Skip to content
laws.sgby aturio
Grounds publishedDecided 21 Aug 2015
Contents

High Court

Rovio Entertainment Ltd v Kimanis Food Industries Sdn Bhd

Feedback & support

[2015] SGHC 216Case no. TA No 11 of 2014

What happened, in plain English

There is no plain-English summary of this judgment yet.

Ask about this judgment

Laws this judgment relies on

Grounds of decision

Official text from eLitigation

  1. George Wei J:

  2. [1] This is an appeal from the principal assistant registrar of trade marks’s (“PAR”) decision, regarding Rovio Entertainment Ltd’s (“the Opponent”) opposition to Kimanis Food Industries Sdn Bhd’s (“the Applicant”) Trade Mark Application No. T1204840G.

  3. The Opponent’s Trade Marks

  4. [2] The Opponent is a software development and entertainment company incorporated in Finland. Its most famous and successful product is the mobile phone game franchise Angry Birds, which was first released in 2009. According to the Opponent, the Angry Birds game has 200 million active users worldwide, and has seen 350 million downloads globally.

  5. [3] The Opponent is the registered proprietor of the following trade marks:

  6. (collectively, “the Opponent Marks”)

  7. [4] For convenience, the registered trade marks will be referred to as the “Red Bird Mark” and the “Angry Birds Word Mark”.

  8. [5] The Opponent also uses a stylised version of the Angry Birds Word Mark (“Fancy Word Mark”). The Fancy Word Mark is exhibited at [10] of the Opponent’s “Amended Grounds of Opposition”. While the Fancy Word Mark is not a registered trade mark in Singapore, the Opponent has asserted that it enjoys copyright in the Fancy Word Mark. The Fancy Word Mark and the Red Bird Mark were registered for copyright in the USA and China. I reproduce the Fancy Word Mark below:

  9. [6] Aside from being used in respect of the Angry Birds game, the Opponent Marks (as well as the Fancy Word Mark) appear to have been used in Singapore on a wide variety of merchandise, including toys, cakes, snacks and fast food. The merchandise is not manufactured and sold by the Opponent itself. It is sold by the Opponent’s licensees. In fact, the Opponent’s business model is at least in part reliant on licensing the Opponent Marks to a wide range of businesses for their use on merchandise. That said, the date on which the use of the Opponent Marks on merchandise commenced in Singapore and the scale of such use are somewhat unclear from the evidence before me.

  10. [7] There is some evidence that the Angry Birds concept has been marketed in Singapore. For example, it was promoted in association with Singapore’s 2011 and 2012 Formula 1 Night Race, and the Singapore Cable Car from June 2012. It has also been advertised at the Singapore Changi airport in November 2012, and at the Takashimaya Department Store in 2011.

  11. The Applicant’s Trade Mark

  12. [8] The Applicant is a snack food manufacturer. It is a company incorporated in Malaysia. It has sold its snacks under the brands “Jimbo” and “D-Jack” in Malaysia and abroad.

  13. [9] The Applicant applied to register the following mark in Trade Mark Application No. T1204840G on 5 April 2012:

  14. (Referred to as the “Angry Bite Mark”)

  15. The Decision below

  16. [10] Before the PAR, the Opponent objected to the registration of the Angry Bite Mark on the grounds of ss 8(2)(b), 8(4) and 8(7) of the Trade Marks Act (Cap. 332, 2005 Rev Ed) (“TMA”).

  17. [11] Five main grounds of opposition were relied on:

  18. (a) Pursuant to s 8(2)(b) of the TMA: The Angry Bite Mark is similar to the Opponent Marks in respect of identical or similar goods, and there exists a likelihood of confusion.

  19. (b) Pursuant to s 8(4)(a) read with s 8(4)(b)(i) of the TMA: The Angry Bite Mark (or an essential part of it) is similar to the Opponent’s earlier well known trade marks and indicates a connection in the course of trade with the likelihood of causing damage to the interests of the Opponent.

  20. (c) Pursuant to s 8(4)(a) read with s 8(4)(b)(ii) of the TMA: The Angry Bite Mark (or an essential part of it) is similar to the Opponent’s earlier well known trade marks, the use of which will cause dilution or take unfair advantage of the earlier well known trade marks.

  21. (d) Pursuant to s 8(7)(a) of the TMA: The Applicant’s use of the Angry Bite Mark constitutes passing off.

  22. (e) Pursuant to s 8(7)(b): The Angry Bite Mark infringes the copyright in the Opponent’s Red Bird Mark as well as the copyright in the Fancy Word Mark.

  23. [12] The evidence for the Opponent comprised two statutory declarations by Mr Harri Koponen and Ms Kati Levoranta. Evidence for the Applicant consisted of a statutory declaration of Mdm Huang Len Len. No expert or survey evidence was provided.

  24. (a) In relation to s 8(2)(b) of the TMA, the marks are more dissimilar than similar in totality, and there is no likelihood of confusion.

  25. (b) In relation to s 8(4) of the TMA, the Opponent Marks are “well known in Singapore” under s 8(4)(a), but are not “well known to the public at large” under s 8(4)(b)(ii) of the TMA. Regardless, the Opponent Marks and the Angry Bite Mark are not identical or similar. Further, the Applicant’s use of the Angry Bite Mark would not indicate a connection between its goods and the Opponent.

  26. (c) In relation to s 8(7)(a) of the TMA, under the law of passing off, there is insufficient evidence to demonstrate goodwill in the Opponent Marks, and there is no misrepresentation.

  27. (d) In relation to s 8(7)(b) of the TMA, there is no copyright infringement because there is no proof that the Opponent Marks are original creations, and there is no proof of copying.

  28. The present appeal

  29. [14] The matter is now before me on an appeal filed by the Opponent against the PAR’s decision. In the present appeal, the Opponent has dropped its objection on the ground of s 8(4)(b)(ii) read with s 8(4)(a) of the TMA which concerns earlier trade marks which are “well known to the public at large in Singapore”.

  30. [15] It pursues the appeal on the ground that the PAR erred in relation to s 8(2)(b), s 8(4)(b)(i) read with s 8(4)(a), s 8(7)(a) and s 8(7)(b) of the TMA.

  31. [16] The issues that arise for my consideration in this appeal can be neatly structured around the statutory grounds upon which the Opponent relies to object to the registration of the Angry Bite Mark.

  32. [17] The first broad issue is whether the Opponent’s objection on the ground of s 8(2)(b) of the TMA can succeed. Specifically, the following points were canvassed:

  33. (a) For the purposes of the similarity comparison in s 8(2)(b) of the TMA, can this court consider the separately registered Opponent Marks as a composite mark, or is the court limited to considering the Angry Bite Mark separately against the Red Bird Mark and the Angry Bird Word Mark respectively?

  34. (b) In light of the conclusion in (a), are the Angry Bite Mark and the Opponent Marks similar?

  35. (c) If the Applicant is allowed to register and use the Angry Bite Mark, is there a likelihood of confusion on the part of the public?

  36. I note in passing that in its submissions, the Applicant did not put in question whether the similarity of goods requirement is satisfied. Thus, this is not a live issue before me.

  37. [18] The second broad issue relates to the Opponent’s objection on the ground of s 8(4)(b)(i) read with s 8(4)(a) of the TMA. The following points arise for consideration:

  38. (a) Is the “whole or an essential part” of the Angry Bite Mark identical or similar to either of the Opponent Marks?

  39. (b) Are the Opponent Marks “well known in Singapore”?

  40. (c) Would the use of the Angry Bite Mark indicate a connection between the goods of the Applicant and the Opponent?

  41. (d) If so, is the connection likely to damage the interests of the Opponent?

  42. [19] The third broad issue relates to s 8(7)(a) of the TMA, namely whether the Defendant’s use of the Angry Bite Mark is liable to be prevented under the tort of passing off. The following sub-issues arise for consideration:

  43. (a) Do the Opponent Marks generate goodwill?

  44. (b) Will the Applicant’s use of the Angry Bite Mark amount to a misrepresentation leading to a likelihood of confusion and damage?

  45. [20] The fourth broad issue is whether the use of the Angry Bite Mark is liable to be prevented for copyright infringement under s 8(7)(b) of the TMA. Specifically, it is necessary to consider,

  46. (a) Whether the Opponent has subsisting copyright in the Opponent Marks or any other relevant works; and

  47. (b) Whether the Applicant copied a substantial part of any of the Opponent’s copyright works in creating the Angry Bite Mark.

  48. Issue 1: s 8(2)(b) of the TMA

  49. [21] Section 8(2)(b) of the TMA provides as follows:

  50. [22] In considering if s 8(2)(b) of the TMA is made out on the facts, the Court of Appeal in Staywell Hospitality Group Pty Ltd v Starwood Hotels & Resorts Worldwide, Inc and another and another appeal [2014] 1 SLR 911 (“Staywell”) at [15] has affirmed the step-by-step approach. This entails assessing the three requirements of similarity of marks, similarity of goods or services, and likelihood of confusion arising from the two similarities, systematically. The first two elements are assessed individually before the final element which is assessed in the round (see [15] of Staywell).

  51. Similarity of marks

  52. [23] There is no dispute that the Opponent’s marks are earlier trade marks under the TMA. An important preliminary issue, however, needs to be addressed before embarking on a comparison of the marks. This concerns whether the Red Bird Mark and the Angry Bird Word Mark can be treated as a composite registered trade mark for the purposes of comparison with the Angry Bite mark.

  53. Preliminary issue: Can the Opponent Marks be considered as a composite mark in the similarity comparison?

  54. [24] Before the PAR, the Opponent submitted that in engaging in the similarity comparison, the court may combine the Opponent Marks, even though they are individually registered as separate trade marks. This means that the court can consider if the Angry Bite Mark is similar to a composite mark made up of both the Red Bird Mark and the Angry Bird Word Mark.

  55. [25] I start by making a preliminary observation that the Opponent Marks were not registered as a series of trade marks under s 17 of the TMA. Indeed, it seems improbable that the Red Bird Mark and Angry Bird Word Mark would qualify as “a series of trade marks”, which is defined in s 17(2) of the TMA as “a number of trade marks which resemble each other as to their material particulars and which differ only as to matters of a non-distinctive character not substantially affecting the identity of the trade mark”. The significance of this shall be discussed below.

  56. [26] The PAR considered the issue, and found that the Opponent Marks should be treated separately. She reasoned that it is the mark as registered that is treated as a property right, and protected under the TMA. The Opponent could gain protection over a composite mark consisting of two of its earlier trade marks only if it had separately registered the composite mark as well.

  57. [27] On appeal, the Opponent submits that the case law cited by the PAR does not support her position that the court may not consider the Opponent Marks as a composite. The submission was that from a policy perspective, requiring trade mark proprietors to register their marks in various permutations in order to be afforded what is perhaps best described as “holistic protection” under the TMA was untenable.

  58. [28] That said, the Opponent submits that the composite mark could have been successfully registered as a trade mark. Given that the individual parts (the Red Bird Mark and the Angry Bird Word Mark) of the composite mark were successfully registered as trade marks (thereby appearing to satisfy the requirements of distinctiveness etc. and see section 101(1)(c) TMA), it followed that the composite mark would also have been registrable. Accordingly, since the Opponent does in fact use the Opponent Marks as a composite on its products, the court should consider it as a composite mark in conducting the similarity comparison.

  59. [29] The Applicant seeks to affirm the PAR’s reasoning, and disputes both the Opponent’s interpretation of the relevant cases as well as its policy reasoning.

  60. The relevant case law

  61. [30] The first case relied on by the PAR and the Applicant is Staywell. The PAR relied on the Court of Appeal’s statement that “assessment of marks similarity is mark-for-mark without consideration of any external matter” (Staywell at [15] and [20]) in concluding that the comparison must be “[a] mark-for-[a]-mark” ([50] of the PAR’s GD).

  62. [31] In my view, the Court of Appeal’s statement must be read in context of the issues raised by the parties and the facts of Staywell. I agree with the Opponent that the Court of Appeal, in making that comment at [20], did not specifically direct its mind to whether two separately registered marks may be considered as a composite at the similarity of marks stage. The Court of Appeal was merely making the point that “external matter”, such as factors that affect the relative importance of different types of similarities, is irrelevant at the similarity of marks stage.

  63. [32] However, another part of the Staywell judgment, not raised by the parties, is perhaps more directly on point.

  64. [33] When the Staywell case was before the High Court, the Judge found that the “ST REGIS” trade mark conveys the idea of royalty. The learned judge held that this idea of royalty was reinforced by the device (the fleur-de-lis device) used by the “ST REGIS” brand of hotels in its advertisements (see Staywell Hospitality Group Pty Ltd v Starwood Hotels & Resorts Worldwide, Inc and another [2013] 1 SLR 489 (“Staywell (HC)”) at [28]). The point is that the fleur-de-lis device, whilst it does not appear (from the report of the decision) to have been registered as a trade mark, was used by the St Regis hotel chain in its advertisements (see Staywell (HC) at [6].

  65. [34] However, the Court of Appeal in Staywell (at [36]) held that

  66. [35] Whilst the Court of Appeal in Staywell was not addressing the case of two related but separately registered trade marks belonging to the same proprietor, the holding lends support to the Applicant’s position that a court may not consider a proprietor’s other registered marks in the similarity of marks comparison. Other registered marks are in this sense also considered “external added matter or circumstances”. Instead, each opponent mark must be compared individually against the applicant mark; what is required is a mark for mark comparison. As the learned author of Ng-Loy Wee Loon, Law of Intellectual Property in Singapore (Sweet & Maxwell, 2nd Ed, 2014) (“Ng-Loy Wee Loon”) opines at para 21.5.18(b), even if the earlier trade mark is always used by its proprietor with a tag line, it is inappropriate to make the comparison between ‘the earlier trade mark plus tag line’ and the later mark.

  67. [36] Of course, the Opponent is entitled to argue that [36] of Staywell does not conclusively decide the preliminary issue before this court. This would be because the Court of Appeal was not explicitly considering the possibility of a composite consisting of and the “ST REGIS” trade mark being the mark in question. Instead, it was undisputed that the court was to compare the “ST REGIS” trade mark against the applicant mark. In that sense, the fleur-de-lis device was naturally “external matter”. However, if the Opponent is correct that the mark that is to be the subject of comparison can consist of a composite of two earlier registered trade marks, then neither of the two earlier registered trade marks would be “external matter”. Instead, they would be part of the mark which is the subject of comparison.

  68. [37] While this argument may be criticised for its reliance on overly fine technical distinctions, there is nevertheless some merit to the view that the Court of Appeal did not directly consider the permissibility of combining two earlier registered trade marks into one composite mark for the purposes of the comparison.

  69. [38] The second case to consider is Infamous Nut Co Ltd’s Trade Marks [2003] RPC 7 (“Infamous Nut”). This was a decision of the Appointed Person, Professor Ruth Annand, of the UK Trade Mark Registry.

  70. [39] In that case, the applicant was attempting to register four device marks, each comprising a monkey eating a monkey nut. The opponent had a series of registered trade marks consisting of monkeys holding a nut. The opponent argued that one factor to consider is that it has a “series” or “family” of “monkey holding a nut” marks. In addressing the sole question before the court, which is whether there was a likelihood of confusion under s 5(2)(b) of the UK’s Trade Mark Act (which is almost in pari materia to Singapore’s s 8(2)(b) of the TMA), the court held at [36] that “where the opponent relies on proprietorship of more than one earlier trade mark, the registrability of the applicant’s mark must be considered against each of the opponent’s earlier trade marks separately”.

  71. [40] Prima facie, this case lends support to the PAR’s and the Applicant’s position. However, the Opponent submits that Infamous Nut is inapplicable because that case concerned a common theme over a “family of marks”, rather than an attempt to combine two separately registered marks.

  72. [41] While the Opponent is correct in arguing that unlike the present case, Infamous Nut was concerned with the thematic similarity across a “family of marks”, it seems to me that the court’s approach is generally of guidance in situations “where the opponent relies on proprietorship of more than one earlier trade mark” (Infamous Nut at [36]).

  73. [42] The key finding of the Court in Infamous Nut is that “the registrability of the applicant’s mark must be considered against each of the opponent’s earlier trade marks separately” (Infamous Nut at [36]). This was based (at least in part) on the court’s interpretation of s 5(2)(b) of the UK’s Trade Marks Act 1994 (“UK TMA 1994”), which contemplates refusal of registration on the basis of an earlier trade mark as defined in the same statute. Nothing in this reasoning suggests that the court’s findings are or ought to be limited to its specific facts or trade marks registered as a series. Rather, it is of general guidance when considering situations where an Opponent relies on more than one earlier registered trade mark in opposing the registration of a new mark.

  74. [43] Indeed, I repeat my earlier observation that trade marks are only registered as a series when they resemble each other as to their material particulars and differ only as to matters of a non-distinctive character which do not substantially affect the identity of the trade mark. The UK’s statutory provision on when marks may be registered as a series is found at s 41 of the UK TMA 1994, which is similar to s 17 of the TMA. If it is impermissible to take into account other registered trade marks within a series for the purpose of comparison with the applicant’s mark, the same must follow (and even more so) where the earlier registered trade marks are not registered as a series.

  75. [44] The Opponent also cited McDonald’s Corp v Future Enterprises Pte Ltd [2005] 1 SLR(R) 177 (“McDonalds”) and The Polo/Lauren Co. LP v Shop in Department Store Pte Ltd [2006] 2 SLR(R) 690 (“Polo”) in support of its position that the court should take into account the way the earlier trade marks are used in the similarity of marks comparison.

  76. [45] In McDonalds, McDonald’s Corporation (“McDonalds”) opposed the applicant’s registration of three trade marks, each of which contained a stylised American eagle with the words “MacNoodles”, “MacTea” and “MacChocolate”. McDonalds had its own registered trade marks – “BIG MAC”, “MAC FRIES”, “SUPERMAC”, EGG McMuffin”, “McChicken”, “McNuggets” and “McDonald’s”. Given the applicant’s application for registration was made in 1995, the Court of Appeal had to consider s 15 of the previous Trade Marks Act (Cap 332, 1992 Rev Ed) (“Old TMA”), which states:

  77. [46] In mounting the attack on the applicant marks, McDonalds stressed that the applicant had copied the “distinctive prefix of the appellant’s family of marks”, namely, “Mc” (see McDonalds at [7]). In brief, the argument was that McDonalds had established a reputation and goodwill in the prefix “Mc”, and that the prefix had become an essential feature in the McDonalds’ family of marks. On this basis, it was claimed that by adopting the prefix “Mac”, the applicant may confuse the public by giving the impression that their products come from McDonalds. McDonalds’ opposition failed and its appeal to the Court of Appeal was dismissed.

  78. [47] In the present case, the Opponent relies on the Court of Appeal’s dicta in McDonalds at [26]. In considering the similarity between the marks in question under s 15 of the Old TMA, the Court of Appeal took into account the fact that McDonalds’ logo (the golden arches) was often used with McDonalds’ registered trade marks even though the logo does not form a part of the registered trade marks. Presumably, this is argued as lending support to the claim that in assessing similarity, one must consider how the Opponent’s registered trade marks are actually used. Specifically, it is relevant to consider if the Opponent Marks are used in conjunction with anything else (or with each other).

  79. [48] While the Court of Appeal in McDonalds did take into account how the earlier trade marks were used under s 15 of the Old TMA, this lends little support to the Opponent’s position in the present case. As the Applicant points out, the dicta in McDonalds at [26] was made in the context of a different statutory provision. Importantly, the court was assessing the likelihood that the applicant’s marks would “deceive or cause confusion”. What s 15 required was an examination of whether the use of the applicant’s trade mark or a part of the trade mark was likely to deceive or cause confusion or otherwise be disentitled to protection in a court of justice. The emphasis was on the use of the applicant’s mark and the likelihood of deception or confusion. By contrast, what is in issue under s 8(2)(b) of the present TMA is the similarity of marks comparison. This involves a mark-for-mark comparison, which precludes consideration of external matters such as the actual or potential use of the trade marks (Staywell at [20]). What is in dispute is whether the “mark” under the current TMA can consist of two previously registered trade marks combined into one composite mark. The McDonalds case does not speak to this question. It merely stands for the proposition that in assessing the likelihood of deception or confusion, actual use may be relevant.

  80. [49] The final case to consider is Polo. The appellant was the registered proprietor of six trade marks: “POLO”, “POLO BY RALPH LAUREN”, “RALPH LAUREN & Polo Player device”, “POLO RALPH LAUREN & polo player device”, “LAUREN RALPH LAUREN” and “RALPH LAUREN”. Whilst there were six registered marks, only the “POLO” word mark was in issue.

  81. [50] In brief, the appellant claimed that the respondent’s use of the “POLO PACIFIC” mark in a particular font, typeface and design infringed its “POLO” trade mark under s 27(2)(b) of the Old TMA (which is in pari materia with s 27(2)(b) of the current TMA). In particular, the appellant submitted that the judge below was wrong to hold, inter alia, that the mark “POLO” was not distinctive (Polo at [12(a)]).

  82. [51] At this juncture, I pause to note that in Polo, the appellant made an “argument based on a family of trade marks with the prefix POLO” (see Polo at [17]). With respect, it is not entirely clear to me what the appellant’s argument entailed. The appellant’s argument might have been that since the word “POLO” was the common denominator of some of its registered trade marks, it had acquired distinctiveness and a reputation in the minds of the public. On this basis, it appears that what was being claimed is that the “POLO” word mark should be regarded as similar to “POLO PACIFIC” despite the differences between the two marks.

  83. [52] In its judgment, the Court of Appeal applied the three-step approach and affirmed the High Court’s finding that the sign “POLO PACIFIC” was not similar to “POLO”. The Court of Appeal also found that there was no likelihood of confusion. What is of interest at this juncture is the court’s decision in relation to similarity of marks.

  84. [53] In assessing similarity of marks, the Opponent rightly noted that the Court of Appeal did consider how the “POLO” mark was used on the appellant’s products. In particular, the court found it relevant that the appellant never used the word “POLO” simpliciter on its products (Polo at [23]). However, it is vital to note that the above observations of the Court of Appeal were made as part of the court’s assessment of the distinctiveness of the “POLO” word mark (see Polo at [17] and [23]). What the Court of Appeal held, in effect, was that the “family of trade marks” argument was a non-starter. On the facts, there was no evidence that the appellant had done anything to promote the “POLO” mark on its own. There was no evidence of distinctiveness acquired through use. Therefore, the court held that the argument based on family of trade marks carried little weight (Polo at [17]).

  85. [54] In coming to its decision, the Court of Appeal referred to the decision of the Appointed Person (Geoffrey Hobbs QC) in 10 Royal Berkshire Polo Club Trade Mark [2001] RPC 32 (“Royal Berkshire”). In that case, the court found that the mark “POLO” was not similar to “10 Royal Berkshire POLO CLUB” even though the evidence before the Appointed Person established that the average consumer would expect “POLO” brand toiletries to come directly or indirectly from the same undertaking (Royal Berkshire at [27]). The Appointed Person held that the marks were not similar because he did not think that people exposed to the use of the applicant’s mark would notice that it contained the word “POLO” without also noticing that it contained the words “ROYAL BERKSHIRE” and “CLUB”. The message of the applicant’s mark was in the combination (Royal Berkshire at [31]).

  86. [55] Accordingly, I am of the view that the Court of Appeal was not establishing a proposition that the court ought to take into account how a mark is used in conducting a similarity of marks comparison. The decision does not support the Opponent’s position that its earlier marks can be combined and considered as a composite. I thus find that this case is of no assistance to the Opponent.

  87. [56] For completeness, I note that the Applicant also cited Time Inc. v Li San Zhong [2014] SGIPOS 14. However, given that the conclusions of the principal assistant registrar in that case were based largely on the PAR’s decision in our present case, I agree with the Opponent that little weight should be placed on it in the current appeal.

  88. [57] To sum up, based on what has been argued before me, there is no direct local authority on the issue of whether two earlier registered trade marks may be combined into a composite whole for the purposes of the similarity of marks comparison (even if Staywell may be argued to suggest that it should be disallowed). On the other hand, the English approach, based on the Infamous Nut, seems to be that any such combination is likely to be disallowed.

  89. My decision

  90. [58] In the absence of legally binding judicial authority, it is apposite to start from the words of s 8(2)(b) of the TMA, and consider them in light of the objectives of trade mark law.

  91. [59] Section 8(2)(b) of the TMA states:

  92. [60] The key interpretive question for the Court is whether “an earlier trade mark” can include a composite of two previously registered trade marks. Section 2 of the TMA defines an “earlier trade mark” as, inter alia

  93. The other parts of the definition of an “earlier trade mark” are irrelevant for our present purposes.

  94. [61] In her judgment (at [49]), the PAR alluded to the argument that given the consistent use of the singular (eg “an earlier trade mark”, “a registered trade mark”), a plain reading of the statutory provisions above suggests that an “earlier trade mark” must refer to a single trade mark. This militates against the Opponent’s position that in considering the similarity of marks, the Court may consider the applicant mark against a composite of two or more separately registered trade marks. Against this argument, the Opponent submits that the use of the singular and/or plural in statutes should not be taken literally unless there is evidence that Parliament specifically intended the singular and/or plural.

  95. [62] However, leaving aside for the moment well known unregistered marks, even assuming the Opponent is correct, under s 2 of the TMA, an “earlier trade mark” must still be “a registered trade mark” [emphasis added]. It is undisputed that the composite of the Opponent Marks is not “a registered trade mark”. The fact that it may have been registrable is neither here nor there. This thus brings into serious question the possibility of adopting the Opponent’s approach to the similarity of marks comparison (ie treating an “earlier trade mark” as being a composite of two separately registered trade marks) without doing violence to the statutory wording.

  96. [63] Of course, while the words of a statute are, and should be, the cornerstone of statutory interpretation, courts have on occasion departed from the literal meaning of statutory language in light of the purposes behind the statute. Such purposive interpretation is mandated under s 9A of the Interpretation Act (Cap. 1, Rev Ed 1985).

  97. [64] Is any such departure in favour of the Opponent justified in this case? The Opponent’s main submission on this front is that it is untenable to require trade mark proprietors to register every permutation of their registered trade marks. It is submitted that it is not uncommon for trade mark proprietors to use their registered trade marks in a variety of combinations. If the law did not protect composites of two or more separately registered trade marks belonging to the same proprietor, this would severely reduce protection for registered trade marks. This, it was argued, would be contrary to what was intended under the statutory regime. Moreover, allowing the registration of the Angry Bite Mark potentially gives the Applicant legal grounds to object to the Opponent registering the Opponent Marks as a composite should it wish to in future. This, it was said, would be unfair.

  98. [65] The Applicant submits that there is nothing untenable about requiring trade mark proprietors to register all the permutations of their existing marks that they wish to use. In fact, they assert that this is currently done by some trade mark proprietors. Moreover, allowing proprietors to mix and match different registered marks to form a composite, de facto registered mark provides excessive (indeed unwarranted) protection.

  99. [66] Having considered the submissions, I am of the view that the statutory purpose behind trade mark protection does not compel a departure from the plain words of the statute. On the contrary, it points against allowing the court to combine the Opponent Marks for the similarity of marks comparison.

  100. [67] The argument that it is impractical or untenable to expect trade mark proprietors to register each combination of the marks they intend to use has not been substantiated. Indeed, whilst there was little evidence on this before the court, it appears that businesses frequently register various permutations of their registered marks. In some cases, the registrations will be pursuant to separate and distinct applications. In other cases, a series of trade marks may be registered pursuant to a single application. It is true that over the course of time, a business may well evolve its trade marks and its use of its trade marks such that the registered marks may no longer sufficiently protect the trade marks that are used in practise. However, it does not follow that it would be unduly onerous or “unfair” to require a trade mark proprietor to register the evolved mark. On the contrary, the reverse is true. Certainty as to what is the mark protected by registration is important.

  101. [68] Section 36 of the TMA provides that a registered trade mark is personal property. Property rights are enforceable in rem. Registration confers rights both to object to the registration of other marks, and to sue other persons for trade mark infringement. As with in rem rights generally, liability for infringement is strict. In some cases, criminal provisions may also come into play. A key objective of the trade mark registration system and the trade mark register must be to give what is best described as fair warning to other business and the public generally. Members of the public whose duty it is not to infringe registered trade marks are entitled to know with objective certainty which trade marks (and what they comprise) are protected by registration.

  102. [69] Allowing a trade mark proprietor to claim protection over all the permutations of its registered marks would be granting excessive protection of a wholly uncertain scope. It limits competition from other businesses, opens them up to unpredictable risks of infringement, and potentially increases trade mark litigation generally. Even granting protection over a composite of just two separately registered marks which are used together in practise (as proposed by the Opponent in the present case) would create considerable uncertainty. By granting protection, what is meant is to treat the two separately registered trade marks as a single composite trade mark for the purpose of determining infringement under the Trade Marks Act. There is no reason why the trader should not be expected to register the composite mark either at the time when he applies to register the first two marks, or subsequently when his use of the registered trade marks and branding strategies makes registered protection of the composite mark desirable.

  103. [70] Given the above analysis, I affirm the decision of the PAR. Under s 8(2)(b) of the TMA, two earlier separately registered trade marks cannot be combined so as to form a single composite mark. This is also consistent with the substantive finding in Staywell that a court may not consider extraneous matter in comparing an earlier registered trade mark against the applicant mark.

  104. [71] Quite belatedly, at the hearing of the appeal itself, the Opponent attempted to argue that in any case, a composite mark consisting of the Opponent Marks has acquired the status of an unregistered trade mark. The Opponent submits that the unregistered trade mark is an “earlier trade mark” for the purposes of s 8(2)(b) of the TMA, given that it falls under the definition in s 2 of the TMA. While s 2 of the TMA undoubtedly leaves open the possibility that an unregistered trade mark may be an “earlier trade mark” if it is a well–known trade mark, I am of the view that the Opponent has not adduced sufficient evidence to show that a composite of the Opponent Marks has acquired the status of an unregistered trade mark. This point was not argued before the PAR (see [7] of the PAR’s GD), nor was it properly argued by the Opponent in his written or oral submissions for the appeal. Indeed, all that was produced was a few photographs where the Opponent Marks were used together. This is clearly insufficient to prove that such an unregistered trade mark exists.

  105. [72] For these reasons, I affirm the PAR’s decision on the preliminary issue, and shall proceed on the basis that the Angry Bite Mark is to be compared against the Red Bird Mark and the Angry Bird Word Mark separately.

  106. [73] It is well-established that in considering the similarity of two marks, the court should look at visual, aural, and conceptual similarity. These three aspects of similarity are merely meant to guide the court’s inquiry. Ultimately, the marks must be appraised as a whole: Staywell at [17]–[18].

  107. [74] Significantly, previous judicial authorities have uniformly affirmed the importance of what has been termed “the technical distinctiveness of a trade mark” in the similarity of marks comparison. As the Court of Appeal stated in Staywell at [25], “a mark which has greater technical distinctiveness enjoys a high threshold before a competing sign will be considered dissimilar to it”. I pause to say a few words on the meaning of “distinctiveness”.

  108. [75] In Staywell (at [24]), the Court of Appeal held that distinctiveness in the “technical sense” was to be understood “in contradistinction to descriptiveness”. As an example of what was meant by “technical distinctiveness”, the Court of Appeal (at [28] of Staywell) gave the example of a trade mark which comprised (in whole or part) a “non-descriptive word” or an “elaborate and inventive graphic device.” On the basis of these examples, an invented word (especially one that has been freshly coined and is meaningless) would also likely possess “technical distinctiveness”. Whilst the expressions “technical distinctiveness”, and indeed “inherent distinctiveness”, have found their way into judgments, I observe that the TMA does not in fact use the terms “technical” or “inherent”. Instead, what is stated in s 7(1)(b) of the TMA is that “trade marks which are devoid of any distinctive character” are not registrable. This does not however prevent registration where the trade mark has “acquired a distinctive character as a result of the use made of it”: s 7(2) of the TMA.

  109. [76] In my view, the essential point made by the Court of Appeal in Staywell was that in comparing the applicant mark to the opponent’s mark, the court is entitled to take into account the “dominant” components or features of the opponent’s mark (Staywell at [23]). Where the opponent mark has a feature that is clearly dominant, the fact that this feature is also found in the applicant mark may lead to a finding of similarity even if some other features are different or absent. Whether there is a dominant feature as such is a matter of impression and degree. The observations made by the Court of Appeal in Staywell at [23]–[29] go no further than the proposition that where a component or feature of a mark is technically distinctive (invented, non-descriptive etc), that feature may well (as a matter of fact and impression) be regarded as a dominant feature of the mark and hence affect the overall assessment of similarity. That said, the Court of Appeal underscored the point that the distinctiveness of the separate components of a mark must ultimately be related back to the impression given by the mark as a whole (Staywell at [26]).

  110. [77] Finally, it bears emphasis that the similarity of marks comparison is “more a matter of feel than science”: see Wagamama Ltd v City Centre Restaurants plc [1995] FSR 713 at 732. Indeed, the comparison of any two marks should take place at an impressionistic level, rather than on a quantitative or mechanistic basis: Staywell at [17].

  111. Role of the appellate court

  112. [78] Before proceeding further, it is important to clarify the role of an appellate court hearing an appeal from the PAR. In Future Enterprises Pte Ltd v McDonald’s Corp. [2007] 2 SLR(R) 845 (“Future Enterprises”) at [5], the Court of Appeal quoted SC Prodal 94 SRL v Spirits International NV [2003] EWHC 2756 (Ch), which states:

  113. [79] The Court in Future Enterprises continues at [7], stating:

  114. [80] Bearing in mind that the appellate court’s role is not to decide the case (and in particular, the factual findings) de novo, but to intervene only if there has been a material error of fact or law, I proceed to consider the similarity of marks.

  115. Comparison between the Red Bird Mark and the Angry Bite Mark

  116. [81] On visual similarity, the Opponent submits that the Red Bird Mark and Angry Bite Mark are similar because of the presence of the angry eyes, eye brows, and the bird like features in the latter (including the rather small tail feathers and beak like tongue). It further submits that the angry eyes and eye brows are the dominant features of both the Red Bird Mark and Angry Bite Mark.

  117. [82] The Applicant repeats the PAR’s reasoning on visual similarity. The PAR found that the only possible similarity is the “angry eyes device”. To be clear, by “angry eyes device”, what is being referred to are “the eyes and eyebrow” of the red bird. These are said to depict anger, hence the description “angry eyes device”. The angry eyes are not a device separate from the Red Bird Mark. The applicant submits that even if the angry eyes were the dominant component of the two marks, the court cannot focus on that alone. The marks are to be compared as a whole. The Red Bird Mark comprises angry eyes, accented eyebrows, a sharp beak and a red bird. On this basis it is said that, the Angry Bite Mark shows a creature biting angrily into something, the “ANGRY BITE” words, and the “D-Jack” words at the corner. These elements are absent from the Red Bird Mark, and create dissimilarity.

  118. [83] The Angry Bite Mark is a composite mark. It comprises a number of elements. Some of these have been described in the previous paragraph. The “creature” is depicted face-on with a mouth biting down on an object. The phrase “ANGRY BITE” appears in the middle of the face. The word “D-Jack” in smaller lettering appears on a banner to the left, just above the creature’s right eye. A small (indeed tiny) mark (said to represent feathers) also appears next to the letter “B”.

  119. [84] On visual similarity, I affirm the PAR’s decision, and do not find any material error of fact or law in her reasoning or conclusion. Indeed, while the angry eyes and eye brows may (on one view) be the dominant and distinctive components of both marks, the PAR has given due weight to this fact. She was correct in stating that notwithstanding the similarity between the alleged dominant components of the marks, the marks still are to be assessed as a whole. I do not find that she erred in coming to the factual conclusion that the Red Bird Mark and the Angry Bite Mark are visually dissimilar on the whole. Indeed, the actual depiction of the eyes and eyebrows in the Red Bird Mark and the Angry Bite Mark whilst similar, are not identical.

  120. [85] On aural similarity, neither side disputes the PAR’s finding that there can be no aural similarity here given the Red Bird Mark has no aural component. I find no reason to disturb this finding.

  121. [86] Finally, on conceptual similarity, the PAR found the Red Bird Mark and the Angry Bite Mark to be dissimilar. She noted that what is important is “the impression of the average consumer” (see [65] of the PAR’s GD). The Red Bird Mark conveys a bird with angry eyes; the Angry Bite Mark conveys an idea of a creature biting angrily. Only the emotion of anger is common.

  122. [87] The Opponent submits that not only is the “anger emotion” common, the Angry Bite Mark also has elements of a bird. It points to the feather at the bottom left corner of the Angry Bite Mark, and the mouth of the creature which is said to look like a beak. In fact, the Opponent adduced evidence that the Malaysian trade marks registry described the creature in the Angry Bite Mark as a “bird”.

  123. [88] Having reviewed the PAR’s reasoning on conceptual similarity, I find no reason to overturn her finding. Notwithstanding the apparent position taken by the Malaysian trade marks registry, I do not find that the Angry Bite Mark conveys the impression or idea of a bird. In particular, I do not see how the mouth of the creature looks like a bird’s beak at all. Whilst the Opponent claims that there is a feather device shown in the Angry Bite Mark (next to the letter “B” in “BITE”), I am unable to draw any conclusion as to what was being depicted. The device is very small and in my assessment, makes little to no visual or conceptual impact at all even though there is a similar device on the Red Bird Mark.

  124. [89] In summary, having affirmed the PAR’s finding that there is no visual, aural or conceptual similarity between the Red Bird Mark and the Angry Bite Mark, I naturally also affirm her overall conclusion that the abovementioned marks are dissimilar on the whole. To be clear, even accepting that both marks share “anger” as a common emotion, I find that the marks on the whole are not similar.

  125. Comparison between the Angry Birds Word Mark and the Angry Bite Mark

  126. [90] The PAR found that while there is aural similarity between the Angry Birds Word Mark and the Angry Bite Mark, the two marks were visually and conceptually dissimilar.

  127. [91] Visually, the PAR found that similarity can only be found in the letters “ANGRY BI”. The other elements of the Angry Bite Mark clearly distinguish it from the Angry Birds Word Mark. The Opponent merely repeated the observation that the letters “ANGRY BI” are similar. It did not address the PAR’s observation that there are other elements to the Angry Bite Mark (namely, the creature) that distinguishes it from the Angry Birds Word Mark. I agree with the PAR’s observation that the pictorial elements of the Angry Bite Mark create visual dissimilarity, and thus see no reason to disturb the PAR’s decision that the Angry Bite Mark and the Angry Birds Word Mark are visually dissimilar.

  128. [92] In terms of aural similarity, the PAR found that there was some aural similarity. She reasoned that the competing marks have more syllables in common than not. The Opponent added that the two marks have the same number of syllables (ie three). The Applicant however submits that the “I” in the word “BITE” sounds different from the “I” in the word “BIRDS”, creating a degree of aural dissimilarity. In my view, the PAR’s finding that there is some aural similarity is consistent with both the Applicant’s and Opponent’s submissions. Indeed, we may all agree that while the “ANGRY” component in both marks are aurally identical, “BIRD” and “BITE” contain some aural dissimilarity.

  129. [93] Finally, on conceptual similarity, the PAR found that the only similarity is the emotion of anger conveyed. The concept of a creature angrily biting something is missing from the Angry Birds Word Mark. The Opponent however repeats its submissions that the concept of anger is a significant similarity, and there are hints of a bird in the Angry Bite Mark. In this regard, I repeat my findings above on the ideas conveyed by the Angry Bite Mark. I see no reason to disturb the PAR’s factual finding that the marks are conceptually dissimilar on the whole.

  130. [94] Given that there is some aural similarity, I will set out in more detail why I agree with the PAR that the Angry Bird Word Mark and Angry Bite Mark are dissimilar on the whole, bearing in mind the need for the court to weigh up the relative importance of similarities/dissimilarities in order to decide whether the marks are similar on the whole.

  131. [95] In this regard, following Hai Tong Co Pte Ltd v Ventree Singapore Pte Ltd and another and another appeal [2013] 2 SLR 941 (“Hai Tong”) at [40], the PAR (at [68] of the PAR’s GD) took into account the fact that the average consumer is one who would exercise care and a measure of good sense in her purchases, rather than an unthinking person in a hurry. She found that the snack products were bought by consumers of all ages, not just the young. As such, she concluded that to the average consumer as described above, the marks would be more dissimilar than similar on the whole. The Opponent, however, contends that children are the target, and indeed, the main, audience of the snack food products.

  132. [96] At this juncture, a question arises as to the relevance of the nature of consumers purchasing the parties’ products at the similarity of marks stage. As the court in Staywell at [20] clarified, at the similarity of marks stage, we are concerned with a mark-for-mark comparison, rather than the “effect of objective similarity between the marks”. It may be argued that the actual type of consumers purchasing the product is irrelevant to the assessment of the objective similarity between two marks. It is relevant only in assessing the confusing effect of that similarity (where, for example, children are presumably more likely to be confused than adults by a certain objective similarity).

  133. [97] We should first consider the position taken by the Court of Appeal in Hai Tong. The Court of Appeal stated at [40(c)] that:

  134. [98] Prima facie, the PAR’s approach seems to be supported by the passage in Hai Tong cited above, which suggests that similarity is to be assessed from the perspective of the “average consumer”. However, the question is whether the Court of Appeal in Hai Tong meant the viewpoint of the average consumer who actually purchases the parties’ products, or whether the “average consumer” at the similarity stage is a notional “average consumer”, having no regard to the actual goods or services in question. The PAR seemed to have adopted the former interpretation in considering whether children were the main consumers of the Applicant’s and the Opponent’s products.

  135. [99] In my view, the latter interpretation of the above passage in Hai Tong is preferable. In Hai Tong at [85(c)(iii)] and [87], the court classified the “particular characteristics of the relevant segment of the public” (referring to the likely purchasers of the parties’ products) as an extraneous factor which could be taken into account when assessing “likelihood of confusion”, but not the similarity of marks.

  136. [100] This is the position taken in Staywell at [96(b)] also, where “the likely characteristics of the relevant consumers and whether they would or would not tend to apply care or have specialist knowledge in making the purchase” was considered an extraneous factor relevant at the “likelihood of confusion stage”. The court in Staywell held (at [20]) that the “issue of resemblance between competing marks is distinct from the question of the effect of such resemblance” [emphasis in original], and that it was impermissible to consider extraneous factors at the similarity of marks stage. This did not mean that the court ignored the reality that the effect of aspects of similarity might depend on circumstances such as the nature of the goods and the perception of customers (Staywell at [20]). However, the Court of Appeal held that these considerations were better reserved for the confusion stage of the inquiry, where the court was asked to assess the effect of the objective similarity between the marks.

  137. [101] Applying Staywell and Hai Tong, it follows that a court in deciding whether the applicant’s mark is similar to the opponent’s mark (the first stage of the three step test) must make the assessment on an objective basis. The points of similarity and dissimilarity must be assessed in relation to the marks as a whole. This is a matter of impression for the judge. While the judge should put himself/herself into the position of a reasonable or notional average consumer (as opposed to an unthinking person in a hurry), the court should not, following Staywell and Hai Tong, delve into the detailed characteristics and perceptions of the actual consumers of the parties’ products at the similarity of marks stage. Thus, I find that the PAR’s conclusion that the Angry Bite Mark and the Angry Birds Word Mark are dissimilar (on the whole) is fully supportable.

  138. [102] To sum up, having considered the visual, aural and conceptual similarities and dissimilarities between the marks, and having considered the marks as a whole, I do not think that the PAR erred in finding that the Angry Bite Mark and Angry Birds Word Mark are dissimilar. While the Opponent did submit that the Court should place some weight on the fact that the designer of the Angry Bite Mark was inspired by the Opponent Marks, the argument does not carry the Opponent any further. Trade mark law does not have as one of its goals the objective of stopping competition simply on the basis that the competitor’s trade mark was inspired by the trade mark proprietor’s trade mark, branding or marketing. In the present case, the Opponent must still prove that the relevant marks are similar. The fact that the Angry Bite Mark was inspired by the Opponent Marks may give rise to a heightened possibility of similarity, but in no way establishes it.

  139. [103] In conclusion, I affirm the PAR’s findings that the Angry Bite Mark is not similar to either the Angry Birds Word Mark, or the Red Bird Mark. While this is sufficient to dispose of any opposition on the ground of s 8(2)(b) of the TMA, I shall nevertheless proceed to consider the “likelihood of confusion” in case I am wrong on similarity of marks.

  140. Likelihood of confusion

  141. [104] As discussed above, at the “likelihood of confusion stage”, the court is broadly concerned with the “effect of objective similarity between the marks, on the perception of consumers” [emphasis in original]: Staywell at [20]. In Hai Tong at [74], the Court of Appeal stated that there are “at least two specific aspects to the element of confusion”. The first is “mistaking one mark for another”, and the second is where “the relevant segment of the public may well perceive that the contesting marks are different, but yet remain confused as to the origin which each mark signifies”. The court in Hai Tong at [75] went on to distinguish “mere association” from a “likelihood of confusion”. Specifically, “it is not sufficient that the relevant segment of the public would recognise or recollect similarities between the contesting marks if there is no likelihood of confusion as to origin ensuing”.

  142. [105] In assessing the likelihood of confusion, the PAR (at [73]–[74] of the PAR’s GD) considered it relevant that the Opponent Marks were used on products together, but stated that “this only renders the marks somewhat more similar”. She found that the overall impression given by the Angry Bite Mark (that of a creature angrily biting) is different from the overall impression given by the Opponent Marks. Further, the PAR weighed the following factors in concluding that there was no likelihood of confusion:

  143. (a) The reputation of the ANGRY BIRD brand is likely to reduce the likelihood of confusion.

  144. (b) Given the snacks are self-serve items on super market shelves, the visual and conceptual aspects figure more prominently than the aural aspects at the point of selection and sale (recall that she found there was some degree of aural similarity, but no visual or conceptual similarity).

  145. (c) The purchasing public is likely to exercise a greater degree of care in selecting the snack products given these are to be ingested.

  146. (d) The low price of the snack products is outweighed by the other factors pointing against a likelihood of confusion.

  147. [106] On appeal, the Opponent submits as follows:

  148. (a) The likelihood of confusion is enhanced by the similarity in colours used on the Applicant’s packaging, as well as the depiction of a catapult at the bottom right corner of the Applicant’s packaging (given that a catapult is a key feature of the Opponent’s Angry Bird mobile phone game). Pictures of crisps packets in identical shades of yellow, green, red etc were presented to the court.

  149. (b) The PAR erred in finding that the aural similarity is less significant than visual and conceptual similarities for self-serve items such as those sold in supermarkets.

  150. (c) The PAR erred in finding that the reputation of the Opponent’s brand detracts from the likelihood of confusion. Given the imperfect recollection of consumers, the fact that the Opponent’s brand is entrenched in the minds of its consumers furthers the likelihood of confusion.

  151. (d) The PAR erred in finding that a degree of care would be taken in choosing the snack products because they are to be ingested. On the contrary, food safety does not weigh heavily on the minds of consumers in Singapore when buying snacks in a supermarket.

  152. (e) The snack products are likely to be purchased by children and adolescents, who are unlikely to apply care or fastidiousness in the selection of snacks.

  153. [107] In response, the Applicant repeats the PAR’s reasoning, and additionally submits the following:

  154. (a) Following Staywell, extraneous factors like packaging or other “superficial marketing choices” are irrelevant in opposition proceedings, even at the “likelihood of confusion” stage.

  155. (b) Consumers who purchase goods from the supermarket are ordinary sensible members of the public who are not easily deceived or hoodwinked.

  156. [108] Based on the above, the following issues on the “likelihood of confusion” inquiry under s 8(2)(b) of the TMA arise for consideration:

  157. (a) Can extraneous factors such as the colour of the Applicant’s packaging and the catapult on the Applicant’s crisp pack be taken into account?

  158. (b) Does the reputation of the Opponent’s brand enhance or detract from a likelihood of confusion?

  159. (c) What is the relevance of the fact that the Applicant’s and Opponent’s goods are ingestible self-serve items sold in supermarkets?

  160. (d) Who are the likely consumers of the Applicant’s and Opponent’s goods?

  161. Can extraneous factors such as the colour of the Applicant’s packaging and the catapult on the Applicant’s crisp pack be taken into account?

  162. [109] Extraneous factors may be taken into account in opposition proceedings at the likelihood of confusion stage. However, this general position is subject to important qualifications: Staywell at [77] and [83]. The Court of Appeal laid down the following guidelines in Staywell at [95]:

  163. [110] The Opponent rather ingeniously argues that the limits or qualifications placed on extraneous factors apply only when an applicant seeks to distinguish its goods on the basis of extraneous factors. It submits that Staywell only refers to certain extraneous factors being impermissible for the purpose of differentiating goods that are otherwise confusingly similar (for example, see Staywell at [86], [89], and [90]). It does not preclude an opponent using extraneous factors to show an enhanced likelihood of confusion. The Opponent submits that this is in line with the principle that an existing proprietor has monopoly rights over its trade mark which deserves full protection.

  164. [111] Reading Staywell as a whole, I find it hard to agree with the Opponent’s position on extraneous factors. It is true that the Court of Appeal in Staywell focused on the impermissibility of extraneous factors being used to negate a likelihood of confusion. However, this is readily explained by the fact that the Court of Appeal was asked to consider that very question.

  165. [112] A closer look at Staywell would reveal that the court did not intend to limit its remarks on extraneous factors to the negating of a likelihood of confusion.

  166. [113] At [86] of Staywell, the Court of Appeal stated:

  167. [114] Given that confusion must stem from the similarity of marks and goods, rather than from extraneous factors, the Court of Appeal’s remarks on the “impermissible extraneous factors” clearly apply to both negating and enhancing the likelihood of confusion. In fact, based on the passage cited above, the impermissibility of admitting certain extraneous factors to negate the likelihood of confusion seems to be premised on the impermissibility of basing a finding of confusion on those very same factors.

  168. [115] Indeed, even looking at the matter from the point of view of legal principles and policy, this conclusion makes sense. Registered trade mark law gives proprietors proprietary rights over trade marks only; it does not give them rights over certain types of packaging, or marketing techniques. It follows that proprietors are protected only to the extent that confusion (or the likelihood of confusion) arises from a use of a mark similar to the registered trade marks in respect of similar or identical goods or services.

  169. [116] In this case, the Opponent does not have an exclusive right to use the coloured packaging for its crisps. Whilst the evidence on this was not explored in detail, I note that the same or similar coloured packaging is in fact used by many crisps manufacturers, not just the Applicant and the Opponent. The Opponent does not have any registered trade mark over the colour of the packaging. Also, since the Opponent is not claiming a trade mark over the catapult, any confusion caused by it (or likely to be caused by it) must be irrelevant.

  170. [117] I thus find that factors to do with the Applicant’s packaging are superficial marketing choices and hence impermissible extraneous factors which may not be considered by this court in determining the likelihood of confusion arising from the alleged similarities between the marks.

  171. Does the reputation of the Opponent’s brand enhance or detract from a likelihood of confusion?

  172. [118] Under the registered trade mark system, new marks which have not been used previously (whether at all, or in respect of a particular category of goods and services) may be registered. While it is necessary for the applicant to have a bona fide intention to use the trade mark (s 5(2)(e)(ii) of the TMA), it is not necessary to show prior use before applying to register a trade mark. It follows that a new trade mark may be properly registered even though it has not yet acquired any reputation. The question that arises in the present case is whether the court can take into account the reputation generated by the use of the mark as a factor in determining whether there is a likelihood of confusion. Nothing in the question posed suggests that a registered trade mark must have a reputation in order for there to be a likelihood of confusion. The issue raised is really the reverse: where reputation in the opponent’s mark is established, is this relevant in determining the likelihood of confusion?

  173. [119] The PAR was of the view that, on the facts, where a mark or brand becomes so well known that it is entrenched in the minds of consumers, this could have the effect of reducing the likelihood of confusion. The PAR cited [96(a)] of Staywell in support of her conclusion. There, the Court of Appeal cites Polo at [34] for the proposition that the “reputation of the marks” is a factor that relates to the impact of marks-similarity on consumer perception. At [34] of Polo, the Court of Appeal states “the greater the exposure and use of a particular registered mark, the greater its reputation is likely to be, and therefore the greater the protection likely to be afforded to it”. None of these authorities seem to lend support to the PAR’s conclusion that where a mark is well known, that reduces the likelihood of confusion. On the contrary, [34] of Polo seems to suggest that where a mark is well known, it ought to be afforded more protection, and hence, it should be easier to find a likelihood of confusion.

  174. [120] However, the Applicant did cite some authorities which more directly support the PAR’s position. At [64] of McDonalds, the Court of Appeal stated “the very success of the appellant, which is inseparable from its logo, is also the very reason why confusion is unlikely”.

  175. [121] Indeed, I note that there is some Australian authority to that effect. In Mars Australia Pty Ltd v Delfi Chocolate Manufacturing SA [2014] ATMO 113 (Mars Australia) at [28], the court stated that if the reputation in the “MALTESERS” trade marks were taken into consideration (coupled with the recognition of the Australian consumer’s imperfect recollection), this would tend to reduce the likelihood of deception or confusion between it and the applicant trade mark, “MALTITOS”. Also, the learned authors of Mark Davidson & Ian Horak, Shanahan’s Australian Law of Trade Marks and Passing Off (Lawbook Co, 5th Ed, 2012) at para 50.2410 state “[i]n some cases the particularly well known nature of the opponent’s mark has in fact told against its argument that confusion is likely”.

  176. [122] That said, the relevant Australian statutory provisions are not the same as those in Singapore. The observation in Mars Australia was made in respect of s 44 of Australia’s Trade Marks Act 1995 (Cth) (“Australian TMA”), and opposition was based on the claim that the applicant’s trade mark was “deceptively similar”. Whilst under the Australian TMA, a mark is deceptively similar if it is likely to deceive or cause confusion, the difference in statutory language does in my view affect the persuasive weight of Mars Australia in Singapore.

  177. [123] It is also worth mentioning that in Mars Australia, whilst the objection under s 44 of the Australian TMA failed, the opposition under s 60 succeeded. Section 60 of the Australian TMA reads as follows:

  178. [124] The court in Mars Australia held that opposition to registration succeeded under s 60 of the Australian TMA because an assessment of the reputation of the trade mark went beyond mere examination of turnover of goods, and included the “credit, image and values projected” and the “esteem” component of the reputation (Mars Australia at [40]). On this basis, the trade mark officer found at [51] that deception or confusion would arise from the demonstrated reputation in the opponent’s trade mark, taking into account the extent of use and the way in which the reputation was established in Australia. Even though the trade mark officer had earlier found that the applicant and opponent’s marks were not deceptively similar, there was some similarity (Mars Australia at [48]). In an opposition under s 60, what was in issue was whether the use of the applicant’s mark would cause deception or confusion. On this basis, the finding at [50] was that the demonstrated reputation in the opponent’s mark would “reduce rather than reinforced the effect of the differences”. I note that the provisions in Singapore on well known earlier trade marks (considered below) are not the same as the Australian provisions.

  179. [125] Returning to s 8(2)(b), I do not find it necessary to review the jurisprudence on this matter in any more detail than I already have (especially as Mars Australia was not cited in argument). In my view, it is plain that there can be cases where the reputation of an opponent mark reduces the likelihood of confusion as a matter of fact. The more a mark is etched in the minds of consumers, and the more familiar consumers are with the details of a mark, the more discerning they may be in some cases, especially where on the facts, the customers will likely be able to detect differences at a glance. But of course, it all depends on the facts. I do not think case law goes so far as to establish a rule of law that the greater the reputation of a mark, the less the likelihood of confusion. I do not think such a rigid rule is conducive to guiding what is fundamentally a fact specific inquiry either. Whether there is a likelihood of confusion is a question of fact for the court, taking into account, inter alia, the reputation of the marks (as stated in Staywell at [96(a)]). The exact effect the reputation of a mark has may differ on the specific facts of a case. As the learned author of Ng–Loy Wee Loon states at para 21.5.29(c), whilst the reputation of an earlier trade mark is relevant, it is not “the determinant factor”. The approach is not “reputation-therefore-confusion” (or the reverse).

  180. [126] In this case, I do not think that the PAR erred in her factual conclusion that “the Opponent’s reputation in the current case is likely to have an effect contrary to a likelihood of confusion”. In my view, the reputation and association between the “angry bird” character and the Opponent’s Angry Bird brand renders the absence of a clear bird image in the Angry Bite Mark a factor that tends to reduce the likelihood of confusion.

  181. What is the relevance of the fact that the Applicant’s and Opponent’s goods are ingestible self-serve items sold in supermarkets?

  182. [127] First, the PAR found that the ingestible nature of the Applicant’s and the Opponent’s snack food products made it more likely that consumers will pay greater attention to the labels and trade marks before buying the goods. On the facts and evidence, I do not agree that this proposition has been made out. In my view, the “ingestible” nature of the snack food products is a neutral factor in relation to the overall assessment of a likelihood of confusion.

  183. [128] Second, the PAR found that on the facts, the visual and conceptual similarities were more important than the aural similarity because the goods are self-serve items and sold in supermarkets. While I do not disagree that the visual and conceptual similarities play a significant role when consumers scan the supermarket shelves and pick out a packet of crisps, it does not follow that aural similarity is irrelevant. The more difficult question is the relative weight to be ascribed to aural similarity. In Singapore, the material time at which confusion must exist is at the time of purchase: Ng-Loy Wee Loon at para 21.5.26, citing Staywell (although questions may arise on post sale confusion). This explains the learned author’s comment at para 21.5.29(d) that the normal way under which the average consumer would purchase the goods or services is a relevant factor in assessing the likelihood of confusion. Where the similarity lies mainly in their aural aspect, the risk of confusion would be greater if the goods are normally purchased by placing verbal orders. Given the snack products in the present case are not purchased by placing verbal orders, I find no reason to disturb the PAR’s assessment that aural similarity is less important on the facts of the present case.

  184. Who are the likely consumers of the Applicant’s and Opponent’s goods?

  185. [129] The PAR disagreed with the Opponent’s submission that children were the primary consumers of the Applicant’s and Opponent’s goods. No evidence was given on this issue. The court is thus left to come to a decision based on its general impression of who the likely purchasers of snack food products are.

  186. [130] On that basis, there is no reason to disturb the PAR’s conclusion that people who buy snacks are not merely, or even primarily, children. After all, people of all ages purchase snacks whether for themselves or others. It is likely that in many cases the parents or adults will be the actual purchasers (even if crisps are meant for consumption by the child).

  187. Conclusion on the likelihood of confusion

  188. [131] I find that on the whole, the PAR is justified in concluding that there is no likelihood of confusion. In coming to this conclusion, I have ignored the alleged similarities in packaging between the parties’ goods.

  189. [132] Given my conclusion that the marks are not similar, and that there is, in any event, no likelihood of confusion, I find that the opposition on the ground of s 8(2)(b) of the TMA must fail.

  190. Issue 2: s 8(4)(b)(i) read with s8(4)(a) of the TMA

  191. [133] I set out the relevant parts of s8(4) of the TMA:

  192. Are the Opponent Marks “well known in Singapore”?

The judgment continues past this page. Read the rest at eLitigation.