be likely to deceive or cause confusion; ... shall not be registered as a trade mark."

  141. [141]

    The foregoing language of "use" which is "likely to deceive or cause confusion" is, of course, the language of s 88(2)(c).

  142. [142]

    Section 61(1)(b) deployed the criteria relating to the initial registrability of a mark (in s 28) in order to determine the ongoing validity of an already registered mark. This gave rise to a dilemma: was the question of whether the use of the mark was "likely to deceive or cause confusion" to be determined at the date when the action seeking its expungement from the Register for invalidity was brought under s 61(1)(b), or was it to be ascertained at the date of registration, in the past? There was a conflict in the authorities concerning this issue. Windeyer J favoured the former test.

  143. [143]

    In his Honour's reasoning, Windeyer J quoted the 9th edition of Kerly's Law of Trade Marks to affirm the proposition that: "One of the most difficult problems of trade mark law is whether and to what extent it is a basic principle of the law that a properly registered mark which becomes deceptive may then be removed from the Register." It was in this context that Windeyer J rhetorically asked whether a mark which had been validly registered could ever be removed for causing confusion, because of the subsequent "assiduous efforts of an infringer". This is an example of the application of the maxim set out above in a trade mark dispute. It underpinned the conclusion that the ongoing validity of a mark – and thus the issue of rectification of, or expungement from, the Register – was not to be tested at the date of registration, in the past, because favouring such an interpretation could permit a person to take advantage of their own wrongdoing in the intervening period.

  144. [144]

    In Riv-Oland Marble Co (Vic) Pty Ltd v Settef SpA, the appellant had misappropriated the respondent's mark, using it on identical goods as those sold by the respondent over many years. The appellant had been the respondent's distributor in Australia, but continued to use the respondent's mark after the termination of their relationship. After 11 years of use of the mark in Australia by the appellant, the respondent successfully registered the mark. It then sued the appellant for infringement. The appellant relied upon s 28(a) of the 1955 Act to contend that the respondent's mark was likely to deceive or cause confusion because of the appellant's prior use of the mark for some time in this country. Bowen CJ and Lockhart J (in separate judgments) decided that the confusion was the product of the appellant's misconduct and found for the respondent. Once again, a literal construction of s 28(a) was rejected. Bowen CJ said as to that issue: "I am, I think, conscious of leaning against a construction of s 28 in regard to blameworthy conduct on the part of a proprietor which would lead to a situation described by Windeyer J where the proprietor of a mark might be denied or lose registration by reason of 'the assiduous efforts of an infringer'".

  145. [145]

    Lockhart J described any outcome whereby the assiduous infringer succeeds in gaining rectification of the Register as "an extraordinary result". His Honour said: "If ... s 28(a) has an operation which is not restricted ... then a trader who uses the registered trade mark of another trader may exploit that infringing use of the mark in the market-place to the point where it can be said that the use of the registered mark by the proprietor of the mark would be likely to deceive or cause confusion. ... It would be an extraordinary result if this conduct, which constituted infringing conduct, could produce either the invalidity of the mark or its liability to expungement. It is no answer to say that a court has a residual discretion to decline to order expungement even where a mark is shown to be invalid. The fact that the infringing conduct of a rival trader, in the circumstances to which I have referred, could produce the consequence that a registered mark becomes invalid is itself so extraordinary a result that its correctness could only be accepted as a last resort."

  146. [146]

    This Court returned to the case of the assiduous infringer in New South Wales Dairy Corp v Murray Goulburn Co‑op Co Ltd. In that case, the issue of when to apply the test in s 28(a) of the 1955 Act arose again in an application for rectification of the Register. Mason CJ and Brennan J (in separate judgments) and Dawson and Toohey JJ (in a joint judgment) agreed that one applied s 28(a) as at the date of the application for rectification. In this respect, their Honours broadly agreed with Windeyer J in Re Bali Brassiere. Once again, the threat of the assiduous infringer was used to favour this construction of s 28(a).

  147. [147]

    In Murray Goulburn, the appellant was a statutory corporation with responsibility for the production, supply and distribution of milk in New South Wales. It was the owner of a mark, being the word "Moove", registered for goods in class 29 of the 1955 Act, which included milk. It sold a flavoured milk product under the name "Moove" in New South Wales. The respondent was a large farm co-operative which manufactured and sold dairy products, including milk. It was the owner of a mark consisting of the word "Moo", also registered in respect of class 29. In 1988, it commenced the sale of flavoured milk in New South Wales under the name "Moo". The appellant sued the respondent for infringement of the mark "Moove" and for rectification of the Register to remove the mark "Moo".

  148. [148]

    In Murray Goulburn, four members of this Court held that a mark could be expunged if, following its registration, it became likely to deceive or cause confusion, and this had resulted from some blameworthy act of the owner (or a predecessor to the owner) or from some other disentitling act. Dawson and Toohey JJ found that s 28 required that, where a proprietor of a mark used it in a way which would be likely to deceive or cause confusion, the proprietor would not be entitled to protection by the courts, and, as such, their Honours were of the "view that a mark should be removed from the register if circumstances arise after the date of original registration which leave its proprietor without protection". This would include, but would not be limited to, a blameworthy act of the owner. For each of these four members of the Court, however, and for the reasons set out below, the threat of the assiduous infringer was a reason for rejecting the proposition that s 28(a) should be applied literally in every case. The same phrase is used precisely in s 88(2)(c) of the present Trade Marks Act. Again, for the reasons set out below, a literal reading of s 88(2)(c) should be rejected.

  149. [149]

    Murray Goulburn was, on this point, subsequently overruled in Campomar Sociedad, Limitada v Nike International Ltd. In the latter case, this Court unanimously decided that determining the likelihood of a mark deceiving or causing confusion required a prospective inquiry and no regard was to be had to the manner in which the mark had in fact been used in the past. However, the reasons in Campomar did not address the conduct of an assiduous infringer, because such conduct was not relevant to the Court's preferred construction of s 28(a) of the 1955 Act. But such conduct is now relevant to the application of s 88(2)(c) of the Trade Marks Act, as explained below. It follows that the following passages from the reasons in Murray Goulburn retain their relevance and authority.

  150. [150]

    First, Mason CJ in Murray Goulburn expressly agreed with Windeyer J's observation about the assiduous infringer in Re Bali Brassiere. His Honour said: "echoing the words of Windeyer J in Re Bali Brassiere ... , it is legitimate to lean against the literal construction of s 28(a) on the ground that it would not protect the registered proprietor from 'the assiduous efforts of a misappropriating user'".

  151. [151]

    The foregoing supported Mason CJ's construction of s 28(a) to the effect that a mark is only liable to be expunged if the use of it becomes likely to deceive or cause confusion and that likelihood is due to the fault or blameworthy conduct of the registered proprietor.

  152. [152]

    Second, Brennan J in Murray Goulburn reasoned that a literal interpretation of s 28(a) whereby rectification would be justified "whenever the use of the mark becomes likely to deceive or cause confusion" had to be rejected. That was because such a construction would: "expose the registered proprietor's statutory right to destruction at the hands of any person who creates the likelihood of deception or confusion. A premium would be placed on what Windeyer J called 'the assiduous efforts of an infringer'".

  153. [153]

    Third, Dawson and Toohey JJ in Murray Goulburn likewise rejected a literal interpretation of s 28(a), preferring instead the approach set out above. Any other construction, their Honours reasoned, would "allow" a person who "assiduously infringed" a mark "to exploit the infringement by producing a situation in which the use of the registered mark would be likely to deceive or cause confusion". Their Honours reasoned that in such a case: "The greater the infringement, the greater the probability that the infringing user would have a ground for claiming rectification against the registered user. There may be a residual discretion to refuse relief, but it seems preferable to construe the provision in such a way as to give no encouragement to such conduct." Campomar and the enactment of s 88(2)(c) of the Trade Marks Act

  154. [154]

    Nothing said in Campomar cast doubt on the foregoing particular aspect of the reasons in Murray Goulburn. Whilst Campomar did not follow the proposition that s 28(a) of the 1955 Act had ongoing ambulatory effect after the date of registration of a mark, as mentioned above, it did not refer at all to the case of the assiduous infringer, post registration. On the construction of s 28(a) preferred by the Court, there was no need to address the possibility of such a person so acting.

  155. [155]

    In any event, Campomar was not concerned with s 88(2)(c), as the 1955 Act applied in that case. The origin of the new Trade Marks Act, including s 88(2)(c), may be found in the 1992 report of the Working Party to Review the Trade Marks Legislation, which made recommendations to the then Minister for Science and Technology. One of the recommendations sought to resolve the conflict in the authorities referred to by Windeyer J concerning when to test the likelihood of deception or confusion. The Working Party favoured a test which required the likelihood of deception or confusion to be tested at the date of commencement of rectification proceedings, thus making relevant the circumstances applicable at that time, including the use of the mark. Necessarily, those circumstances also needed to include the conduct of any assiduous infringer. That recommendation, which led to the enactment of s 88(2)(c) in the Trade Marks Act, aligned the test for determining the likelihood of deception or confusion with that favoured by Windeyer J in Re Bali Brassiere and by Mason CJ, Brennan, Dawson and Toohey JJ in Murray Goulburn. Section 88(2)(c) and the assiduous infringer

  156. [156]

    Because, by its terms, s 88(2)(c) requires an examination of "circumstances applying at the time when the application for rectification is filed" the conduct of an assiduous infringer is rendered relevant to the issue of rectification; it is one of the "circumstances" to be considered. In that respect, the applicable reasoning from Murray Goulburn becomes compelling. It would, with respect, be startling if an outcome which was regarded as "extraordinary" for the purposes of ss 61(1)(b) and 28(a) of the 1955 Act could be accepted as a result intended by Parliament for the purposes of s 88(2)(c), when both statutory schemes address the grounds for rectification of the Register. It cannot be right that Parliament intended from 1996 onwards to reward the assiduous infringer. The target is clear.

  157. [157]

    It follows that the phrase "likely to deceive or cause confusion" in s 88(2)(c) cannot be literally applied in every case. As is well known, "[i]f the target of a legislative provision is clear, the court's duty is to ensure that it is hit rather than to record that it has been missed". The phrase must be construed in a restricted or "strained" way so that it does not include such deception or confusion which has occurred because of the infringing conduct of an applicant for rectification. That was exactly the approach adopted by a majority of this Court in Murray Goulburn, and there is no reason not to adopt the same approach to s 88(2)(c). This reading down of s 88(2)(c) is compelled by the necessity to avoid what Lockhart J described as an "extraordinary result". Parliament cannot have intended such a glaringly wrong outcome. The presence of section 89(1)

  158. [158]

    The presence of s 89(1) in the Trade Marks Act, which is set out in the reasons of Jagot J, does not compel any contrary conclusion. That is so for the following reasons.

  159. [159]

    First, s 89(1) addresses the conduct of the proprietor of the mark, rather than that of another person such as an assiduous infringer. It is concerned with a determination as to whether the proprietor's "act or fault" gave rise to the confusion alleged by the person seeking rectification. It is not directed at confusion caused by an assiduous infringer. In that respect, the ground for rectification in s 88(2)(c) is engaged where there exists a likelihood of deception or confusion for reasons other than persistent infringement. That might be because of changes in the market; changes in the products being sold; or because of the conduct of the proprietor of the mark, including how the mark has been or will be used. Section 89(1) is a dispensing power where rectification may be declined, notwithstanding the likelihood of deception or confusion arising from use of the registered mark, where the cause of that deception or confusion is not the conduct of the registered proprietor. Construed thus, s 89(1) operates harmoniously with s 88(2)(c).

  160. [160]

    Second, a registered proprietor should not be left with only a discretionary remedy, where that owner's mark has been knowingly and persistently attacked. I agree with Dawson and Toohey JJ's observation in Murray Goulburn that, in the face of a possible residual discretion in that case to refuse relief, "it seems preferable to construe [s 28 of the 1955 Act] in such a way as to give no encouragement" to the assiduous infringer. I also prefer not to give encouragement to those who seek to undermine the intellectual property rights of registered owners. In the recent decision of the Full Court in Firstmac Ltd v Zip Co Ltd, Katzmann and Bromwich JJ (with whom Perram J agreed) refused to rectify the register to reward the conduct of an assiduous infringer. Their Honours said that: "where, as here, the primary judge found that Firstmac [the registered owner] was without fault, and we have found that the respondents did not prove that their use of the Firstmac Mark was honest, to exercise the discretion to remove the Firstmac Mark would reward 'the assiduous efforts of a misappropriating user'. To pick up on what Brennan J said in Murray Goulburn ... , it would place a premium on those efforts."

  161. [161]

    It may thus well be the case that the principle also has a role in an application of s 89(1).

  162. [162]

    It may otherwise be accepted that delay in prosecuting infringement may be relevant to the exercise of that discretion. As McHugh J observed in Murray Goulburn: "[r]egistered proprietors have a duty to be vigilant in protecting their rights". But here, for the reasons given by the primary judge, the appellant had good reason to delay the bringing of proceedings against the respondents – she simply did not have the financial capacity to take on so famous and so successful a person as the American artist, and the companies that the American artist controls. When the appellant obtained that financial capacity, she took the step of commencing proceedings against the respondents. Disposition

  163. [163]

    The Full Court concluded that at the time of the application for rectification, when the reputation of the "Katy Perry" mark was even stronger than it was in 2008, consumers of the appellant's clothing "were likely to be deceived or confused". No other reason was suggested for such likely deception or confusion. For the reasons given by Jagot J, that conclusion was incorrect. However, even assuming the Full Court's conclusion to have been correct, no rectification was justified. That is because if there had been a likelihood of deception or confusion, it would have been the product of the respondents' (in addition to Bravado's) persistent and assiduous infringing of the appellant's registered mark. It thus would not have satisfied the ground for removal from the Register in s 88(2)(c), for the reasons given above.

  164. [164]

    It follows that there is no need to consider any application of s 89(1).

  165. [165]

    This appeal must be allowed with costs, and the orders proposed by Jagot J should be made. Introduction

GLEESON J.

  1. [166]

    The appellant is an Australian fashion designer, whose birth name is Katie Jane Perry. She has designed and sold clothes, particularly luxury loungewear, since about 2007, using the name "Katie Perry" as her brand. On 29 September 2008, the appellant applied for registration of the word mark "Katie Perry" in accordance with the scheme for registration of trade marks established by the Trade Marks Act 1995 (Cth). The appellant sought registration of the mark in class 25 for clothes.

  1. [167]

    On 21 July 2009, the appellant's mark was entered on the Register of Trade Marks and she became the registered owner of Australian trade mark number 1264761 ("the Katie Perry Mark" or "the Designer's Mark"). The date of registration of the Katie Perry Mark, which is the "priority date" for the registration of that mark within the meaning of the Act, is 29 September 2008. As a result of the registration of the Katie Perry Mark, the appellant obtained rights, including the right to relief under the Act if the trade mark is infringed.

  2. [168]

    The second respondent, known worldwide as Katy Perry, is a music artist and performer who adopted that stage name in 2002. The other respondents are companies associated with her. In June 2008, the second respondent released her second studio album "One of the Boys" which included the song "Ur So Gay" and singles including "I Kissed a Girl" and "Hot n Cold". The appellant accepted that, from around July 2008, Katy Perry has been a nationally and internationally famous pop star. On 26 June 2009, the second respondent lodged an application for registration of the word mark "KATY PERRY" as a trade mark in Australia in classes 9, 25 and 41. The second respondent withdrew her application for a trade mark in class 25 after an adverse examination report in response to the application expressed the view that the proposed trade mark in class 25 "closely resembles [the Katie Perry Mark] because the trade marks are visually very similar and phonetically identical". On 7 November 2011, the second respondent became the registered owner of Australian trade mark number 1306481 comprising the name "KATY PERRY" ("the Katy Perry Mark" or "the Pop Star's Mark") in classes 9 and 41.

  3. [169]

    The background to the commencement of these proceedings is set out in the judgment of Gordon A-CJ and Beech-Jones J. In 2019, the appellant commenced proceedings in the Federal Court of Australia, claiming relevantly that the respondents had infringed the Designer's Mark. The primary judge (Markovic J) granted relief in the appellant's favour, including a declaration that the third respondent had infringed the Designer's Mark as a joint tortfeasor by advertising, offering for sale and selling infringing goods: (1) during the Australian leg of the second respondent's world tour, called "Katy Perry: The Prismatic World Tour", between 7 November 2014 and 15 December 2014; (2) at pop-up stores in Sydney and Melbourne during the Prismatic Tour; and (3) on the "Bravado" website, which included the "Katy Perry" webstore. The primary judge also ordered for an inquiry as to damages or an account of profits in respect of the third respondent's infringing conduct.

  4. [170]

    By cross-claim filed in the Federal Court on 20 December 2019, the first and second respondents sought rectification of the Register by cancelling the registration of the Designer's Mark, relying on the grounds set out in s 88(2)(a) and (c) of the Act. The ground in s 88(2)(a) is "any of the grounds on which the registration of the trade mark could have been opposed under this Act". The ground in s 88(2)(c) is "because of the circumstances applying at the time when the application for rectification is filed, the use of the trade mark is likely to deceive or cause confusion".

  5. [171]

    The primary judge dismissed the cross-claim but a Full Court of the Federal Court of Australia allowed the respondents' appeal from that aspect of the primary judgment and ordered that the Register of Trade Marks be rectified by cancelling the registration of the Designer's Mark. In summary, the Full Court relevantly concluded that both grounds for cancellation of the Designer's Mark were established. The Full Court decided that the primary judge should have found: (1) as to the ground in s 88(2)(a) (read with s 60 of the Act), (i) the Pop Star's Mark had acquired a reputation in Australia before the priority date of the Designer's Mark, being 29 September 2008, so that the registration of the Designer's Mark could have been opposed under s 60 of the Act (s 60(a)), and (ii) because of the reputation of the Pop Star's Mark, the use of the Designer's Mark on clothes would be likely to deceive or cause confusion (s 60(b)); and (2) as to the ground in s 88(2)(c), consumers were likely to be deceived or confused by use of the Designer's Mark as at the date that the cross-claim was commenced, that is, 20 December 2019.

  6. [172]

    The Full Court also concluded that the discretion not to rectify the Register in s 89 was not enlivened in circumstances where that Court considered that the circumstances leading to the rectification order had arisen through acts or faults of the appellant.

  7. [173]

    It is unnecessary to consider other aspects of the Full Court's judgment because the appeal to this Court concerns only the first and second respondents' cross-claim seeking cancellation of the Designer's Mark. The appellant contends that the Full Court erred in finding that those respondents established: (1) the ground for cancellation in s 88(2)(a) (read with s 60 of the Act); and (2) the ground for cancellation in s 88(2)(c). Further, the appellant contends that the Full Court erred in finding that the discretion not to rectify the Register in s 89 of the Act was not enlivened. That issue arises only if the Full Court was correct in finding a ground for cancellation of the Designer's Mark. Section 88(2)(a): Cancellation of registration of a trade mark on the ground in s 60 of the Act

  8. [174]

    Section 60 is entitled "[t]rade mark similar to trade mark that has acquired a reputation in Australia". It provides that the registration of a trade mark "in respect of particular goods or services" may be opposed on the ground that: "(a) another trade mark had, before the priority date for the registration of the first-mentioned trade mark in respect of those goods or services, acquired a reputation in Australia; and (b) because of the reputation of that other trade mark, the use of the first-mentioned trade mark would be likely to deceive or cause confusion."

  9. [175]

    In this case, the Designer's Mark was registered in respect of a single kind of goods, being clothing. Accordingly, registration of that mark could only have been opposed in respect of the particular class 25 goods that are clothing.

  10. [176]

    Read together and in the context of s 88(2)(c), which is expressed in broader terms, ss 60 and 88(2)(a) permit cancelling the registration of a trade mark in respect of particular goods or services by reference to a ground that could have been relied upon to oppose the registration of the registered trade mark. That ground concerns the likelihood of deception or confusion from the use of a trade mark on account of the reputation of another mark, being a reputation that had been acquired in Australia before the priority date. Thus, s 60 directs attention to the reputation of the earlier mark before the priority date. As explained below, s 60 requires identification of the "notional normal and fair use" of the opposed trademark and an evaluation of the likelihood of deception or confusion from that use of the opposed mark "because of" the identified reputation of the earlier mark. Trade mark acquiring a reputation in Australia

  11. [177]

    The purpose of s 60 is to provide protection for well-known marks, whether registered or not. Such a purpose has been explained on the basis that "the registration or use of a confusingly similar mark will in most cases amount to an act of unfair competition, and may also be considered prejudicial to the interests of those who will be misled". A trade mark is described, in s 17 of the Act, as a "sign" used or intended to be used "to distinguish goods or services dealt with or provided in the course of trade by a person from goods or services so dealt with or provided by any other person". For a trade mark, its "reputation" consists in its recognition as a trade mark by the public generally, that is, as a signifier of the source of goods or services. The text of s 60 does not require any such reputation to be identified by reference to one or more of the classes into which goods and services have been divided for the purposes of the Act. In many cases, the reputation of a trade mark will relate to specific goods or services and not to a class established by the Regulations.

  12. [178]

    The broad definition of "sign" in the Act includes, relevantly, a name. Evidently, a name may have a reputation (or general public recognition) distinct from any reputation that the name may have as a trade mark, as in the case of an individual who has a reputation in relation to their personal qualities or talents. Those qualities might be related or unrelated to the use of the individual's name as a trade mark. Generally, an individual's reputation as to their personal character is distinguishable from any reputation of their name as a trade mark, although a personal reputation may affect the nature or extent of public recognition of that name when used as a trade mark.

  13. [179]

    The word "acquired" is an ordinary English word, not defined in the Act and not a term of art. "Acquire" has shades of meaning. One meaning is "to come to have or possess" or "obtain" a quality or feature, while another is "[t]o attract or achieve" regard or a specific reputation. The reference to achievement implicates the object, so that to "acquire" has also been defined to mean "to gain for oneself through one's actions or efforts". In the absence of any specification in s 60 that the relevant reputation must be obtained by the use of the trade mark, the phrase "another trade mark had ... acquired a reputation in Australia" in s 60(a) should be understood to mean that the trade mark had, in fact, come to possess general public recognition as a trade mark whether through its use, its intended use, or otherwise. Whether a trade mark had acquired a reputation in Australia by a relevant date is a question of fact which can be difficult to ascertain. Reputation is commonly inferred from a high volume of sales, together with substantial advertising expenditures and other promotions.

  14. [180]

    In this case, the relevant date for ascertaining whether the Pop Star's Mark had acquired a reputation in Australia is 29 September 2008. The primary judge found the Pop Star's Mark had acquired such a reputation in Australia by that date in relation to entertainment and music but not in relation to clothing. In reaching this conclusion, the primary judge relied upon the actual use of the Pop Star's Mark in Australia in relation to the provision of music and entertainment. The primary judge found that the name "Katy Perry" was used as a trade mark in Australia before the priority date in connection with the marketing and offering for sale of several singles, including "Ur So Gay" and "I Kissed a Girl", and the album "One of the Boys". The primary judge correctly distinguished between the reputation of the name "Katy Perry" as a trade mark and any separate reputation of the name as the name of the second respondent. The primary judge found that media coverage of the second respondent did not constitute trade mark use, although it referred to the Pop Star's Mark.

  15. [181]

    Accepting the primary judge's finding that the Pop Star's Mark had acquired a reputation in Australia by the priority date, the Full Court considered that the primary judge erred by finding that the reputation of the Pop Star's Mark was "confined" to music and entertainment. The Full Court considered that the evaluation of the reputation of the Pop Star's Mark should have taken into account the common practice of pop stars to sell merchandise including clothing at concerts, and to launch their own clothing labels, as well as the second respondent's personal reputation as Katy Perry.

  16. [182]

    As explained above, the Full Court was correct that s 60(a) does not require the relevant trade mark's reputation to be specific to the particular goods or services in respect of which the opposed mark is registered: the text does not express or imply such a limit on the relevant reputation. To the contrary, the text of s 60(a) distinguishes between the kind of trade mark registration that can be opposed ("a trade mark in respect of particular goods or services") and the kind of trade mark that can ground the opposition ("another trade mark [that] had ... acquired a reputation in Australia"). However, to the extent that the Full Court considered that the common practice of pop stars was relevant to the identification of the reputation of the Pop Star's Mark, that was an error in the absence of any evidential basis that the common practice, in fact, extended the reputation of the Pop Star's Mark beyond its actual use. Specifically, there was no evidence that the Pop Star's Mark had been used in Australia in accordance with the common practice. Arguably the primary judge's finding, accepted by the Full Court, that the Pop Star's Mark had acquired a reputation in relation to music and entertainment was less precise than was justified by the available evidence which demonstrated a reputation in relation to pop music and pop music performances, but neither party suggested that the primary judge's finding was not available on the evidence. Confusion because of the reputation of an earlier trade mark

  17. [183]

    The requirement of likelihood of causing confusion "because of" the reputation of the earlier mark directs attention to the degree and nature of that reputation, and the likely effect of that reputation upon the effectiveness of the opposed mark in performing its function as a trade mark. The relevant confusion is about the source of goods or services and whether the particular goods or services on which the opposed trade mark is used or intended to be used come from the same source as goods or services on which the earlier trade mark is used. As with the similar test in s 120(2) of the Act, the question is whether there is a "real, tangible danger" of confusion occurring. Again, as with the similar test in s 120(2), the relevant confusion to be considered is confusion of ordinary people who are customers likely to buy the particular goods or services that are the subject of the opposed trade mark registration.

  18. [184]

    The relevant "use" of the opposed mark is the use to which the mark could properly be put within the ambit of its proposed registration, often referred to as the "normal and fair use". The particular goods or services may be a class of goods or services within the classification system established by the Regulations, or a subset of a class or classes. In this case, the particular goods in respect of which registration of the Designer's Mark could have been opposed are a subset of class 25, namely, clothing.

  19. [185]

    Thus, the relevant test is whether, because of the reputation of the earlier trade mark acquired in Australia before the priority date, use of the opposed trade mark in relation to particular goods or services would be likely to cause ordinary people who would be likely to buy those goods or services to wonder if there was a trade connection between the two marks. Applied to this case, and subject to one qualification, the question is whether because of the reputation of the Pop Star's Mark in Australia, acquired before 29 September 2008, use of the Designer's Mark in relation to clothes would be likely to cause ordinary people who would be likely to buy clothes to wonder if there was a trade connection between the two marks.

  20. [186]

    The qualification concerns the requirement of "normal and fair" use. It would be inappropriate to hypothesise a use of the Designer's Mark that would seek to deceive or confuse consumers as to a connection with the Katy Perry Mark that did not exist. Such a use would be unfair.

  21. [187]

    The Full Court found that: "[o]nce the common practice [of pop stars selling clothing merchandise and launching clothing labels] is taken into account, together with the primary judge's finding that the Katy Perry Mark had a reputation in music and entertainment, and [the second respondent's] own reputation in [her stage name], and the correct comparison is made, ... the use by [the appellant] of her mark on clothes would be likely to deceive or cause confusion".

  22. [188]

    The so-called "correct comparison" was "between the reputation of the Katy Perry Mark, and a notional normal and fair use of" the Designer's Mark. The Full Court identified four matters said to lead to the conclusion that the primary judge erred in failing to be satisfied that, because of the reputation of the Pop Star's Mark as at 29 September 2008, the use of the Designer's Mark would be likely to deceive or cause confusion.

  23. [189]

    The first matter was that the primary judge "unduly confined the reputation of the Katy Perry Mark". As explained above, there was no such error. To the contrary, the Full Court erred by conflating the reputation of the Katy Perry Mark as a trade mark with the reputation of Katy Perry as an individual, and by considering that the Katy Perry Mark had a "reputation in clothes" in Australia, despite the absence of any use of that mark in respect of clothes.

  24. [190]

    The second matter was the primary judge's comparison of the Designer's Mark and the Pop Star's Mark, notwithstanding the parties' common ground that the two marks were deceptively similar. The Full Court criticised the primary judge's findings of differences between the marks and considered that the close similarity of the two marks was "highly relevant as to whether or not there is a likelihood of confusion".

  25. [191]

    The expression "deceptively similar", defined in s 10 and used elsewhere in several provisions of the Act, is not used in s 60. For the purposes of the Act, a trade mark is taken to be "deceptively similar" to another trade mark if it so nearly resembles that other trade mark that "it is likely to deceive or cause confusion". Thus, deceptive similarity under the Act is a deemed characteristic of a trade mark based on the resemblance between two marks. As the test in s 60(b) is concerned with the reputation of the earlier mark and its effect on use of the opposed mark, that test necessarily contemplates the effect of the opposed mark on an ordinary consumer who is aware of the earlier mark, rather than a side by side comparison of the two marks. As with the test for deceptive similarity, the notional buyer's imperfect recollection of the earlier mark is at the heart of the test for likely confusion because of the earlier mark's reputation. The important difference between the two tests is that the test in s 60 is not confined to the issue of resemblance between two marks.

  26. [192]

    The primary judge found that use of the Designer's Mark affixed to the same types of clothing a performer such as the second respondent might offer for sale and sell would not have caused deception or confusion. I agree that there is an apparent inconsistency between this finding and the parties' common ground as to the deceptive similarity of the two marks. That inconsistency indicates error by the primary judge in the assessment of the likelihood of confusion for the purposes of applying s 60 of the Act.

  27. [193]

    The third matter was the primary judge's placement of "undue weight" on the absence of actual confusion at and since the priority date. The Full Court considered that the absence was explicable by the small scale of the appellant's business and significant weight should not have been placed upon it when the relevant test involved consideration of notional normal and fair use and was not limited to actual use. The primary judge made two relevant findings on this matter. The first was that the continued use of the Designer's Mark since the priority date "without a skerrick of evidence of any confusion on the part of any member of the public tells against the existence of any deception or confusion in any person". The second was that the absence of any evidence of actual confusion was "a powerful factor" in the context of the continued use of the Designer's Mark, including in relation to t-shirts bearing the name on their front in large font as early as July 2008. These are matters of evaluation of the evidence, not indicative of error by the primary judge.

  28. [194]

    The fourth matter was the primary judge's reliance on actual use of the Pop Star's Mark subsequent to the priority date. In this Court, the appellant did not dispute the Full Court's finding of error by the primary judge in this respect. Actual use of a mark subsequent to the priority date is not relevant to the extent or nature of the mark's reputation prior to the priority date and therefore does not bear upon an assessment of the effect of that reputation upon the likelihood of confusion within the meaning of s 60(b).

  29. [195]

    It follows that the Full Court was correct to find that the primary judge erred in finding that the ground in s 60 was not made out. However, the Full Court, in its turn, erred in its finding that the ground in s 60 was established. The first error was in finding that the common practice of pop stars was relevant to ascertaining the reputation that had been acquired by the Pop Star's Mark. The second error was taking into account the second respondent's reputation in the name "Katy Perry" in ascertaining the reputation of the Pop Star's Mark.

  30. [196]

    Section 60 requires a complex assessment of the facts that bear upon the effect of the reputation of the earlier mark, assessed before the priority date, upon a hypothetical future use of the Designer's Mark. The hypothetical "normal and fair use" may involve normative considerations to the extent that the hypothesised use must be both normal and fair. Relevant facts may include: (1) the goods or services in respect of which the reputation has been acquired, and whether they are the goods or services of the kind in respect of which the opposed mark obtained registration; (2) characteristics of the earlier mark that affect an assessment of the reputation of that mark, including the degree of similarity between the two marks; (3) aspects of the earlier trade mark's reputation, which may affect the likelihood of confusion such as the strength of the reputation; and (4) any common practice concerning the use of certain kinds of trade marks, to the extent that such practice is capable of demonstrating likely confusion on account of the identified reputation of the earlier trade mark.

  31. [197]

    Having found error in the reasoning of the Full Court, it is necessary to decide what judgment should have been given on this ground for cancellation of the Designer's Mark. In my view, the primary judge's conclusion that the ground in s 60 was not made out is to be preferred for the following reasons. The reputation of the Pop Star's Mark, acquired in Australia before the priority date, was in relation to particular genres of entertainment and music, that is, pop music concerts and pop music. The reputation was in respect of different goods and services from the category of "[c]lothing" in respect of which the Designer's Mark was sought to be registered.

  32. [198]

    The respondents did not demonstrate that the use of the Designer's Mark would be likely to deceive or cause confusion because of the reputation of the Pop Star's Mark before the priority date, notwithstanding that the two marks were so similar that the parties agreed they were "deceptively similar". It is necessary to identify the possible permissible uses of the Designer's Mark as a registered trade mark against which the likelihood of confusion is to be tested. These must be normal and fair uses that could realistically have given rise to a possibility that the two marks had a common trade source before the priority date. Examples of clothes that were later sold in Australia using the Pop Star's Mark are hoodies, tank tops, t-shirts and booty shorts. Accordingly, possible normal and fair uses of the Designer's Mark are in respect of clothes of these types, either prominently on the clothes themselves or, less prominently, on a label affixed to the clothes or on other marketing materials for the clothes.

  33. [199]

    It is reasonable to think that an ordinary customer for clothes would have been aware that a pop star, or the owner of a mark with a reputation in relation to pop music concerts and pop music, might seek to enhance the reputation of the mark by merchandising clothes. Further, such an ordinary customer for clothes would have been aware that a pop star like Katy Perry might seek to build her brand by launching a clothing label.

  34. [200]

    Taking these matters into account, the likelihood of confusion was remote because ordinary customers who would be likely to buy clothes bearing the Designer's Mark would, if they thought about the possibility of connection with the Pop Star's Mark, also notice the necessary absence of any other features of the use of the Designer's Mark to connect it to the goods or services in respect of which the Pop Star's Mark had acquired a reputation. That absence is a corollary of the requirement that the hypothesised use of the Designer's Mark must be a normal and fair use. Ordinary people, who were aware of the Pop Star's Mark, and its reputation in relation to pop music concerts and pop music, would expect clothes sourced from an entity associated with that mark to convey that the wearer was a fan of Katy Perry. They would probably expect clothes associated with the Pop Star's Mark to bear some association with Katy Perry as a pop star apart from the bare name "Katie Perry", such as by displaying details of a concert tour by Katy Perry, or the name of a song or album released by Katy Perry. They would dismiss the possibility of a shared trade source where such clothes were not sold in a context that identified an association between Katy Perry and the clothes, such as at a musical performance by Katy Perry or on a website clearly associated with Katy Perry. That expectation would be reinforced by the assumption of an ordinary purchaser that the owner of the Pop Star's Mark would seek to use that mark to build the reputation of Katy Perry as a pop star.

  35. [201]

    Supposing that the Designer's Mark were to be used on clothing in some less obtrusive way, such as on a label, an ordinary person who recognised the Pop Star's Mark would be unlikely to be confused about whether that less obtrusive use demonstrated a trade association with the Pop Star's Mark. It is relevant that the reputation of the Pop Star's Mark was acquired only relatively shortly before the priority date. In that context, it is unlikely that such an ordinary person might think that Katy Perry had launched a clothing label in the short period after the reputation of the Pop Star's Mark was acquired, without prominently associating herself with the label, consistent with her persona as a pop star. Section 88(2)(c): Cancellation of registration because of the circumstances applying at the time when the application for rectification of the Register is filed The "circumstances applying"

  36. [202]

    The expression "circumstances applying" is broad. As first enacted, s 88(2)(c) appears to have been intended to apply to a case where a properly registered trade mark had subsequently become deceptive or confusing, as a result of circumstances applying at the time of filing the application for rectification of the Register, but which did not apply at the time of registration. Circumstances indicating that a trade mark has lost distinctiveness, being circumstances that apply at the relevant time, may establish the ground for rectification of the Register in s 88(2)(c). Thus, Gummow J once observed that "there will be cases where the concepts of lack of distinctiveness and deceptiveness tend to coalesce".

  37. [203]

    It was not in dispute that by the date of the filing of the cross-claim (20 December 2019), the second respondent had achieved international fame and the Pop Star's Mark had a "significant reputation". The primary judge noted that an increasing range of merchandise was marketed and sold to the public under the Pop Star's Mark, including a shoe collection as well as other items of clothing.

  38. [204]

    The primary judge did not explicitly identify the relevant "circumstances applying" as at 20 December 2019 but found that the ground in s 88(2)(c) was not made out in the absence of evidence of actual confusion over a period of ten years during which both marks were in use, having regard to the "strength" of the second respondent's reputation and the reputation of the Pop Star's Mark.

  39. [205]

    The Full Court concluded that the ground in s 88(2)(c) was made out for the same reasons that the ground in s 88(2)(a) (read with s 60) was satisfied. Their Honours did not find the absence of any evidence of actual confusion persuasive where that absence was explicable by the small scale of the use of the Designer's Mark. They found that the likelihood of confusion arose whether the relevant use was the actual or notional use of the Designer's Mark.

  40. [206]

    In this Court, the appellant argued that the ground in s 88(2)(c) should be assessed by reference to the actual use of the registered trade mark. That approach would substantially defeat the purpose of the s 88(2)(c) provision, which is to provide for the possible cancellation of the statutory monopoly conferred by the registration of a trade mark, being a monopoly over all uses in respect of which the mark is registered. The use of the present tense "is" in s 88(2)(c) (in contrast to "would be" in s 60) does not detract from this conclusion. Rather, it reflects the application of the test of "likely to deceive or cause confusion" to a presently registered mark, and not a mark in respect of which registration is sought for the future.

  41. [207]

    Having adopted its own reasons in relation to s 60, the Full Court erred in finding that the ground in s 88(2)(c) was made out for those reasons. The primary judge's conclusion was correct, albeit for the following different reasons.

  42. [208]

    The "circumstances applying" as at 20 December 2019 were: (1) Katy Perry and the Pop Star's Mark had achieved reputations, at least in relation to pop music concerts and pop music, that were stronger than as at the priority date of 29 September 2008; (2) the Pop Star's Mark was also used on clothing and shoes; and (3) the facts concerning actual use of each of the two marks after the priority date.

  43. [209]

    The likelihood of deception or confusion is to be tested against the possible permissible uses of the Designer's Mark as a registered trade mark, identified in relation to whether the ground in s 60 was made out. It remains relevant that an ordinary customer for clothing would have been aware that a pop star, or the owner of a mark with a reputation in relation to pop music concerts and pop music, might seek to enhance the reputation of the mark by merchandising clothes. Further, such an ordinary customer for clothes would have been aware that a pop star like Katy Perry might seek to build her brand by launching a clothing label. The courts below made no finding that the volume of sales of clothing or shoes under the Pop Star's Mark, or that advertising or promotions using the Pop Star's Mark in connection with clothing or shoes, demonstrated that the Pop Star's Mark had acquired a reputation in connection with clothes. The evidence did not support a conclusion that the increased recognition of Katy Perry and the Pop Star's Mark led to a likelihood of confusion from the use of the Designer's Mark, which would not have resulted from the strength of the reputation of Katy Perry and the Pop Star's Mark at the priority date. Section 89: Rectification of the Register may not be granted in certain cases

  44. [210]

    The appellant has established that the Full Court erred in finding that the respondents had established the grounds in s 88(2)(a) (read with s 60) and s 88(2)(c) of the Act in respect of the Designer's Mark. The correct conclusion is that neither ground was established. Accordingly, there is no basis on which to order rectification of the Register, and the issues concerning the discretion in s 89 do not arise for consideration. Conclusion

  45. [211]

    The appeal should be allowed. Orders should be made in the terms proposed by Jagot J. The questions in the appeal

JAGOT J.

  1. [212]

    The appeal arises in these circumstances. The appellant, Katie Jane Taylor (born Katie Jane Perry), was the registered owner of the trade mark "Katie Perry" in Australia in respect of goods within class 25, "[c]lothing, footwear, headgear". Ms Taylor succeeded before the primary judge (Markovic J) in establishing that the third respondent, Kitty Purry Inc, a company controlled by the second respondent, Katheryn Hudson (better known by her stage name of Katy Perry) had infringed the "Katie Perry" trade mark by using the trade mark "Katy Perry" in Australia in respect of clothing. Ms Taylor also succeeded before the primary judge in defending her "Katie Perry" trade mark from the application by the first and second respondents for rectification of the Register of Trade Marks ("the Register") by cancelling and removing Ms Taylor's "Katie Perry" trade mark from the Register.

  1. [213]

    The respondents (collectively "the Katy Perry parties") appealed and Ms Taylor cross-appealed from the orders of the primary judge. The Full Court of the Federal Court of Australia (Yates, Burley and Rofe JJ) would have allowed that part of Ms Taylor's cross-appeal concerning the status of Ms Hudson as a joint tortfeasor in the infringement of Ms Taylor's "Katie Perry" trade mark. That issue became moot, however, as the Full Court allowed the appeal of the Katy Perry parties and ordered the cancellation of Ms Taylor's "Katie Perry" trade mark and its removal from the Register.

  2. [214]

    The general significance of the appeal, beyond the immediate interests of the parties, is that it raises six questions of statutory construction about the provisions of the Trade Marks Act 1995 (Cth) ("the TM Act").

  3. [215]

    The first question is whether a trade mark can have acquired a reputation in Australia, within the meaning of s 60(a) of the TM Act, other than a reputation in respect of particular goods or services. The answer to that question is "no". A trade mark can acquire a reputation in Australia only in respect of particular goods or services.

  4. [216]

    The second question is whether a trade mark that has acquired a reputation in Australia within the meaning of s 60(a) of the TM Act in respect of particular goods or services, can be said for that reason to have acquired a reputation in Australia in respect of other goods or services. The answer to that question is "no". That a trade mark has acquired a reputation in Australia in respect of particular goods or services does not mean that the trade mark has (or can have) acquired a reputation in Australia in respect of other goods or services.

  5. [217]

    The third question is, given that s 60(b) of the TM Act permits registration of a trade mark to be opposed and therefore that s 88(1) and (2)(a) permit the cancellation of that trade mark if registered, can it be said that the use of that mark "would be likely to deceive or cause confusion" as provided for in s 60(b) "because of the reputation of [another] mark" acquired in Australia in respect of goods or services other than those in respect of which the other mark had acquired a reputation in Australia. The answer to that question is "yes". If a trade mark has acquired a reputation in Australia in respect of particular goods or services, it is possible that the use of a trade mark in respect of other goods or services would be likely to deceive or cause confusion "because of the reputation of [the first-mentioned] mark" as provided for in s 60(b) of the TM Act. Whether that is so or not is a question of fact in each case.

  6. [218]

    The fourth question is whether s 88(2)(c) of the TM Act, in saying that an application for rectification of the Register may be made on the ground that "because of the circumstances applying at the time when the application for rectification is filed, the use of the trade mark is likely to deceive or cause confusion", confines consideration of the use of the registered trade mark to the actual use made of that mark at the time the application for rectification of the Register is filed. The answer to this question is "no". Section 88(2)(c), in referring to "use", means the notional use of the trade mark (which means any normal and fair use across the ambit of the registration of the trade mark, be that use actual, intended or possible).

  7. [219]

    The fifth question is whether s 88(1) of the TM Act, in saying that "[s]ubject to subsection (2) and section 89, a prescribed court may, on the application of an aggrieved person or the Registrar, order that the Register be rectified by", relevantly, cancelling the registration, provides a source of power for a prescribed court not to cancel a registration separate from the source of power not to do so in s 89 of the TM Act. The answer to this question is "no". Properly construed, the sole source of power not to order rectification of the Register, in circumstances where the court has found a circumstance in s 88(2) of the TM Act to exist, is the power in s 89 of that Act.

  8. [220]

    The sixth question concerns s 89(1) of the TM Act, which enables the court to "decide not to grant an application for rectification" where a ground for rectification has been established under s 88(2) "if the registered owner of the trade mark satisfies the court that the ground relied on by the applicant has not arisen through any act or fault of the registered owner". The question is whether the "act or fault" of the registered owner can be constituted by nothing more than the fact of the registered owner having made the application for registration of the trade mark. The answer to that question is "no".

  9. [221]

    As will also be explained, the consequence of the answers to these six questions of construction is that the Full Court, while correctly acknowledging that "in undertaking an analysis of questions of fact and degree, involving matters of judgment, minds may well differ on the result [so that] mere difference in opinion as to the result does not mean that appealable error is involved in the judgment under appeal", incorrectly attributed errors to the primary judge.

  10. [222]

    The force of the considerations relevant to appellate jurisdiction summarised by Allsop J in Branir Pty Ltd v Owston Nominees (No 2) Pty Ltd remains compelling in a case such as the present – the "demonstration of error may not be straight-forward where findings or conclusions involve elements of fact, degree, opinion or judgment or when the findings or conclusions in question can be seen as made with the advantage of hearing the evidence in its entirety, presented as it unfolded at the hearing with the opportunity over the course of the hearing and adjournments for reflection and mature contemporaneous consideration and assessment, in particular in a long and complex hearing".

  11. [223]

    With those considerations in mind, this is not a case in which I can conclude either that the primary judge erred in some specific respect or that the primary judge's ultimate evaluations are unreasonable or untenable in order to give rise to material error despite the absence of apparent specific error. Accordingly, the Full Court ought not to have interfered with the orders of the primary judge other than to the extent required to give effect to its conclusion that part of Ms Taylor's cross-appeal should be allowed. The TM Act Provisions

  12. [224]

    A "trade mark", by s 6(1), has the meaning given in s 17. Section 17 provides that: "A trade mark is a sign used, or intended to be used, to distinguish goods or services dealt with or provided in the course of trade by a person from goods or services so dealt with or provided by any other person."

  13. [225]

    Section 6(1) defines "sign" expansively.

  14. [226]

    The "priority date" for the registration of a trade mark in respect of particular goods or services means: (a) if the trade mark is registered – the date of registration of the trade mark in respect of those goods or services; or (b) if the registration of the trade mark is being sought – the day that would be the date of registration of the trade mark in respect of those goods or services if the trade mark were registered.

  15. [227]

    The "use of a trade mark", by s 6(1), has a meaning affected by s 7(1)-(3). Section 7(1)-(3) expand the circumstances in which certain uses are use of a trade mark. Relevant to this case are the other provisions of s 7 as follows: "(4) In this Act: use of a trade mark in relation to goods means use of the trade mark upon, or in physical or other relation to, the goods (including second-hand goods). (5) In this Act: use of a trade mark in relation to services means use of the trade mark in physical or other relation to the services."

  16. [228]

    By s 19: "(1) A trade mark may be registered in accordance with this Act in respect of: (a) goods; or (b) services; or (c) both goods and services. (2) The registration of a trade mark may be in respect of goods or services of more than one class. ... (3) The regulations may provide for the classes into which goods and services are to be divided for the purposes of this Act."

  17. [229]

    By s 20(1): "If a trade mark is registered, the registered owner of the trade mark has, subject to this Part, the exclusive rights: (a) to use the trade mark; and (b) to authorise other persons to use the trade mark; in relation to the goods and/or services in respect of which the trade mark is registered."

  18. [230]

    By s 27(1) a "person may apply for the registration of a trade mark in respect of goods and/or services if" specified conditions are satisfied. By s 27(3)(b) the application must "specify, in accordance with the regulations, the goods and/or services in respect of which it is sought to register the trade mark". By s 27(5) an "application may be made in respect of goods and services of one or more of the classes provided for in [the] regulations". By s 32 the Registrar may decide the class in which goods or services are comprised.

  19. [231]

    By s 41(1) an "application for the registration of a trade mark must be rejected if the trade mark is not capable of distinguishing the applicant's goods or services in respect of which the trade mark is sought to be registered (the designated goods or services) from the goods or services of other persons".

  20. [232]

    By s 43 an "application for the registration of a trade mark in respect of particular goods or services must be rejected if, because of some connotation that the trade mark or a sign contained in the trade mark has, the use of the trade mark in relation to those goods or services would be likely to deceive or cause confusion".

  21. [233]

    Section 44 also relevantly requires an application to be rejected if the trade mark is substantially identical with or deceptively similar to a trade mark registered by or sought to be registered by another person in respect of similar goods or services or closely related goods or services, in specified circumstances.

  22. [234]

    Section 60 provides that: "The registration of a trade mark in respect of particular goods or services may be opposed on the ground that: (a) another trade mark had, before the priority date for the registration of the first-mentioned trade mark in respect of those goods or services, acquired a reputation in Australia; and (b) because of the reputation of that other trade mark, the use of the first-mentioned trade mark would be likely to deceive or cause confusion."

  23. [235]

    By s 88: "(1) Subject to subsection (2) and section 89, a prescribed court may, on the application of an aggrieved person or the Registrar, order that the Register be rectified by: (a) cancelling the registration of a trade mark; or ... (2) An application may be made on any of the following grounds, and on no other grounds: (a) any of the grounds on which the registration of the trade mark could have been opposed under this Act; ... (c) because of the circumstances applying at the time when the application for rectification is filed, the use of the trade mark is likely to deceive or cause confusion; ..."

  24. [236]

    By s 89: "(1) The court may decide not to grant an application for rectification made: ... (b) on the ground that the trade mark is liable to deceive or confuse (a ground on which its registration could have been opposed, see paragraph 88(2)(a)); or (c) on the ground referred to in paragraph 88(2)(c); if the registered owner of the trade mark satisfies the court that the ground relied on by the applicant has not arisen through any act or fault of the registered owner. ..." Construction

  25. [237]

    Because a trade mark is a sign used, or intended to be used, to distinguish the goods or services of one person from the goods or services of another person "in the course of trade" (s 17), a trade mark must be in respect of goods and/or services (s 19) and can be used only in relation to goods or services (s 7(4) and (5)). For the same reason, the exclusive rights a registered owner of a trade mark obtains are relevantly the exclusive rights to use the trade mark in relation to the goods or services in respect of which the trade mark is registered (s 20(1)). Consistent with this, an application to register a trade mark must be rejected if the trade mark is not capable of distinguishing the applicant's goods or services in respect of which the trade mark is sought to be registered from the goods or services of other persons (s 41(1)).

  26. [238]

    Within this statutory scheme it is apparent that when s 60(a) refers to "another trade mark" which has "acquired a reputation in Australia" (which need not be a registered trade mark) the provision presupposes that the trade mark has acquired a reputation in Australia in respect of particular goods or services. Section 60(a) does not expressly refer to the reputation of the trade mark in Australia being in respect of particular goods or services, but that is because the scheme of the TM Act as whole exposes that a trade mark is a sign in respect of goods or services enabling a reputation in respect of goods or services in Australia to be acquired. That is, the TM Act does not contemplate that there may be an unregistered or registered trade mark in respect of which the owner may have acquired a reputation in Australia unconnected to the goods or services the source of which the trade mark distinguishes from other goods or services.

  27. [239]

    So understood, s 60 contemplates two trade marks each in respect of particular goods or services. Those goods or services need not be the same goods or services, similar goods or services, or closely related goods or services. Rather, s 60 requires that because of the reputation the trade mark (registered or unregistered) has acquired in Australia in respect of goods or services, the use of the trade mark sought to be registered in respect of goods or services (in accordance with s 7(4) or (5)) would be likely to deceive or cause confusion.

  28. [240]

    These matters explain the answers given above to the first, second and third questions.

  29. [241]

    Construing s 88(2)(c) as if it confines the relevant consideration to the actual use of the registered trade mark at the time the application to rectify the Register is filed would fail to give effect to the text, context and purpose of that provision. It may be accepted that s 88(2)(a) concerns "grounds on which the registration of the trade mark could have been opposed under this Act", meaning that the prescribed court must find that such a ground or grounds existed at the time of the making of the application for registration. This does not mean that the scope of s 88(2)(c) is logically confined to the actual use of the trade mark at the time the application for rectification is filed. Section 88(2)(c), in saying that "because of the circumstances applying at the time when the application for rectification is filed ... the use of the trade mark is likely to deceive or cause confusion", is referring to the notional use of the trade mark. The notional use of a trade mark encompasses the actual, intended and possible use of the trade mark across the scope of its registration, provided that each such use is relevant only if it is a normal and fair use (and not an illegitimate or improper use) of the trade mark.

  30. [242]

    That s 88(2)(c) uses the present tense twice ("application ... is filed" and "use ... is likely to") does not have the effect proposed on behalf of Ms Taylor. The first use of the present tense is unavoidable as it fixes the relevant time of the consideration, being the time at which the application for rectification is filed. The second use of the present tense reflects only that the "use of the trade mark" means notional use (a concept which includes actual, intended and possible use of the trade mark if normal and fair in each case). On that basis, while s 88(2)(c) could have said "the use of the trade mark is [or would be] likely to deceive or cause confusion" (reflecting the terms of s 60(b)), that it does not do so is no indicator of a legislative intention to confine the scope of s 88(2)(c) to actual use of the trade mark at the time the application for rectification is filed. It is an indicator only that under s 88(2)(c) the trade mark must be registered, whereas under s 60 the trade mark is only proposed to be registered.

  31. [243]

    The illogicality of the contrary proposition is readily demonstrable. A registered trade mark may not be being used at the time an application for rectification is filed. In that event, s 88(2)(c) is nevertheless available by reference to the notional use of that trade mark. A registered owner of a registered trade mark, before an application for rectification is filed, may declare an intention to expand the use of the trade mark. On the construction proposed for Ms Taylor that declared intention would apparently not be within the scope of relevant "circumstances applying at the time when the application for rectification is filed" because the declared intention does not affect the actual use of the trade mark at that time and whether that actual use is likely to deceive or cause confusion. This cannot be correct.

  32. [244]

    This explains the answer to question four.

  33. [245]

    Section 88(1) of the TM Act says that subject to s 88(2) and s 89, a prescribed court may, relevantly, cancel the registration of a trade mark. Section 88(2) specifies the exclusive grounds on which an application for, relevantly, cancellation of a trade mark may be made. Section 89 specifies the circumstances in which the court (being the prescribed court referred to in s 88(1)) may decide not to grant an application for rectification. Given these provisions, the principle established in Anthony Hordern & Sons Ltd v Amalgamated Clothing and Allied Trades Union of Australia is engaged. The principle is that: "When the Legislature explicitly gives a power by a particular provision which prescribes the mode in which it shall be exercised and the conditions and restrictions which must be observed, it excludes the operation of general expressions in the same instrument which might otherwise have been relied upon for the same power."

  34. [246]

    Consistent with this orthodox approach to statutory construction the sole source of the power not to rectify the Register is to be found in s 89(1) and (2) of the TM Act, not s 88(1).

  35. [247]

    This explains the answer to question five.

  36. [248]

    The condition in s 89(1) of the TM Act ("if the registered owner of the trade mark satisfies the court that the ground relied on by the applicant has not arisen through any act or fault of the registered owner") presupposes that the trade mark is registered. The provision enables the court not to grant an application for rectification of the Register despite a ground for rectification being established if the registered owner satisfies the court as to the condition in s 89(1). In this context, the mere fact of having applied for and obtained registration of the impugned trade mark cannot itself be an "act or fault" within the terms of the condition in s 89(1). This answer, however, does not preclude the circumstances in which such an application was made, including the state of mind of the registered owner when making the application, from constituting the making of the application as an "act or fault" within the terms of the condition in s 89(1).

  37. [249]

    This explains the answer to question six. Facts Up to 29 September 2008

  38. [250]

    The relevant time for determination of the s 60 issues is "before the priority date for the registration of the first-mentioned trade mark", meaning the priority date of Ms Taylor's trade mark, "Katie Perry", being 29 September 2008, noting that this trade mark was registered in respect of goods within class 25.

  39. [251]

    The primary judge's key findings of fact leading up to that time included the following: (1) Ms Hudson adopted the stage name "Katy Perry" in 2002 and obtained a talent manager in 2004; (2) a single Katy Perry recorded but did not release was included as a track in a movie released in Australia in June 2005; (3) Katy Perry collaborated with and performed a song with a Christian band on The Tonight Show with Jay Leno in the United States in January 2006 and featured in a band's music video later that year; (4) Katy Perry transferred from another record label to Virgin in about April 2007; (5) Katy Perry released a single "Ur So Gay" and an associated music video on social media in November 2007 which "went viral"; (6) Katy Perry released an extended play ("EP") including the "Ur So Gay" single in January 2008; (7) Katy Perry performed the song "Ur So Gay" on a television variety program in the United States in April 2008; (8) Katy Perry released an album "One of the Boys" including the songs "Ur So Gay", "I Kissed a Girl", "Hot n Cold", "Thinking of You" and "Waking Up in Vegas" in June 2008, which ranked on music charts including in Australia and attracted media attention in Australia; and (9) Katy Perry went on tour in the United States between 20 June and 17 August 2008 performing about 40 concerts, and then toured in other countries commencing in the United Kingdom from 9 September 2008, with tickets for her Australian concerts going on sale on 12 September 2008 for a tour of Australia scheduled for 10 to 14 October 2008.

  40. [252]

    The primary judge also found that Katy Perry's "One of the Boys" album was number seven in Sanity Music's bestselling albums in Australia in the week prior to 7 August 2008. By 4 September 2008 "I Kissed a Girl" had been at number one in the Australian charts for six weeks. The singles "I Kissed a Girl" and "Hot n Cold" were both in the Australian "top ten" in the week of 15 September 2008, were reported as numbers four and five respectively on iTunes in 2008 and were both in ARIA's "End of Year Charts – Top 100 Singles in 2008" in Australia.

  41. [253]

    The primary judge recorded that Ms Taylor accepted that Katy Perry, by the priority date, was a nationally (ie, in Australia) and internationally famous pop star.

  42. [254]

    Steven Jensen, a partner of Direct Management Group Inc, a talent management agency which manages Katy Perry's career, first arranged for "Katy Perry" branded merchandise including limited items of clothing, heart-shaped glasses, lip balm (cherry chapsticks) and tote bags to be sold at Katy Perry concerts in the United States from 20 June 2008, having applied for trade marks of "Katy Perry" in the United States for "apparel" on 12 June 2008 and for "music, CDs and other goods" and "entertainment and other services" on 26 June 2008.

  43. [255]

    The effect of the evidence is that, before the priority date of 29 September 2008, there was no "Katy Perry" branded merchandise available for sale in Australia. Further, only a very limited range of "Katy Perry" branded merchandise had been made available for sale at Katy Perry's concerts in the United States and other countries as part of her first tour which started on 20 June 2008. Up to 20 December 2019

  44. [256]

    The relevant time for determination of the s 88(2)(c) issues is "the time when the application for rectification is filed", being 20 December 2019.

  45. [257]

    The primary judge's key findings of fact leading up to that time included the following: (1) the "Katy Perry web store" went live in October 2008 and enabled consumers internationally to purchase "Katy Perry" branded merchandise; (2) Katy Perry's tour of Australia from 10 to 14 October 2008 was successful and had major media coverage; (3) Katy Perry went on another world concert tour between January and November 2009, performing in Australia in August 2009; (4) throughout 2010 Katy Perry appeared as a guest performer on several prominent television shows in the United States; (5) Katy Perry released another album in early 2011 and again went on a world tour performing in 21 countries including Australia; (6) Katy Perry released a series of singles in 2012 and 2013 performed at various events in those years; (7) Katy Perry released another album in 2013 followed by another world tour between May and October 2015, performing in 38 countries including Australia, and also performed a single at the White House for an event honouring the Special Olympics; (8) Katy Perry performed at the Super Bowl half-time show in 2015 which was viewed by 118.5 million people, reported as the most-watched Super Bowl half-time show in history; (9) Katy Perry performed at multiple events in 2016 and 2017 including in 2017 at the Glastonbury Festival in the United Kingdom; (10) Katy Perry released an album in 2017 which was followed by a tour of 33 countries including Australia between September 2017 and August 2018 and became a judge on a television show "American Idol", broadcast in Australia until 2020; and (11) Katy Perry continued performing in 2019 at several significant events.

  46. [258]

    The primary judge also found that after the "Katy Perry web store" went live in October 2008 enabling consumers internationally to purchase "Katy Perry" branded merchandise, the "Katy Perry" branded merchandise range expanded, with new merchandising arrangements being made and merchandise being sold not only at concerts and online but also in retail stores in Australia. The first items made available for sale on the "Katy Perry web store" included a limited range of "Katy Perry" branded t-shirts and singlets, with the range expanding significantly over time to include shoes in 2016 and thereafter handbags. Mr Jensen's evidence

  47. [259]

    Mr Jensen first became aware of Ms Taylor's application for registration of the "Katie Perry" trade mark in respect of clothes in May 2009. Mr Jensen was concerned that Ms Taylor might be attempting to "obtain a financial advantage" from Katy Perry. Legal threats and negotiations ensued, the upshot of which was that Katy Perry offered a co-existence arrangement with Ms Taylor which Ms Taylor refused in August 2009. By that time Mr Jensen believed it to be in Katy Perry's best interests not to oppose the registration of Ms Taylor's "Katie Perry" trade mark as the dispute between Ms Taylor and Katy Perry had attracted publicity adverse to Katy Perry and, in any event, "he did not consider there to be any likelihood that members of the public would confuse clothing and other merchandise bearing the 'Katy Perry' name on it with Ms Taylor's clothes". Mr Jensen continued to hold this latter belief at all times thereafter.

  48. [260]

    There is no finding below as to whether Mr Jensen continued to believe that Ms Taylor might be attempting to "obtain a financial advantage" from Katy Perry after the cessation of the negotiations in August 2009. For its part, the Full Court observed that at "all times Mr Jensen did not consider there was any likelihood of any consumers being deceived or confused". Sections 88(2)(a) and 60 – application to facts

  49. [261]

    Section 88(2)(a) enables an application for rectification of the Register on "any of the grounds on which the registration of the trade mark could have been opposed" under the TM Act, which includes the ground in s 60. Section 60(a) Question

  50. [262]

    The question under s 60(a) is whether the trade mark "Katy Perry" had, before the priority date for the registration of the "Katie Perry" trade mark in respect of particular goods or services (29 September 2008), acquired a reputation in Australia. As explained, a trade mark, whether registered or not, can acquire a reputation in Australia only in respect of particular goods or services. A trade mark cannot acquire a reputation in Australia disconnected from the goods or services the source of which the trade mark distinguishes from the source of other goods or services. Answer

  51. [263]

    Before the priority date of Ms Taylor's "Katie Perry" trade mark, Katy Perry had acquired a reputation in Australia in her trade mark "Katy Perry" in respect of the service of "entertainment" and in respect of the goods of "recording discs" used to communicate recorded music. As noted, for the purpose of s 60, it is immaterial that there was no registered trade mark "Katy Perry" in Australia in respect of this service and these goods before that date. The primary judge and the Full Court were correct to so conclude. As the Full Court put it, on the facts in this case, the "reputation in [Katy Perry's] name was of necessity a reputation in her name as a mark", that reputation in the mark being in respect of the service of "entertainment" and the goods of "recording discs". Section 60(b) Question

  52. [264]

    The question under s 60(b) is whether, before the priority date, because of the reputation in Australia of the trade mark "Katy Perry" in respect of "entertainment" and "recording discs", the use of Ms Tayor's "Katie Perry" trade mark in respect of goods within class 25 would be likely to deceive or cause confusion. Primary judge's answer

  53. [265]

    The primary judge answered this question "no" because the primary judge considered that: (1) before the priority date, the reputation in the "Katy Perry" trade mark was a reputation in respect of entertainment and music; (2) before the priority date, the reputation in the "Katy Perry" trade mark did not extend to a reputation in respect of clothes; (3) there was no evidence of any confusion before the priority date or subsequently; (4) while the two trade marks are aurally identical, they are visually slightly different ("Katy" versus "Katie"); and (5) "Katy Perry" branded clothes usually included graphics or photographs of Katy Perry whereas "Katie Perry" branded clothes do not. Full Court's answer

  54. [266]

    In contrast to the negative answer the primary judge gave to this question, the Full Court answered this question "yes" because: (1) in its view, the primary judge unduly confined the reputation of the "Katy Perry" trade mark to entertainment and music when it extended to clothing due to the "common practice" of pop stars to sell clothing; (2) it was common ground that the "Katy Perry" trade mark and the "Katie Perry" trade mark were deceptively similar as defined in s 10 of the TM Act ("a trade mark is taken to be deceptively similar to another trade mark if it so nearly resembles that other trade mark that it is likely to deceive or cause confusion"); (3) the primary judge placed undue weight on the lack of evidence of actual confusion which was explicable by the fact that Ms Taylor's business always operated at a very small scale; (4) the statutory question had to be answered by reference to the notional use of the "Katie Perry" trade mark, being the normal and fair use of the mark across the range of goods in respect of which it was registered; and (5) the statutory question had to be answered by reference to the reputation in Australia of the "Katy Perry" trade mark before the priority date, not the "Katy Perry" branded clothes in fact sold in Australia after the priority date.

  55. [267]

    The Full Court concluded that: "Once the common practice is taken into account, together with the primary judge's finding that the Katy Perry Mark had a reputation in music and entertainment, and Ms Hudson's own reputation in the name 'Katy Perry', and the correct comparison is made, we consider that because of the reputation in the Katy Perry Mark, the use by Ms Taylor of her mark on clothes would be likely to deceive or cause confusion." Consideration

  56. [268]

    Contrary to the submissions for Ms Taylor, the Full Court should not be understood to have suggested that because it was common ground that the two trade marks were "deceptively similar" as defined in s 10 of the TM Act, it had to follow that s 60 was satisfied. Determining infringement of a registered trade mark in s 120 of the TM Act involves the test of the use of a trade mark being "deceptively similar" to the registered trade mark. The concept of trade marks being "deceptively similar" also determines if an application for registration of a trade mark must be rejected under s 44. In the context of these provisions, the "essential task is one of trade mark comparison; the resemblance between the two marks must be the cause of the likely deception or confusion ... the marks [being] judged as a whole, taking into account both their look and their sound" and consideration of the reputation of the registered (or to be registered) trade mark is impermissible. Section 60 has a different focus. The relevant effect ("would be likely to deceive or cause confusion") must be "because of the reputation of that other trade mark", being its reputation acquired in Australia which must be a reputation in respect of particular goods or services.

  57. [269]

    Given this, it also cannot be said, as Ms Taylor proposed, that the Full Court erred when it said that "[w]hilst some die-hard fans of [Katy Perry] may recognise the incorrect spelling ['Katie Perry'], the ordinary consumer with an imperfect recollection of the Katy Perry Mark confronted with a garment with the ['Katie Perry' trade mark] on it would be likely to be confused as to the source of the item and wonder whether it was associated with [Katy Perry]" merely because this involves consideration of a person's imperfect recollection. That s 60 focuses on the reputation of a trade mark in Australia does not mean that a person's possibly imperfect recollection of that mark is irrelevant to s 60. But, as the submissions for Ms Taylor said, because s 60 focuses on reputation, the strength of that reputation may be capable of dispelling any likelihood of deception or confusion for s 60 purposes even if the trade marks are "deceptively similar" for ss 44 and 120 purposes. Moreover, because a person would see the "Katie Perry" trade mark only in connection with "Katie Perry" branded clothing (be it on a website, in a store, or as worn by a person) the visual difference between "Katie Perry" and "Katy Perry" is also relevant to the s 60 question as the primary judge rightly recognised.

  58. [270]

    The Full Court's attribution of error to the primary judge in respect of her Honour's conclusion that the "Katy Perry" trade mark did not have a reputation in clothes is problematic. The Full Court reasoned that because there was evidence that before the priority date it was "the common practice of pop stars to sell merchandise including clothing at concerts and to launch their own clothing labels", the reputation of the "Katy Perry" trade mark could not be confined to "music and entertainment", so that the primary judge erred in "concluding that the Katy Perry Mark did not have a reputation in clothes".

  59. [271]

    The fact, however, is that not a single item of "Katy Perry" branded clothing had been sold in Australia before the priority date (29 September 2008), so there could be no reputation of the "Katy Perry" trade mark in respect of clothes in Australia at that time. Even in the United States, the sale of "Katy Perry" branded merchandise, which included very limited items of clothing, was sold only at Katy Perry concerts until the "Katy Perry web store" became operational in October 2008, after the priority date. A "common practice" of pop stars extending their commercial activities into other goods, including clothing, cannot give to the trade mark under which the pop star performs a reputation in Australia in respect of those other goods. The primary judge did not err in so concluding.

  60. [272]

    The primary judge was also correct to recognise that, given the terms of s 60 ("because of the reputation of", relevantly, the "Katy Perry" trade mark), the nature and expected consequences of the reputation as acquired in Australia in the minds of ordinary Australian people was relevant. This is because the question is whether, because of the reputation the "Katy Perry" trade mark had acquired in Australia (in respect only of "entertainment" and "recording discs"), the use (meaning the notional use as described above) of the "Katie Perry" trade mark on goods within class 25 in Australia would be likely to deceive or cause confusion.

  61. [273]

    That the "Katy Perry" trade mark had not acquired a reputation in Australia in respect of clothing is relevant to s 60. The s 60 question, adapted to the present case was whether – because of the reputation of the "Katy Perry" trade mark in Australia before the priority date (which, in fact, was a reputation in respect of the service of "entertainment" and the goods of "recording discs") – Ms Taylor's (notional) use of the "Katie Perry" trade mark in respect of goods within class 25 would be likely to deceive or cause confusion. That is, would such a use involve "a real, tangible danger of a reasonable number of people being caused to wonder whether a trade connection existed between" clothing bearing the "Katie Perry" trade mark and Katy Perry as the trade source of the "Katy Perry" trade mark in respect of "entertainment" and "recording discs".

  62. [274]

    If (as the Full Court concluded) the primary judge, in deciding the s 60 question, had given weight to the fact that "Katy Perry" branded clothes when sold in Australia after the priority date usually included graphics or photographs of Katy Perry whereas "Katie Perry" branded clothes did not, the primary judge would have erred, first, by not recognising that the relevant use of the "Katie Perry" trade mark was its notional use and, second, by impermissibly shifting focus from the reputation the "Katy Perry" trade mark had acquired in Australia before the priority date to the actual use of the "Katy Perry" trade mark in Australia after the priority date. As to the first purported error, the primary judge's reasons expose a full appreciation that "use" in s 60 means "notional use". Indeed, the primary judge rejected contentions for both parties about the relevant nature of the goods on the basis that "it is necessary to consider the notional and fair use of [Ms Taylor's 'Katie Perry' trade mark] on the Registered Goods, clothes, in the course of trade and, in particular, whether such use would be likely to cause deception or confusion". As to the second purported error, given the primary judge's reasons as a whole, no perceived infelicity of language in one or two paragraphs of those reasons can support a reasonable inference that her Honour lost focus of the relevance of the priority date and the reputation acquired in the "Katy Perry" trade mark before that time.

  63. [275]

    In these circumstances, it seems most unlikely that the primary judge committed the errors of reasoning attributed by the Full Court. This is reinforced by the fact that the impugned paragraphs of the primary judge's reasons are not concerned with the reputation of the "Katy Perry" trade mark per se but the question whether the use of the "Katie Perry" trade mark on goods within class 25 would be likely to deceive or cause confusion because of the reputation of the "Katy Perry" trade mark. Moreover, in those paragraphs the primary judge was rebutting a contention by the Katy Perry parties to the effect that her Honour should consider the notional use of the "Katie Perry" trade mark on "the types of clothes that are generally sold by touring pop stars: hoodies, T-shirts, tank tops and booty shorts". In that context, it is likely that the primary judge was trying to make a different point from the errors the Full Court perceived, a point which involves no error.

  64. [276]

    The different point has two aspects to it. The first aspect is that the notional use of the "Katie Perry" trade mark on goods within class 25, being confined to its normal and fair use, does not extend to an improper or illegitimate use or a use in bad faith. As such, it could not be posited under s 60 that, for example, the use of the "Katie Perry" trade mark might include an image of Katy Perry or a reference to the name of or logo or pattern associated with a song, EP, album or concert tour by Katy Perry or, for example, that "Katie Perry" branded clothing might be sold at or outside a concert of Katy Perry. No such use of the "Katie Perry" trade mark would be normal and fair.

  65. [277]

    The second aspect is that, in contrast, given the terms of s 60, it could (and should) be posited that, as the "Katy Perry" trade mark had a reputation in respect of "entertainment" and "recording discs" in Australia before the priority date, the nature and extent of that reputation is relevant to whether an ordinary person at that time in Australia would be likely to be deceived or caused to be confused by the use of the "Katie Perry" trade mark on goods within class 25. Given the nature and extent of the reputation of the "Katy Perry" trade mark in respect of "entertainment" and "recording discs" in Australia before the priority date, an ordinary person in Australia in fact aware of the reputation of that trade mark would expect that any use of the mark would expose, even highlight, the connection between the good or service (be it clothing or anything else) and Katy Perry, the pop star about to perform her first concerts in Australia. That expectation derives from the nature of the reputation in Australia of the "Katy Perry" trade mark and is part of the "surrounding circumstances" relevant to that reputation in which the possibility or likelihood of a person being deceived or caused to be confused must be evaluated.

  66. [278]

    With this in mind, it is also apparent that the fact that the "Katy Perry" branded clothing released in Australia immediately after the priority date included the names of Katy Perry songs highlighting the connection with Katy Perry as the source of the "Katy Perry" trade mark was probative evidence – not of the actual use of the "Katy Perry" trade mark after the priority date (which was irrelevant), but of the content of the expectations of ordinary people in Australia before the priority date "because of the reputation" the "Katy Perry" trade mark had acquired in Australia.

  67. [279]

    Further, if such an ordinary person in Australia was in fact aware of the reputation of the "Katy Perry" trade mark and saw clothing bearing the trade mark "Katie Perry" and nothing else to signify a connection between the clothing and Katy Perry, the pop star, then the primary judge's conclusion that it was not likely that such a person would be deceived or confused is readily understandable. At the least, the conclusion is unaffected by the errors the Full Court attributed to the primary judge.

  68. [280]

    Nor is it apparent that the primary judge, as the Full Court posited, gave "undue weight" to the lack of actual confusion at and since the priority date. The primary judge recognised that Ms Taylor's business was very small. Importantly, what the primary judge also recognised, albeit implicitly, was that it was for the Katy Perry parties to prove the likelihood of deception or confusion being caused. This is because the Katy Perry parties had applied for rectification of the Register by the cancellation of the "Katie Perry" trade mark and therefore were the moving party in respect of the s 60 question. The burden of proof being on the Katy Perry parties plainly underlies the primary judge's observation that Ms Taylor had given discovery of documents which would have exposed any evidence of confusion communicated to her, but there was not "a skerrick of evidence of any confusion on the part of any member of the public" even in the face of Katy Perry's increasing fame. It is only in that context, and given that Ms Taylor had been selling "Katie Perry" branded t-shirts since as early as July 2008, that the primary judge found the lack of any evidence of any confusion at any time "a powerful factor". It is not apparent why this analysis involved the giving of "undue weight" to the lack of evidence of confusion.

  69. [281]

    For these reasons, none of the errors the Full Court attributed to the primary judge are apparent from the primary judge's reasons. On that basis, given the evaluative nature of the statutory exercise, the Full Court should not have set aside the primary judge's conclusion about the s 60 question unless it considered that conclusion unreasonable or untenable despite its inability to find any specific error in the primary judge's reasoning process. Otherwise, the Full Court would be doing no more than substituting its own opinion on a contestable matter for that of the primary judge who had the benefit of hearing the entirety of the matter when the Full Court's appellate jurisdiction is confined to the correction of error. For the reasons that follow, I am not persuaded that the primary judge's conclusion on s 60 is unreasonable or untenable.

  70. [282]

    It may be accepted that, before the priority date, Katy Perry was a famous pop star in Australia and internationally. However, before the priority date: no "Katy Perry" branded clothing had been sold in Australia; only very limited "Katy Perry" branded clothing had been sold at concerts in the United States and in other countries commencing 20 June 2008; Katy Perry had not yet toured in Australia (tickets for her first Australian tour having gone on sale on 12 September 2008); and while it is possible that a person who had acquired a clothing item at a Katy Perry concert in the United States or another country wore the item in Australia before such items became available at the first Katy Perry concert in Australia or from the "Katy Perry web store" when it became operational (both in October 2008), the overwhelming likelihood is that exceedingly few, if any, people could have seen any "Katy Perry" branded clothing in Australia before the priority date.

  71. [283]

    In these circumstances, I am unable to conclude that there was anything unreasonable or untenable in the primary judge's conclusion that the Katy Perry parties simply had not adduced sufficient evidence to discharge their onus of proving the existence of a real, tangible danger of confusion.

  72. [284]

    Analysing the question from a different perspective from that taken by the primary judge, before the priority date, an ordinary person in Australia may have: (1) never heard of Katy Perry – in which event the person would have held no doubt at all, let alone a "reasonable doubt", about the source of the "Katie Perry" branded clothing or other goods within class 25; (2) been an ardent fan of Katy Perry – in which event it must be inferred that the person would have known not only that "Katie Perry" was a misspelling of "Katy Perry" but also that Katy Perry had not yet toured Australia and there was no "Katy Perry" branded clothing yet available in Australia; or (3) heard of Katy Perry, to a greater or lesser extent, but not have been an ardent fan of Katy Perry – in which event it must be inferred that it is most likely that the person would never have seen any item of clothing bearing the "Katy Perry" trade mark in Australia and, by reason of the nature of the reputation the "Katy Perry" trade mark had acquired in Australia at that time in respect of "entertainment" and "recording discs", would have expected any clothing bearing the "Katy Perry" trade mark in Australia to highlight the connection between the clothing and Katy Perry, the pop star, such as by including her image or the name of or logo or pattern associated with one of her songs, her EP, or her concert tour which was coming shortly to Australia. What such a person would not have expected at that time is for any clothing the source of which was Katy Perry, the pop star, to expose no such relationship with Katy Perry.

  73. [285]

    As a result, if such an ordinary person was confronted by the "Katie Perry" trade mark in respect of goods within class 25 in Australia before the priority date used in a normal and fair (and not illegitimate or improper) way as required by the concept of notional use, the reputation in the "Katy Perry" trade mark in Australia in respect of "entertainment" and "recording discs" was such that there was no real and tangible danger of confusion on the part of the ordinary person. It is possible that such a person might momentarily wonder if the "Katie Perry" trade mark was somehow connected to the "Katy Perry" trade mark but, given that no notional normal and fair use of the "Katie Perry" trade mark could extend to any form of trading off the reputation of the "Katy Perry" trade mark, the person would quickly realise in the circumstances, including the absence of any apparent connection with Katy Perry, the pop star, that the clothing was unconnected to Katy Perry, the pop star and trade source of the "Katy Perry" trade mark. The danger of confusion is possible but has not been proved by the Katy Perry parties to be real or tangible.

  74. [286]

    For these reasons, the primary judge's conclusion that the use of the "Katie Perry" trade mark in respect of goods within class 25 would not be likely to deceive or cause confusion because of the reputation that the "Katy Perry" trade mark had acquired in Australia before the priority date as required by s 60 of the TM Act cannot be said to be erroneous. Section 88(2)(c) – application to facts The question

  75. [287]

    The question under s 88(2)(c) is whether, because of the circumstances applying at the time when the application for rectification is filed, the use of the trade mark is likely to deceive or cause confusion.

  76. [288]

    Given the case as put before the primary judge, the Full Court and this Court, and the conclusions below, it is not necessary to determine whether the criterion "because of the circumstances applying at the time when the application for rectification is filed" in s 88(2)(c) of the TM Act can include circumstances caused or materially contributed to by wrongful conduct of the person said to be an "aggrieved person" who is applying for rectification of the Register. The primary judge's answer

  77. [289]

    The primary judge answered the question under s 88(2)(c) "no". In effect, the primary judge considered Katy Perry's celebrity and "star power" by 20 December 2019 meant that her name and trade mark "Katy Perry" had a "heightened reputation" compared to the position before the priority date. The primary judge accepted that it was "arguable" that an ordinary person in Australia might "be caused to wonder" if there was a connection between clothes bearing Ms Taylor's "Katie Perry" trade mark and Katy Perry at 20 December 2019, but concluded that s 88(2)(c) was not engaged for two reasons. First, there was no evidence of actual confusion over the entire period of ten years in which goods within class 25 had been sold by both Ms Taylor under the "Katie Perry" trade mark and by Katy Perry under the "Katy Perry" trade mark. Second, the heightened strength of the reputation of Katy Perry under the trade mark "Katy Perry" meant that an ordinary person in Australia, when confronted with the "Katie Perry" trade mark in respect of goods within class 25, after a "moment's reflection" would conclude that the goods were not from or connected with "Katy Perry". The primary judge characterised this latter consideration as "compelling". The Full Court's answer

  78. [290]

    The Full Court considered that the primary judge had erred and answered the question under s 88(2)(c) "yes". The Full Court explained that, in accordance with its reasoning in respect of s 60, it did "not consider that the absence of any actual evidence of confusion is persuasive, where that absence is explicable by the small scale of use of Ms Taylor’s Mark ['Katie Perry'] over the 10 years since it was adopted". Further, the cases to which the primary judge referred in respect of the relevance of the strength of the "Katy Perry" trade mark by 20 December 2019 were cases in which the trade marks were not deceptively similar (in contrast to the common position of the parties in the present case). The Full Court concluded that the heightened reputation of the "Katy Perry" trade mark by 20 December 2019 compared to the priority date in fact reinforced the conclusion that the "Katie Perry" trade mark was likely to deceive or cause confusion because of the circumstances at 20 December 2019. Consideration

  79. [291]

    The principal difference between the circumstances as they existed before the priority date of 29 September 2008 and at the time when the Katy Perry parties filed the application for rectification on 20 December 2019 is that it is apparent that, by the later date, Katy Perry had transformed from a nationally and internationally famous pop star into a nationally and internationally famous pop star and "celebrity". By this, it is meant that the evidence indicates that by around 2014 onwards Katy Perry was transforming from a person famous for her songs and their performance to a person famous as Katy Perry.

  80. [292]

    To the extent the Full Court's conclusions about s 88(2)(c) rely on its reasoning in respect of s 60, the conclusions are affected by material error (as explained above). Again, to the extent those conclusions attribute error to the primary judge, the attributed error is not discernible. The primary judge appreciated that Ms Taylor's use of the "Katie Perry" trade mark was very small compared to the Katy Perry parties' use of the "Katy Perry" trade mark. As the primary judge put it, Ms Taylor's small use of the "Katie Perry" trade mark "may be an explanation for the absence of evidence of past confusion". The primary judge continued, however, by observing "[t]hat said, ten years is a long time and in that period Ms Taylor’s business and the ['Katie Perry' trade mark] had online presence and continued to operate in an environment where [Katy Perry's] fame was on the rise. Despite that there was absolutely no evidence of confusion."

  81. [293]

    It must be recalled again that, in respect of the application for rectification, the Katy Perry parties bore the onus of proving a ground for rectification. Further, it is not as if the "Katie Perry" trade mark had not been used at all or used only to a trivial extent. At all relevant times Ms Taylor maintained a database of contacts which she regularly updated about "Katie Perry" clothing. She established and maintained websites to sell her clothes under the "Katie Perry" brand. She sold her "Katie Perry" clothing at Paddington Markets for seven years from August 2008, and from 2011 at the Mosman Market and The Rocks Markets. Ms Taylor sold her "Katie Perry" clothing from a showroom in Mosman Tuesdays to Thursdays and then from a shop in Paddington until 2015 before opening in a new location in Paddington in 2016. Ms Taylor established and maintained accounts for her "Katie Perry" clothing on Facebook, Instagram and thereafter YouTube.

  82. [294]

    It may be accepted that evidence of actual confusion is often persuasive whereas an absence of evidence of such confusion may be no more than an absence of evidence. The primary judge's evaluation, however, is unaffected by material error: ten years is a long time for two deceptively similar trade marks to be on the market in Australia with no evidence of any confusion, despite the attempts by the Katy Perry parties to find such evidence by way of discovery. That period was sufficiently long to support an inference from the absence of evidence of confusion that in fact it was not likely that people had been or would be deceived or confused. Indeed, in a different context (s 89(1)), the Full Court itself said that the "absence of any actual confusion over the many years of co-existence suggests that there has been no confusion on the part of consumers". The primary judge therefore did not err in drawing that inference.

  83. [295]

    Similarly, while the primary judge compared the strength of the "Katy Perry" trade mark at 20 December 2019 to the strength of the trade marks in the cases to which her Honour referred, the primary judge did not suggest that the conclusions in those cases dictated the outcome in the present case. It was no error for the primary judge to consider that the strength of the "Katy Perry" trade mark at 20 December 2019 was capable of supporting the conclusion that the "Katie Perry" trade mark in respect of goods within class 25 would not be likely to deceive or cause confusion. The trade channels used by Ms Taylor to sell her "Katie Perry" clothing (websites, social media, markets, showrooms) are evidence of the available trade channels for the purpose of assessing the notional use of the "Katie Perry" trade mark in respect of goods within class 25. Those available trade channels all expose the "Katie Perry" trade mark as a mark that, overwhelmingly, would be seen rather than heard, a fact which substantially decreases the materiality of the aural identity of the "Katie Perry" trade mark and the "Katy Perry" trade mark. Once it is accepted that at 20 December 2019 Katy Perry was a pop star and celebrity whose reputation has transcended her pop singer status, it is not possible to discern any error in the primary judge's analysis or conclusion that the Katy Perry parties in the cross-claim "[had] not made out" the ground for cancellation of the "Katie Perry" trade mark under s 88(2)(c). As a result, I prefer to express no view on the question whether s 88(2)(c) is to be construed as excluding any circumstance which has arisen through wrongful conduct of an applicant for rectification and, if so, whether that exclusion would apply to the Katy Perry parties on the facts of this case. Section 89 – application to facts

  84. [296]

    For the above reasons, the primary judge was correct to conclude that the Katy Perry parties had not established any ground for cancellation of the "Katie Perry" trade mark, with the consequence that no question arose under s 89 of the TM Act.

  85. [297]

    If a ground for cancellation had been established the power in s 89(1) not to rectify the Register would have been available only if Ms Taylor as the registered owner had satisfied the primary judge that "the ground relied on by the [Katy Perry parties] has not arisen through any act or fault of the registered owner".

  86. [298]

    The Full Court concluded that in respect of the s 60 ground (which it considered established) Ms Taylor had not merely filed her trade mark application for "Katie Perry" in respect of goods within class 25, but had done so knowing about Katy Perry, Katy Perry's trade mark, and the common practice of pop stars to use their trade marks in connection with clothing. In so concluding, the Full Court did not, as Ms Taylor would have it, find that her act of merely applying to register her trade mark was the "act" for the purpose of the condition in s 89(1). The "act" on which the Full Court relied was Ms Taylor's filing of the application in certain circumstances known to Ms Taylor. That analysis does not involve the error proposed by Ms Taylor.

  87. [299]

    Otherwise, the Full Court's analysis of the question whether Ms Taylor, as the registered owner of the "Katie Perry" trade mark, had satisfied the court that the ground under s 88(2)(c) had not arisen through any act or fault of Ms Taylor appears to have assumed that there was no difference between that question and the same question in respect of the ground under s 60. That assumption was incorrect. The condition in s 89(1) is the registered owner satisfying the court in respect of the "ground relied on by the applicant". This can mean only the ground relied on by the applicant as established under s 88(2). If the ground is s 88(2)(a) referring to s 60, it is that ground which is relevant. If the ground is s 88(2)(c), it is that ground which is relevant. The s 88(2)(c) ground is confined to "the circumstances applying at the time when the application for rectification is filed". While those circumstances may derive from the circumstances applying before the priority date, the two time periods are different. The s 89(1) question in respect of s 88(2)(c) is whether the s 88(2)(c) ground has arisen through any act or fault of the owner. Conclusions and orders

  88. [300]

    For these reasons, the primary judge made no material error. The Full Court erred in concluding that the grounds for rectification of the Register by the cancellation of the "Katie Perry" trade mark were established. As the Full Court did not decide grounds 5 to 7 in the notice of appeal filed by the Katy Perry parties the matter must be remitted to the Full Court which should decide the outstanding issues in the appeal including such orders as to costs of the appeal as ultimately appear to it to be appropriate.

  89. [301]

    The following orders are to be made: (1) The appeal be allowed with costs. (2) Orders 1, 3, 4, 5, 6, 7 and 8 made by the Full Court of the Federal Court of Australia on 22 November 2024 be set aside. (3) The matter be remitted to the Full Court of the Federal Court of Australia in respect of grounds 5, 6 and 7 of the appeal to that Court and any other outstanding issue including the costs of that appeal and of the proceeding before the primary judge.

Unofficial copy. Source: NSW Caselaw. Refer to the official version for authoritative text.