Societe Des Produits Nestlé SA and another v Petra Foods Ltd and another

1. Chan Seng Onn J:

2. [1]

   [1] This case deals with claims in trade mark and copyright infringement. A single thread runs through the numerous and varied issues that have been raised by the parties, and that is the question of when one trader may legitimately restrain another trader from marketing goods of a similar shape to its products by asserting intellectual property protection in the shapes of the products themselves. The shapes in question are those comprising the familiar Kit Kat chocolate coated wafer bars, which have been the subject of much litigation throughout the world.

3. The parties

4. [2]

   [2] The plaintiffs belong to the Nestlé group of companies. The first plaintiff, Societe Des Produits Nestlé SA, is a public company incorporated in Switzerland. The second plaintiff, Nestle Singapore (Pte) Ltd, is a company incorporated in Singapore. The plaintiffs are the proprietors of the chocolate covered wafer products under the KIT KAT trade mark.

5. [3]

   [3] The defendants belong to the Petra Foods group of companies. The first defendant, Petra Foods Limited, is a public company incorporated in Singapore. It is the parent company of the second defendant, Delfi Singapore Pte Ltd, as well as PT Perusahaan Industri Ceres (“PT Ceres”), a company incorporated in Indonesia, and Delfi Chocolate Manufacturing S.A (“Delfi SA”), a company incorporated in Switzerland.

6. [4]

   [4] The second defendant is the importer and distributor of a chocolate product under the trade mark “Take-It” or “Delfi Take-It”. The “Take-It” product is manufactured by PT Ceres under licence from Delfi SA.

7. The claims and counterclaim

8. [5]

   [5] The first plaintiff is the registered owner of the following trade marks in Singapore:

9. [6]

   [6] I will refer to the Two Fingers Shape and the Four Fingers Shape collectively as “the Registered Shapes”.

10. [7]

    [7] The first plaintiff also claims to be the owner of the following unregistered trade mark (“the Two Fingers Composite Mark”):

11. [8]

    [8] Finally, the first plaintiff also asserts that it is the owner of the following “artistic work” which it claims was created in or about the late 1990s (“the Artistic Work”):

12. [9]

    [9] The plaintiffs do not in their statement of claim specify what type of “artistic work” the Artistic Work is. I assume that they mean that the Artistic Work is a species of drawing or painting. The defendants do not dispute that it is an “artistic work”.

13. [10]

    [10] The plaintiffs allege that the shape and packaging of the defendants’ Take-It products infringe their intellectual property rights. I set out in the table below the allegedly infringing goods:

14. [11]

    [11] The second defendant does not dispute that it imported and sold the 2-Fingers Take-It in Singapore from at least August 2010 and the 4-Fingers Take-It from at least April 2012.

15. [12]

    [12] The plaintiffs’ claims may be summarised as follows:

16. (a) Trade mark infringement (product)

17. (i) The 2-Fingers Take-It infringes the Two Fingers Shape under ss 27(1) and/or 27(2)(b) of the Trade Marks Act (Cap 332, 2005 Rev Ed) (“TMA”).

18. (ii) The 4-Fingers Take-It infringes the Four Fingers Shape under ss 27(1) and/or 27(2)(b) of the TMA.

19. (b) Trade mark infringement (packaging)

20. (i) The Representation of the 2-Fingers Take-It infringes the Two Fingers Shape under s 27(2)(b) of the TMA.

21. (ii) The Representation of the 4-Fingers Take-It infringes the Four Fingers Shape under s 27(2)(b) of the TMA.

22. (c) Well-known trade marks (product)

23. (i) The Two Fingers Shape is well known and/or well known to the public at large in Singapore, and the defendants should be restrained by injunction from using the 2-Fingers Take-It under ss 55(2), 55(3)(a) and/or 55(3)(b) of the TMA.

24. (ii) The Four Fingers Shape is well known and/or well known to the public at large in Singapore, and the defendants should be restrained by injunction from using the 4-Fingers Take-It under ss 55(2), 55(3)(a) and/or 55(3)(b) of the TMA.

25. (d) Well known trade marks (packaging)

26. (i) The Two Fingers Shape is well known and/or well known to the public at large in Singapore, and the defendants should be restrained by injunction from using the Representation of the 2-Fingers Take-It under ss 55(2), 55(3)(a) and/or 55(3)(b) of the TMA.

27. (ii) The Four Fingers Shape is well known and/or well known to the public at large in Singapore, and that the defendants should be restrained by injunction from using the Representation of the 4-Fingers Take-It under ss 55(2), 55(3)(a) and/or 55(3)(b) of the TMA.

28. (iii) The Two Fingers Composite Mark is well known and/or well known to the public at large in Singapore and the defendants should be restrained by injunction from using the Representation of the 2-Fingers Take-It under ss 55(2), 55(3)(a) and/or 55(3)(b) of the TMA.

29. (e) Copyright infringement (packaging)

30. (i) The Representation of the 2-Fingers Take-It infringes the Artistic Work under ss 32 and/or 33 of the Copyright Act (Cap 63, 2006 Rev Ed) (“Copyright Act”).

31. [13]

    [13] The defendants’ counterclaim may be summarised as follows:

32. (a) The Two Fingers Shape and Four Fingers Shape ought to be invalidated and/or revoked on the following grounds that the Two Fingers Shape and the Four Fingers Shape:

33. (i) were registered in breach of s 7(1)(a) of the TMA in that they are not trade marks within the meaning of s 2(1) of the TMA, being not capable of distinguishing the goods for which they were registered;

34. (ii) were registered in breach of s 7(1)(b) of the TM in that they are devoid of any distinctive character;

35. (iii) should be revoked pursuant to s 22(1)(a) of the TMA as the marks were not put to genuine use in the course of trade in Singapore within five years following the date of completion of registration;

36. (iv) should be revoked pursuant to s 22(1)(b) of the TMA as use of the marks was suspended for an uninterrupted period of five years;

37. (v) were registered in breach of s 7(3)(a) of the TMA in that each of them consists exclusively of a shape which results from the nature of the goods themselves;

38. (vi) were registered in breach of s 7(3)(b) of the TMA in that each of them consists exclusively of a shape which is necessary to obtain a technical result; and

39. (vii) were registered in breach of s 7(3)(c) of the TMA in that each of them consists exclusively of a shape which gives substantial value to goods.

40. (b) The plaintiffs’ threats of legal proceedings for infringement of a registered trade mark were groundless threats of infringement proceedings under s 35(2) of the TMA.

41. Background

42. [14]

    [14] The Kit Kat products which bear the Registered Shapes are essentially moulded chocolate wafer confectionaries. According to Mr Clive Barnes (“Mr Barnes”), the head of communication and information services for Nestle’s global research and development activities, the four fingers Kit Kat product was originally launched as Rowntrees Chocolate Crisp in 1935, which was rebranded as Kit Kat in 1937. It bore the Four Fingers Shape right from the beginning and it remains unchanged till today. The two fingers version was subsequently launched in 1963. Undoubtedly, the first plaintiff has been marketing the Kit Kat products for a very long time.

43. [15]

    [15] Mr Chuang Tiong Liep (“Mr Chuang”), who is an executive director of the first defendant and a director of the second defendant, said that the first defendant and its subsidiaries (including the second defendant) also have a long history of selling moulded chocolate wafers and biscuits. He said that one of the subsidiaries, PT Ceres, first manufactured a moulded chocolate wafer comprising of a single wafer coated with chocolate under the trade mark PATSY in Indonesia as early as 1951.

44. [16]

    [16] Subsequently, PT Ceres decided to develop a cheaper and smaller confectionary to cater to the Indonesian market. As a result, in around 1986 to 1987, PT Ceres manufactured and sold moulded chocolate wafers comprising of two or three “sticks” of wafers joined at the base with chocolate. These were sold in Indonesia, Singapore and Dubai under the WINDMOLEN mark. The Windmolen wafers were made using compound chocolate, which is different from real chocolate in that the former uses vegetable fat in place of cocoa butter. These Windmolen wafers were eventually phased out as they did not do well in the Indonesian market.

45. [17]

    [17] According to Mr Chuang, the experience in producing, distributing and selling the Windmolen wafers eventually led to the development of the Take-It products, which include the 2-Fingers Take-It and 4-Fingers Take-It moulded chocolate wafer products. They are made of real chocolate, as opposed to compound chocolate.

46. [18]

    [18] The plaintiffs have disputed Mr Chuang’s rendition of the defendants’ historical background. They do not believe the Windmolen products ever existed. They say that even if they existed, the defendants still chose to deliberately model the Take-It products after the plaintiffs’ products, and ride on their coat tails, even though they knew that the plaintiffs had registered the Registered Shapes.

47. [19]

    [19] Considering the popularity of the Kit Kat products, it is unsurprising if the defendants did, despite their denials, deliberately emulate the plaintiffs’ products. But this puts the cart before the horse. Even if the defendants did copy, it is not a wrong unless such copying impinges the plaintiffs’ legal rights. Imitation is no less valid a business strategy than innovation. There is nothing unconscionable about competition.

48. Preliminary issue: Whether the second plaintiff has any locus standi or cause of action in these proceedings

49. [20]

    [20] The defendants submit that the second plaintiff has no locus standi or cause of action. This is because all the trade mark rights are pleaded in the name of the first plaintiff. It is also the first plaintiff who is said to be the owner of the copyright in the Artistic Work. Accordingly, the second plaintiff should be struck out for having no cause of action.

50. [21]

    [21] In response, the plaintiffs say that this is not the defendants’ pleaded case and the defendants are bound by their pleadings. They also note that prior to the hearing of the main trial, the defendants had attempted unsuccessfully to strike out the plaintiffs’ copyright claim and even then they did not raise this point. They say this is litigation by ambush and the plaintiffs are irreversibly prejudiced.

51. [22]

    [22] The plaintiffs also refer to s 44 of the TMA:

52. [23]

    [23] The plaintiffs merely say that it is clear from sub-ss 44(3) and (5) of the TMA that the second plaintiff has locus standi to sue. I take it that the plaintiffs are essentially saying that the second plaintiff is a licensee, at least as far as the trade mark claims are concerned.

53. [24]

    [24] The basis for the defendants’ assertion is the plaintiffs’ statement of claim. They have every opportunity to examine the plaintiffs’ pleadings from the beginning. They could and should have raised this issue much earlier. To only do so after trial has concluded seriously prejudices the plaintiffs. I therefore agree with the plaintiffs that the second plaintiff should not be struck out.

54. Whether the Registered Shapes are invalid and/or revocable

55. [25]

    [25] I now turn to the substance of the case. I will begin by addressing the defendants’ counterclaim to invalidate and/or revoke the Registered Shapes. If the defendants’ succeed, the plaintiffs’ claims in registered trade mark infringement will necessarily fail.

56. Whether the Registered Shapes consist exclusively of the shape of goods which is necessary to obtain a technical result

57. The key legal principles

58. [26]

    [26] The defendants’ contend that the registrations of the Two Fingers Shape and Four Fingers Shape are invalidated by s 7(3) read with s 23(1) of the TMA. If the Registered Shapes fall within any of the prohibitions under s 7(3) of the TMA, the registrations will be invalid, regardless of how distinctive the Registered Shapes are. Sections 7(3) and 23(1) of the TMA state:

59. [27]

    [27] Of the three grounds set out in s 7(3) of the TMA, I begin with the bar to registration under s 7(3)(b) of the TMA, as this is the focus of the parties’ submissions. I will first summarise a number of relevant propositions of law which are set out in Lego Juris A/S v Office for Harmonisation in the Internal Market (Case C-48/09 P) [2010] ETMR 63 (“Lego Juris”), where the Court of Justice of the European Union (“CJEU”) considered Art 7(1) and in particular Art 7(1)(e) of the Council Regulation (EC) No 40/94 of 20 December 1993 on the Community trade mark (OJ 1994 L 11, p 1) (“Regulation No 40/94”), which is in pari materia with s 7(3)(b) of the TMA. The reasoning in Lego Juris largely follows the earlier case of Philips Electronics NV v Remington Consumer Products Ltd (Case C-299/99) [2003] RPC 2 (“Philips (ECJ)”). For present purposes, I will refer to the bar to registration under Art 7(1)(e) of Regulation No 40/94 and s 7(3)(b) of the TMA, as well as other equivalent statutory provisions found in various jurisdictions, as the “Technical Result Exception”.

60. [28]

    [28] I accept the following propositions of law as being well-founded and applicable to s 7(3)(b) of the TMA:

61. (a) The grounds for refusal under s 7(3) of the TMA must be interpreted in light of the public interest underlying them. The interest underlying the Technical Result Exception is to “prevent trade mark law granting an undertaking a monopoly [which is potentially of unlimited duration] on technical solutions or functional characteristics of a product”. Technical solutions are capable of protection for only a limited period, so that subsequently they may be freely used by all economic operators (Lego Juris at [43] to [46]).

62. (b) No amount of acquired distinctiveness through use can cure a trade mark which offends the Technical Result Exception, for “even if a shape of goods which is necessary to obtain a technical result has become distinctive in consequence of the use which has been made of it, it is prohibited from being registered as a trade mark” (Lego Juris at [47]).

63. (c) If all the essential characteristics of a three-dimensional sign are dictated by the technical solution to which that sign gives effect, the sign consists exclusively of the shape of goods which is necessary to obtain a technical result. However, a sign cannot be refused registration as a trade mark if “the shape of the goods at issue incorporates a major non-functional element, such as a decorative or imaginative element which plays an important role in the shape” (Lego Juris at [52]).

64. (d) A shape may still be considered necessary to obtain the technical result even if the same technical result may be achieved by various solutions, such as by alternative shapes, for the reason that a trade mark may allow its proprietor to prevent other undertakings from not only using the same shape, but other similar shapes (Lego Juris at [54] to [58]). This is especially where the solution incorporated in the shape in question is the “technically preferable solution” (Lego Juris at [60]). Accordingly, the existence of other shapes which can achieve the same technical result does not exclude this ground for refusal for registration (Lego Juris at [83]).

65. [29]

    [29] The CJEU in Lego Juris also provided some guidance on how a court ought to proceed in determining the essential characteristics of a shape mark and whether the shape is exclusively functional:

66. (a) The expression “essential characteristics” must be understood as referring to the most important elements of a sign (Lego Juris at [69]).

67. (b) The identification of these essential characteristics must be carried out on a case-by-case basis, and in so determining, the court may either base its assessment directly on the overall impression produced by the sign, or first examine in turn each of the components of the sign concerned (Lego Juris at [70]).

68. (c) After identifying the sign’s essential characteristics, the court must ascertain whether they all perform the technical function of the goods at issue (Lego Juris at [72]).

69. (d) The presumed perception of the sign by the average customer is not a decisive element but is, at most, a relevant criterion of assessment for the court when it identifies the essential characteristics of the sign (Lego Juris at [76]).

70. (e) It is only necessary to assess whether the essential characteristics perform the technical function of the product concerned. The examination must be carried out by analysing the sign filed with a view to its registration as a trade mark and not signs consisting of other shapes of goods (Lego Juris at [84]).

71. [30]

    [30] In James Mellor et al, Kerly’s Law of Trade Marks and Trade Names (Sweet & Maxwell, 15th Ed, 2011) (“Kerly”), the learned authors argued that it was implicit in Philips (ECJ) and Lego Juris that “[t]he test is an objective one and does not depend upon the subjective intentions of the designer of the shape” (at para 8-161). For that proposition, Kerly cited Rimer J’s decision in Koninklijke Philips NV v Remington Consumer Products Ltd [2005] FSR 17 (“Philips 1 (HC)”), where the learned judge at [32] stated:

72. [31]

    [31] I agree with Rimer J’s observation.

73. Previous decisions on the Four Fingers Shape

74. [32]

    [32] The question of whether the registrations of the Four Fingers Shape and/or Two Fingers Shape are barred by the Technical Result Exception has been the subject of decisions before various tribunals, which I set out below:

75. (a) Four Fingers Shape

76. (i) Societe Des Produits Nestle SA v Cadbury UK Ltd (Case No O-257-13) (“Nestle v Cadbury (Opposition)”), a decision of the UK Intellectual Property Office in opposition proceedings against the first plaintiff’s application to register the Four Fingers Shape as a UK trade mark before Hearing Officer Allan James;

77. (ii) Societe des Produits Nestle SA v Cadbury Holdings Limited (Case R 513/2011-2) [2013] ETMR 25 (“Nestle v Cadbury (OHIM)”), a decision of the Office for Harmonization in the Internal Market (“OHIM”) Second Board of Appeal dealing with an appeal from cancellation proceedings in respect of the registration of the Four Fingers Shape as a Community Trade Mark;

78. (b) Two Fingers Shape

79. (i) The decision of the Paris Court of Appeal in Case 13/09001 (“Nestle (Paris)”), which was the first plaintiff’s appeal against the Director of the National Industrial Property Institute’s (“INPI”) decision to refuse the first plaintiff’s application to register the Two Fingers Mark;

80. (c) Four Fingers Shape and Two Fingers Shape

81. (i) Societe Des Produits Nestle SA and another v International Foodstuffs Co and others (Case Nos 33960/11, 71834/12 and 35379/13) (“Nestle (South Africa)”), a decision by the High Court of South Africa in trade mark infringement proceedings brought by the first plaintiff and Nestle South Africa (Pty) Ltd against the defendants for, inter alia, infringement of the Four Fingers Shape and Two Fingers Shape.

82. [33]

    [33] In Nestle v Cadbury (Opposition), the Hearing Officer considered the essential features of the Four Fingers Shape to be as follows (at [66]):

83. (a) the rectangular “slab” shape of the Four Fingers Shape including the relative proportions of length, width and depth (“Feature 1”);

84. (b) the presence, position and depth of the breaking grooves arranged along the length of the bar, which effectively divide the bar into detachable “fingers” (“Feature 2”); and

85. (c) the number of such grooves which, together with the width of the bar, determine the number of “fingers” (“Feature 3”).

86. [34]

    [34] The Hearing Officer made the following findings (Nestle v Cadbury (Opposition) at [81]):

87. [35]

    [35] The Hearing Officer also found that the Four Fingers Shape is not an aesthetic one, for the reasons above, as well as the further following reasons (Nestle v Cadbury (Opposition) at [85]):

88. [36]

    [36] Interestingly, the Hearing Officer concluded (at [89]) that the Four Fingers Shape is excluded from registration for all the goods applied for under class 30, except cakes and pastries, by s 3(2)(b) of the UK Trade Marks Act 1994 (c 26) (“the UK Act”), which is the UK equivalent of s 7(3)(b) of the TMA, when it would appear from his reasoning that he had only found that Feature 2 and Feature 3 fell under the Technical Result Exception, while Feature 1 was precluded from registration by s 3(2)(a) of the UK Act (the UK equivalent of s 7(3)(a) of the TMA), which bars from registration any shape consisting exclusively of the shape which results from the nature of the goods themselves.

89. [37]

    [37] Nestle v Cadbury (Opposition) was the subject of an appeal to the UK High Court before Arnold J in Societe Des Produits Nestle SA v Cadbury UK Ltd [2014] EWHC 16 (“Nestle v Cadbury (HC)”). The Hearing Officer’s factual findings were not challenged, although there were legal issues which caused Arnold J to refer three questions of law to the CJEU for clarification. As of the time of the writing of this judgment, the CJEU has yet to deliver its decision.

90. [38]

    [38] The OHIM Second Board of Appeal in Nestle v Cadbury (OHIM) reached a different conclusion from the Hearing Officer. In identifying the essential characteristics of the Four Fingers Shape, the Second Board of Appeal found that the sign consists of “four identical trapezoidal bars aligned together on a rectangular base” (at [104]). It concluded that this shape did not fall within the Technical Result Exception, for the following reasons (at [105] to [107]):

91. [39]

    [39] The Hearing Officer in Nestle v Cadbury (Opposition) had the opportunity to refer to the latter decision, but nevertheless differed from the latter’s conclusions. He distinguished the case on the following points (at [110]):

92. [40]

    [40] I now turn to the other two decisions. In Nestle (Paris), the Paris Court of Appeal disagreed with INPI that the Two Fingers Shape was excluded from registration by the Technical Result Exception. The translation of the decision is not entirely easy to understand. It accepted the first plaintiff’s description of the Two Fingers Shape, which was described as follows:

93. [41]

    [41] The Paris Court of Appeal found as follows:

94. [42]

    [42] With all respect to the Paris Court of Appeal, the reasoning is rather terse, and insofar as the finding that the “breakable character” is not exclusively functional because there are other ways to make the bar breakable, it is inconsistent with Lego Juris.

95. [43]

    [43] Finally, in Nestle (South Africa), J W Louw J – who was only dealing with affidavit evidence – also found that the Registered Shapes were not excluded by the Technical Result Exception. The main reason for this finding was based on a South African rule of evidence which meant that unless the evidence of Mr Barnes (who was also a witness in that case) was so far-fetched or untenable, the court was bound to accept his evidence that the various finger wafer shapes used for the Kit Kat product must originally have been designed and used for their visual and aesthetic impact and appeal and for enjoyment when eating, and that the finger wafer shapes are not, obvious ones as they are, in fact, technically difficult and inconvenient to manufacture (at [28]).

96. [44]

    [44] Nevertheless, Louw J went on to assume the correctness of the respondents’ evidence that the shape of the mould and, therefore, the ultimate trapezoidal shape, is entirely a technical requirement (ie, for the release of the chocolate bars from the mould). Even so he did not think the Technical Result Exception applied. He held at [30]:

97. [45]

    [45] I will note however that the respondents in Nestle (South Africa) relied only on the trapezoidal shape of the fingers for the purposes of the Technical Result Exception (at [26]).

98. [46]

    [46] One cannot help but notice that the different tribunals dealing with similar shapes came to different conclusions or similar conclusions based on different reasons. This illustrates that the question of whether a shape is caught by the Technical Result Exception is a highly fact sensitive one in which the evidence presented, and how it is presented, plays a major role. While a judge may take some guidance from prior decisions which examined the same shapes, I am mindful that, in the end, I have to exercise my best judgment based on the material evidence that is actually before me.

99. The relevance of the plaintiffs’ reasons for choosing the Registered Shapes

100. [47]

     [47] Mr Barnes gave evidence as to the origins of the Registered Shapes and the reasons why the shapes were chosen. In essence, he stated that the shapes were chosen for their aesthetic appeal as well as the taste and feel of the product in question.

101. [48]

     [48] To begin, as already noted at [30] above, the test is an objective one and does not depend on the subjective intentions of the designer of the shape, although they may be taken into account. I believe that the Registered Shapes were motivated at least in part by the need for the Kit Kat product to look and taste desirable to the consumer. A purely functional shape that looks unappetising will not likely be adopted by any competent manufacturer of a food product. However, if this desirable shape is achieved by putting together a number of elements, all of which are necessary for a technical result, it would nevertheless be barred from registration by the Technical Result Exception. This must be so even if, subjectively, the registrant of the shape in question chose the shape primarily for other reasons. Indeed, having regard to the policy underpinnings of s 7(3)(b) of the TMA, it is even more important that competitors should be allowed to avail themselves the use of such shapes.

102. The essential features of the Registered Shapes

103. [49]

     [49] The first step is to consider what the essential features of the Registered Shapes are. Unfortunately, one major difficulty that remains after Lego Juris relates to the criteria by which a court determines the essential characteristics of a three-dimensional sign. The essential characteristics are the most important elements of the sign, but most important to whom and in what sense? Indeed, Rimer J in Philips 1 (HC) at [31] had expressed that it was unfortunate that Philips (ECJ) “did not consider it necessary to offer any express guidance as to what it meant by an ‘essential’ feature or, perhaps more relevantly, what it impliedly had in mind as to what might amount to ‘non-essential’ features”. At [135], Rimer J decided that the answer to whether a feature is essential turned on whether it is a feature of the overall shape which has distinctive eye impact. He considered the evidence of the public to be of “primary relevance” on this issue, although in coming to a conclusion he also provided his own assessment of the sign in question (at [137] to [140]).

104. [50]

     [50] The CJEU in Lego Juris rejected the contention that the identification of the essential characteristics of the sign must be carried out from the perspective of the average consumer, since it is not the distinctive character of the sign that is in issue (Lego Juris at [75] to [77]). Nevertheless, considering the purpose underlying trade mark law, and the fact that in infringement proceedings the assessment of similarity between two marks is carried out from the viewpoint of the average customer (see Hai Tong Co (Pte) Ltd v Ventree Singapore Pte Ltd and another and another appeal [2013] 2 SLR 941 at [40(c)]), it seems to me that, in normal circumstances, the presumed perception of the average consumer will be a relevant consideration. This is consistent with the following passage from Mummery LJ’s judgment in the English Court of Appeal case of Koninklijke Philips Electronics NV v Remington [2006] FSR 30 (“Philips 1 (CA)”) (at [52]):

105. [51]

     [51] The holding above is not inconsistent with Lego Juris, which does not deny that there may be situations where the presumed perception of the average customer is important. In my view, the weight that ought to be given to the presumed perception of the average consumer will depend on the facts. Philips 1 (CA) dealt with a common consumer good, namely, electric shavers. Presently we are concerned with a trade mark registration for a chocolate confectionary product, and the reasoning set out in Philips 1 (CA) is applicable, if not more so. The Goods (see [5] above) are mass manufactured snacks meant for everyday consumption which are targeted at the general public. The average consumer of the Goods is unlikely to scrutinise the nooks and crannies of a chocolate confectionary. The Philips 1 (CA) approach is therefore appropriate in the present case.

106. [52]

     [52] The defendants submit that the essential features of the Registered Shapes are as follows (for avoidance of doubt, I use the phrases “essential features” and “essential characteristics” to mean the same thing):

107. (a) one, the rectangular “slab” shape of the Registered Shapes as they appear on the form of the registration, including the relative proportions of length, width and depth (“the First Essential Feature”);

108. (b) two, the presence, position and depth of the breaking grooves arranged along the length of the bar, which effectively divide the bar into detachable “fingers” (“the Second Essential Feature”); and

109. (c) three, the number of such breaking grooves which, together with the width of the bar, determine the number of fingers (“the Third Essential Feature”).

110. [53]

     [53] These are the features identified by the UK Hearing Officer in Nestle v Cadbury (Opposition), except that the defendants have applied his findings to both the Four Fingers Shape and the Two Fingers Shape in the present proceedings.

111. [54]

     [54] The plaintiffs contend that the essential features of the Registered Shapes are instead the following:

112. (a) the two/four straight fingers wherein each finger is in the shape of an elongated trapezoid;

113. (b) joined on a base;

114. (c) in the characteristic overall dimensions of the Registered Shapes; and

115. (d) a “plinth” surrounding the base on all four sides.

116. [55]

     [55] The “characteristic overall dimensions” in feature (c) above refers to the “placement of 2 and 4 individual fingers… along the longer edge of each finger” for the Two Fingers Shape and Four Fingers Shape respectively.

117. [56]

     [56] The first three features identified by the plaintiffs are similar to the OHIM Second Board of Appeal’s characterisation of the Four Fingers Shape as “four identical trapezoidal bars aligned together on a rectangular base” except that instead of one essential feature, there are now three. Another difference is that the plaintiffs characterisation emphasised that the fingers are joined “on a base” while avoiding the word “rectangular”. The “plinth” was also added as an essential element.

118. [57]

     [57] The defendants objected to the plaintiffs’ provision of a list of essential features. This is because the plaintiffs raised these issues only very late in the day. The plaintiffs did not contest the defendants’ characterisation of the essential features in the affidavits of evidence-in-chief (“AEIC”) of their witnesses, or even raise the point in cross examination. They only did so in their closing submissions. The defendants argue that the plaintiffs should be taken to have accepted the defendants’ list. Be that as it may, I must still decide for myself what the essential characteristics of the Registered Shapes are.

119. [58]

     [58] Considering the relative simplicity of the shape, the identification of the essential characteristics of the Registered Shapes is apt to be “carried out by means of a simple visual analysis of the sign” (Lego Juris at [71]). As noted above at [29(b)], when determining the essential characteristics of a sign, the court may base its assessment directly on the overall impression produced by the sign. Based on an overall impression of the striking features of the Registered Shapes when one takes a quick look at them as an average consumer might do when making a purchase of such goods, I agree with the defendants’ list of essential features.

120. [59]

     [59] I deal first with the overall “rectangular slab shape” of the Registered Shapes, but I will first caveat that it seems that the word “slab” is more apt to describe the Four Fingers Shape than the Two Fingers Shape, when one considers their respective ratios of length and width, although nothing turns on this.

121. [60]

     [60] The plaintiffs argue that describing the Registered Shapes as a rectangular slab shape is effectively saying that a row of trapezoids is a large rectangle, which is as good as calling a triangle a square.

122. [61]

     [61] This argument is pitched at too abstract a level. Take this shape (a two-dimensional one, for simplicity):

123. [62]

     [62] One might describe the figure above as two triangles, joined at the hypotenuse. Another might describe it as a rectangle, with one diagonal line connecting the top right and bottom left corner of the rectangle. Does this shape contain one essential characteristic, or two? Does the shape contain two triangles or just one rectangle? Fortunately, it is rarely, if ever, that a court will have to deal with a question that is so completely devoid of context.

124. [63]

     [63] In my view, it is appropriate to describe the overall basic shape of the Registered Shapes as rectangular. The overall rectangular dimensions of the Four Fingers Shape did not escape the perception of both the Hearing Officer and the OHIM Second Board of Appeal, even if the latter only used the word rectangular to describe the base of the shape. The shapes may contain trapezoidal shaped fingers, but that does not alter the overall rectangular shape. The visual impact of this element is especially pronounced when the Registered Shapes are viewed from above, and what the average customer of a chocolate confectionary will immediately notice is the rectangular feature of the shapes. When the Registered Shapes are turned over and are viewed with its base facing upwards, it looks very much like a rectangular slab. An average consumer would not likely notice the minor feature of a plinth or a thin base on which the individual separate fingers sit or that these separate fingers are actually joined along the bottom of their entire lengths by a thin base.

125. [64]

     [64] As for the grooves, they catch the eye from every view except the views from the bottom and from the long edge of the Registered Shapes. In my view, the grooves are an essential feature. Having accepted that the grooves are an essential feature, it follows that the number of grooves which, together with the width of the bar, determine the number of fingers, is also an essential feature.

126. [65]

     [65] The plaintiffs argue that it is wrong to describe grooves as an essential feature of the Registered Shapes because grooves are in fact an empty space and are not part of the marks in that it does not contain chocolate. This submission is without merit. There is no reason that an empty space cannot be used to describe an essential feature of a shape or a sign. A half empty glass is another way of describing the shape that a half full glass of water takes. One might also think of an engraving, a hole in a doughnut, gaps in a chain, or the grooves of a tyre. Such a feature can be highly noticeable, as well as functional. The plaintiffs also assert that by describing the grooves as “breaking”, the defendants have attempted to unfairly influence the subsequent analysis of whether the essential feature is solely attributable to achieve a technical result. I have not allowed myself to be so influenced.

127. [66]

     [66] As for the plinth (which the defendants refer to as “the bumper”), the plaintiffs raise two points which I will quickly deal with. First, they argue that the defendants do not consider the plinth an essential feature when “the evidence… is that the ‘bumper’ ‘clearly’ alters the appearance of the product”. In this regard, the plaintiffs cite the defendants’ own expert, Professor Peter Jonathan Fryer (“Prof Fryer”), who had said that the defendant’s Take-It product in single format which does not have a bumper looked different from chocolate with a bumper. The answer to this point is contained in the passage from Mummery LJ’s judgment in Philips 1 (CA), cited at [50] above, which is that not every “feature which contributes to the overall impression is an essential feature of the mark”.

128. [67]

     [67] Secondly, they argue that Prof Fryer had accepted during cross-examination that the plinth should be an “essential feature” but was omitted from his expert report. This is what he said:

129. [68]

     [68] Prof Fryer did not exactly agree that the plinth is an essential feature. He only agreed that it was a necessary part (and I am inclined to think that he was referring to the plinth being a necessary part during the manufacturing process to serve a functional purpose as a bumper) and he immediately asserted thereafter in his answer that he did not believe that it formed a part of the shape of the mark. Even if it can be said that Prof Fryer was purporting to give his view on what he considered to be the essential features of the Registered Shapes, the opinion of a manufacturing expert is not conclusive on whether or not the plinth is an essential feature of the Registered Shapes, although expert opinion can be taken into account (see Lego Juris at [71]). The question of what constitutes an essential feature is an objective one. In any event, it is clear from the context that Prof Fryer meant that he was of the view that the plinth was a necessary part of the defendants’ chocolate manufacturing process. In this part of the analysis, we are dealing with the essential elements of the sign to be registered as a trade mark, not the necessary elements of the product for the purposes of efficient manufacturing. The latter is only relevant after it has been determined that it is also an essential element of the shape. A shape feature that is a necessary element for a technical function or an efficient manufacturing process does not ipso facto make it into an essential visual element or characteristic of the three-dimensional shape sought to be registered as a trade mark. When determining from an impressionistic visual examination of what the essential characteristics of that shape or sign for trade mark registration are, the technical functions (if any) resulting from any part of the three-dimensional shape are to be totally ignored. It is only after a certain part of the shape has been found to be an essential characteristic, that the question whether that part of the shape as a physical form or shape is necessary to obtain a technical result is to be examined.

130. [69]

     [69] Accordingly, if the plinth is not an essential feature of the Registered Shapes, whether or not it is also necessary for a technical result is irrelevant. Let us say that a sign in the form of a shape has essential features A, B and C. It also contains a non-essential feature D. If A, B and C are all functional and D is non-functional, it is clear that this sign will be precluded from registration by the Technical Result Exception. It cannot make a difference if D is also independently functional since it has been determined not to be an essential feature of the registered shape.

131. [70]

     [70] On the other hand, if A and B are non-functional essential features and are purely aesthetic in nature, but C and D are functional, the sign will not be precluded from registration because not all the essential features A, B and C are functional. This is consistent with the policy that registration of such a three-dimensional shape should not be allowed to exclude another manufacturer M from utilising the functional usefulness of characteristics C and D in the shape of M’s product because characteristic D being a non-essential characteristic of the registered shape is not going to lead to M’s product shape being a trade mark infringement of the registered shape with characteristic D anyway. Similarly, the incorporation by M of shape feature C into its product in order to utilise the functional usefulness of characteristic C is not likely to lead to a trade mark infringement of the registered shape because M’s product only has the essential shape characteristic C but not the rest of the essential features A and B considered to constitute (together with C) the whole of the registered shape. Consequently, there will hardly be any visual similarity between M’s product shape (though having essential features C and D in its shape necessary for the functional aspects but not incorporating any aesthetic features A and B) and the registered shape (having essential features A, B, C and a non-essential feature D) leading to a likelihood of confusion on the part of the public for trade mark infringement considerations. The proprietor of the registered shape has thus not gained an unfair long term monopoly though registration of its three-dimensional trade mark, that would have prevented others from using the useful functional characteristics of shapes C and/or D.

132. [71]

     [71] What if, A and D must operate in combination to provide a technical result X? In such a case, A will be considered necessary to obtain this technical result X, ie, assuming that A is essential, it will not be a non-functional essential feature, but a functional essential feature.

133. [72]

     [72] In determining whether the plinth is an essential feature of the Registered Shapes, I have to keep in mind that the average customer for a chocolate confectionary product is unlikely to pay the same level of scrutiny to a shape as a judge pondering over the essential characteristics of a particular shape. My view that the average consumer would not have paid it any notice is reinforced by Nestle v Cadbury (Opposition), where the Hearing Officer determined (at [67]) that the plinth is unlikely to make any real impact on an average consumer. He referred to survey findings where only one of 473 respondents mentioned a “ridge” around the edge of the bar. The plaintiffs have also adduced survey evidence before me, but they did not contain such relevant details.

134. [73]

     [73] Similarly, Nestle v Cadbury (OHIM) did not mention the plinth at all when determining the essential features of the Four Fingers Shape. It was only mentioned in the decision of Louw J in Nestle (South Africa), who described it as an “apron”. After some consideration, I find that the “plinth” is merely a minor feature and not an essential characteristic of the Registered Shapes, and not “one of the most important elements of the sign”.

135. [74]

     [74] Having found that the defendant’s list of essential features is correct, the next issue is whether all the essential features are functional.

136. Whether all the essential features of the Registered Shapes are functional

137. [75]

     [75] The defendants’ case as to why all the essential features of the Registered Shapes are functional is as follows:

138. (a) First Essential Element: The rectangular slab shape of the Registered Shapes is necessary to create a shape that can be manufactured and packed at the speed required for commercial production.

139. (b) Second Essential Element: Although the manufacturer can choose where and how to run the grooves in the product, the presence of breaking grooves is fundamentally necessary to allow the product to be broken down for consumption, and the angle of those grooves would be dictated by the position and depth of these grooves as release angles of some kind will be required to allow the product to be efficiently removed from the mould.

140. (c) Third Essential Element: the number of breaking grooves and fingers is necessary to achieve the technical result of providing the customer with a desired portion size.

141. (1) The manufacturing issue

142. [76]

     [76] The plaintiffs contend that the Technical Result Exception has no application to the manufacturing process but is instead confined to the examination of the functional features of a shape mark from the perspective of the user of the product. If they are correct, the defendants will fail.

143. [77]

     [77] It should be recalled that the Hearing Officer in Nestle v Cadbury (Opposition) considered that the Second and Third Essential Features are necessary to achieve a technical result. He had found that the angle of the sides of the product, and of the grooves, in the Four Fingers Shape, is constrained by the method of manufacture (Nestle v Cadbury (Opposition) at [78]).

144. [78]

     [78] On appeal, the first plaintiff argued that the Hearing Officer was wrong to conclude that the Technical Result Exception applies to features which related to the manner in which the goods were manufactured, such as the angle of the grooves, as opposed to the manner in which the goods functioned (Nestle v Cadbury (HC) at [68]). Counsel for Cadbury, however, argued that there was nothing in the wording of Art 3(1)(e)(ii) of the European Parliament and Council Directive 2008/95/EC (“the Directive”) to limit it to the function of the goods. He also argued that it would be inconsistent with the policy underlying the Technical Result Exception to limit it in that way (Nestle v Cadbury (HC) at [73]). Art 3(1)(e)(ii) of the Directive is in pari materia with s 7(3)(b) of the TMA.

145. [79]

     [79] Arnold J expressed his preference for Cadbury’s submission, but did not feel like he was able to say that this interpretation of the Directive is clearly correct. As a result, he referred the question below to the CJEU for a preliminary ruling (at [75]):

146. [80]

     [80] Arnold J considered that the answer ought to be yes (at [76]).

147. [81]

     [81] The first plaintiff had submitted before Arnold J, and did so again before me, that Philips (ECJ) at [78] stands for the proposition that the Technical Result Exception has no application to the manufacturing process but is confined to the examination of the functional features of a shape mark from the user of the product’s perspective. I reproduce the relevant paragraph:

148. [82]

     [82] In response, the defendants cite a number of other paragraphs from Philips (ECJ) to show that Philips (ECJ) should not be read so narrowly. They emphasise that the case refers not only to “functional characteristics”, but also to “technical solutions”.

149. [83]

     [83] It seems to me that there is not much value in dissecting the precise language used in Philips (ECJ). In the passages cited, the court was laying the foundation to answer a very specific question, which is essentially whether the Technical Result Exception “must be interpreted to mean that a sign consisting exclusively of the shape of a product is unregistrable by virtue of that provision if it is established that the essential functional features of the shape are attributable only to the technical result” and “whether the ground for refusal or invalidity of the registration imposed by that provision can be overcome by establishing that there are other shapes which can obtain the same technical result” (Philips (ECJ) at [66]). The minds of the judges were not directed to whether the words “technical result” should be interpreted in a narrow or expansive way. At best, all it can be said is that Philips (ECJ) did not consider whether the Technical Result Exception applies to methods of manufacture.

150. [84]

     [84] In any event, Philips (ECJ) is not the only case that has expounded on the policy underpinnings of the Technical Result Exception. It was described by Mummery LJ in Philips 1 (CA) at [19] in the following way:

151. [85]

     [85] Reading s 7(3)(b) of the TMA literally, there is nothing to suggest that the Technical Result Exception is only restricted to the manner in which the goods function. There is authority that it should not be read so narrowly. The defendants refer to the South African High Court case of Triomed (Proprietary) Limited v Beecham Group Plc [2001] FSR 34 (“Beecham (HC)”). The shape mark that was under attack for invalidity in that case was a bi-convex pharmaceutical tablet with a regular oval shape. It was argued that the mark should be expunged pursuant to, inter alia, s 10(5) of South Africa’s Trade Marks Act (No 194 of 1993) (“SA TMA”), which states that “a mark which consists exclusively of the shape, configuration, colour or pattern of goods where such shape, configuration, colour or pattern is necessary to obtain a specific technical result, or results from the nature of the goods themselves” shall be liable to be removed from the register. While the wording of s 10(5) of the SA TMA is different from s 7(3)(b) of the TMA, it appears to share the same purpose.

152. [86]

     [86] The following paragraphs of Beecham (HC) (at [75] and [77]) demonstrate that the judge took into account the functionality of the shape from both a manufacturer perspective and a user perspective:

153. [87]

     [87] Beecham (HC) was upheld by the Supreme Court of Appeal of South Africa (see Beecham Group Plc and Others v Triomed (Pty) Ltd (100/01) [2002] ZASCA 109 (“Beecham (SC)”) at [29] and [30]).

154. [88]

     [88] The plaintiffs argue that the Technical Result Exception cannot apply to the manufacturing process because if it were so, all shape marks would not be registrable because shape marks must by definition be capable of manufacture. It is very easy for any manufacturer to arbitrarily point to any feature of the shape mark in question and argue that because of his own particular way of manufacturing, he must have that particular feature in question, even though there is in fact no monopolisation of the feature in question by the proprietor of the trade mark.

155. [89]

     [89] I am unable to see how extending the Technical Result Exception to the manufacturing process would have the effect of barring shape marks arbitrarily. First, a defendant will not succeed merely because he can show that a shape is capable of manufacture. Indeed, in the present case, there is no assertion that the presence of the grooves in the Registered Shapes or the number of grooves is necessary to obtain a technical result in the manufacture of chocolate products whether by the plaintiffs or the defendants using their particular way of manufacturing, and I suspect that they would not be. The need for grooves and the number of grooves are not dictated by the manufacturing process. Having regard to the policy underlying the Technical Result Exception, a defendant should be allowed to show that a plaintiff’s trade mark should not be registrable not only on the basis that a particular essential feature of the shape in issue is needed to create a useful functional feature in the goods for the user but also that it is necessary for the efficient manufacture of the goods in question. This is not at all arbitrary.

156. [90]

     [90] Secondly, even if a feature is not necessary in one party’s particular way of manufacturing but necessary in another party’s particular way of manufacturing, it does not mean that the Technical Result Exception should therefore not apply to the manufacturing process. When determining whether a feature of a shape is necessary for the efficient manufacture of a good, one must take into account the fact that in the marketplace for goods, there will be players of different sizes, resources and sophistication. What may not be “necessary” for a global firm with deep pockets may be crucial for a smaller start-up that wishes to compete. Trade mark law should not be misused to unjustifiably set up additional barriers of entry and impede the freedom of competition. There is undoubtedly a deleterious effect on the efficient operation of the market if the registration of a three-dimensional sign has the effect of forcing competitors to resort to a more expensive manufacturing process, or to use a cheaper process which would result in a product that is less attractive to its intended customer. As Prof Fryer said, a production line can become very sophisticated if one were to throw enough engineers at a problem. I do not doubt that Nestle’s production methods are sophisticated indeed and it may be that Nestle’s sophisticated production methods do not require a certain product shape for efficient manufacturing. However, that does not preclude the same product shape from being a technical solution for its less sophisticated competitors. Accordingly and thirdly, the fact that the proprietor of the shape mark may not require a feature to manufacture the product efficiently (at the present time) will not be determinative of the issue of whether or not the feature is functional and therefore falls within the Technical Result Exception.

157. [91]

     [91] Upon a closer look, the plaintiffs’ point regarding the absence of “monopolisation” is really just a variant of the discredited argument that a shape is not necessary for a technical result if it can be shown that other shapes can get the same result. This is reinforced by Lego Juris at [59] and [60] that states that the need to avoid unduly impairing the opportunity for competitors to place on the market goods whose shapes incorporate the same technical solution applies a fortiori in a case where the competent authority has found that “the solution incorporated in the shape of goods examined is the technically preferable solution for the category of goods concerned”. For example, I think the South African courts were right in the Beecham cases to regard the fact that the bi-concave shape facilitates the coating of a tablet as a relevant technical result. There may be other ways of manufacturing the tablet so that the tablet may be coated without using a “bi-concave shape” but for the reasons stated in Philips (ECJ) and Philips 1 (CA) about the policy underpinnings of the Technical Result Exception, this is irrelevant.

158. [92]

     [92] Having found that the Technical Result Exception is not confined to the examination of the functional features of a shape mark from the perspective of the user of the product, but also applies to the manufacturing process, I will now examine the essential features in turn to see if they fall foul of the exception.

159. (2) The First Essential Element: The rectangular slab shape

160. [93]

     [93] On the question of whether the First Essential Feature is necessary to obtain a technical result, the defendants rely on Prof Fryer’s expert evidence that the rectangular slab shape, including the relative proportions of the length, width and depth are controlled by the requirements to create a shape that can be manufactured and packed at the speed required for commercial production.

161. [94]

     [94] According to Prof Fryer, mass produced chocolate is most commonly made through the methods of “enrobing” or “moulding”. Enrobing is the method where the centre product (which may not be chocolate) is coated with molten chocolate when progressing along a conveyer belt. Moulding involves first depositing molten chocolate into moulds which give the chocolate its shape and later (after cooling) removing the chocolate from the moulds. Both the Kit Kat products and the Take-It products are moulded.

162. [95]

     [95] Typically, moulding is used when a precise and particular consistent chocolate shape or detailing is required for consumer appeal or in order to fit particular packaging requirements. To be clear, the choice of enrobing or moulding can lead to goods of differing quality. Mr Barnes gave evidence that enrobing tends to be much cheaper than moulding. Enrobing results in a product that looks more generic as it will just take the shape of the product that is underneath them (as opposed to moulded chocolate, which takes the shape of the mould). Enrobing also tends to be “less premium” in terms of its finish. Mr Chuang also said that enrobed and moulded chocolate confectionary taste differently and the latter is generally perceived by consumers to be of higher quality and value.

163. [96]

     [96] To make a solid chocolate moulded product, molten tempered chocolate is poured into warm moulds. They are then vibrated to remove any air bubbles and ensure an even distribution of the chocolate in the mould. If the chocolate fills the mould, the top of the mould (which becomes the bottom of the chocolate after demoulding) may be scraped flat, unless the mould is not filled entirely (in which case it will settle into a flat base). For chocolate covered wafers, there will be certain additional steps relating to the placement of the wafers.

164. [97]

     [97] The demoulding process involves the mould being turned upside down and the underside being knocked to release the chocolates, which fall out onto a conveyor belt that usually conveys them to a hopper in a wrapping or packaging machine.

165. [100]

     [100] Finally, Prof Fryer also stated that the base must be generally flat for placement and to allow for units to be stacked alongside each other. He said a base that is not flat would lead to technical difficulties and would make manufacturing by moulding impossible.

166. [101]

     [101] The plaintiffs argue that I should not place much weight on Prof Fryer’s evidence as it is based “entirely on academic knowledge”. I do not think that made his evidence unreliable. I have no reason to doubt his credentials with respect to the design of moulds and the manufacture of moulded chocolates. While he is mainly a scholar, he has worked very closely with major food companies as well. For example, it is his evidence that he is familiar with a number of major Cadbury products. In my view, his evidence is compelling, credible, and accords with common sense. Thus, for the reasons he has provided, I find that the rectangular slab shape is necessary for a technical result.

167. [102]

     [102] Before I turn to the next point, I note that the Hearing Officer in Nestle v Cadbury (Opposition) found that the First Essential Element resulted from the nature of the goods themselves, but did not actually say that it is necessary for a technical result. This was what he said at [72] and [73]:

168. [103]

     [103] On appeal, Arnold J did not accept Cadbury’s argument that the Hearing Officer’s findings supported the conclusion that the First Essential Feature was necessary to obtain a technical result, albeit without any elaboration (Nestle v Cadbury (HC) at [65]).

169. [104]

     [104] However, the decisions of the English courts though persuasive are not binding on me. On a separate note, I can think of why a feature dictated by the manufacturing process might be regarded as resulting from the nature of the goods. Where an essential feature presents an obvious technical solution that contributes significantly to the efficient and economical manufacture of the specific good in question, one would expect that particular essential feature to be commonplace among goods of the same kind and for the shape of these goods to take a somewhat similar form, which then suggests that that is the general or inherent nature of the shape of those goods. In this sense, there is a certain degree of overlap between sub-ss 7(3)(a) and (b) of the TMA.

170. (3) The Second Essential Element: The presence, position, and depth of the breaking grooves

171. [105]

     [105] The defendants submit that the presence of the grooves allows the product to be broken up for consumption. The depth of the grooves is a compromise between (a) the minimum depth required for the product to retain structural integrity in manufacture and transit, and (b) the maximum permissible depth before consumers would have difficulty in breaking the bar cleanly into regular shaped and conveniently consumable fingers. The grooves also have to be angled in order for the product to be effectively demoulded, even if the precise angle itself is not an essential feature of the shape.

172. [106]

     [106] The Hearing Officer in Nestle v Cadbury (Opposition) and Arnold J in Nestle v Cadbury (HC) both took the view that the grooves were necessary for a technical result, while the OHIM Second Board of Appeal in Nestle v Cadbury (OHIM) considered that the ease of portioning of a product was not a technical result. It seems to me that the difficulty here is, again, the drawing of boundaries. To borrow the language of Beecham (SC) at [30], a shape is necessary for a technical result if it is a reasonable technical solution to a problem. The problem itself does not have to be “technical” one. So in Beecham (SC), the “ease of swallowing” of a pill was regarded as a specific technical result.

173. [107]

     [107] Of course, a chocolate product can be broken up even without the presence of breaking grooves if one accepts untidy and uneven broken pieces of chocolate instead of neat and cleanly broken smaller pieces of long fingers of chocolate. The question is whether this means that the presence of the grooves “is an unnecessary and redundant solution to begin with”, as the plaintiff has submitted. I think this argument is misguided. In the first place, even if feature X is not necessary as a technical solution to achieve outcome Y per se, a more efficient or otherwise desirable method such as having a feature X to attain outcome Y is nevertheless to be regarded as necessary to obtain that technical result Y. It is unquestionable that the presence of grooves makes a chocolate bar easier to break apart. Moreover, where there is a groove, a break will be cleaner and more aesthetically pleasing. Finally, the presence of grooves also allows for more precise apportionment of the chocolate product, such as in halves or quarters, with respect to the Registered Shapes (a point which overlaps with the Third Essential Feature, as we shall later see). Accordingly, I regard the use of the longitudinal grooves in the three-dimensional shape to be necessary to obtain the desirable technical result of facilitating the breaking up of a larger rectangular slab of chocolate bar into smaller neat finger like pieces of chocolate bars with the appropriate proportions of a suitable size each time for eating.

174. [108]

     [108] As for the angle of the sides of the grooves, it is constrained by the need for the chocolate product to be demoulded. Without going into details, the sides of a moulded chocolate product have to be sloped for the efficient demoulding of the product. This is called the “release angle”. According to Prof Fryer, a release angle of eight to ten degrees is considered the minimum. Of course, there are other shapes and angles (as long as they are greater than eight degrees) which can be efficiently demoulded. Mr Barnes pointed out that even a square shape can be demoulded using the appropriate technique (albeit a more complicated one). Nevertheless, it seems to me that the average consumer will not notice minor variations in the angles of the sides and the grooves of the Registered Shapes. Other considerations in determining the angle of the slope of the grooves in a chocolate wafer product is the consumer’s need to grasp the wafer and separate the fingers by hand as well as certain other manufacturing and packaging requirements. Taken together the precise angle is necessarily constrained. I regard the Second Essential Element comprising the shape of the breaking grooves with its particular release angles as being necessary to obtain a technical result that satisfies s 7(3)(b) of the TMA.

175. (4) The Third Essential Element: The number of breaking grooves and fingers

176. [109]

     [109] In this regard, Prof Fryer’s evidence is that the number of grooves, together with the width of the bar, determines the number of “fingers”, and that the number of fingers is given by the need to provide the consumer with individual pieces. Similarly, the Hearing Officer in Nestle v Cadbury (Opposition) at [77] found that the number of fingers for the Four Fingers Shape is determined by the size of the chosen portion. In contrast, Louw J in Nestle (South Africa), the other decision which dealt squarely with the issue of the number of fingers for the Registered Shapes, considered that a “chocolate slab obviously does not need to be in the shape of two or four fingers” (at [30]).

177. [110]

     [110] As Prof Fryer had noted, the size of a chocolate bar is constrained by handsize and convention (bars that are shorter or longer are difficult to put into shop displays). Once a chocolate bar is breakable, the next question is: What is the portion size? This will be determined by the consumers’ need for more or less of the product depending on their appetite at the time of purchase (Nestle v Cadbury (Opposition) at [77]). Much like the story of Goldilocks and the Three Bears, the consumer wants each portion to be just right. Considering these restraints, a chocolate bar or wafer can only be apportioned in so many ways before a portion is simply too big or too small for the average consumer. Accordingly, the number of grooves and the number of fingers in the Registered Shapes is not an arbitrary one. As Arnold J put it, “it was necessary for there to be three grooves in order for there to be four fingers each of which was of the desired portion size” (Nestle v Cadbury (HC) at [67]).

178. [111]

     [111] For the foregoing reasons, I find that each and every one of the essential features of the Registered Shapes is necessary for a specific though different technical result, and the registrations of the Registered Shapes are therefore invalid pursuant to s 7(3)(b) read with s 23(1) of the TMA.

179. Whether the Registered Shapes result from the nature of the goods themselves

180. The key legal principles

181. [112]

     [112] The defendants argue that the registrations of the Registered Shapes are also prohibited under s 7(3)(a) of the TMA for being shapes which result from the nature of the goods themselves. Again, for simplicity, I will refer to the bar to registration under the TMA, as well as other equivalent statutory provisions found in various jurisdictions, as the “Nature Exception”.

182. [113]

     [113] The Nature Exception was recently the subject of analysis in Hauck GmbH & Co KG v Stokke A/S, Stokke Nederland BV, Peter Opsvik and Peter Opsvik A/S (Case C-205/13) (“Hauck”), a decision which was handed down by the CJEU on 18 September 2014. The case was a reference for a preliminary ruling from the Netherlands and the shape in issue was that of the “Tripp Trapp” children’s chair, which was registered for “chairs, especially high chairs for children”. One of the questions referred to the CJEU was the following:

183. [114]

     [114] The CJEU made the following findings:

184. (a) The Nature Exception pursues the same objective as the other two exceptions under the European equivalent of s 7(3) TMA (at [20]).

185. (b) To apply the Nature Exception, it is necessary to identify the essential characteristics of the sign on a case-by-case basis (at [21]).

186. (c) The Nature Exception is not limited only to (i) “natural” products (which have no substitute) and (ii) “regulated” products (the shape of which is prescribed by legal standards) as such shapes could not be registered in any event because of their lack of distinctive character (at [24]).

187. (d) The concept of a “shape which results from the nature of the goods themselves” also includes shapes with essential characteristics which are inherent to the generic function or functions of the goods. The reason for this is that it would otherwise make it difficult for competitors to give their goods a shape suited to the use to which those goods are intended. It is clear that those are essential characteristics which consumers will be looking for in the products of competitors as they are intended to perform an identical or similar function (at [25]).

188. [115]

     [115] Accordingly, the CJEU found that the Nature Exception may apply to a sign which “consists exclusively of the shape of a product with one or more essential characteristics which are inherent to the generic function or functions of that product and which consumers may be looking for in the products of competitors” (Hauck at [27]).

189. [116]

     [116] To be clear, I do not think the CJEU is saying that the Nature Exception applies only by reference to the function of a good. This would not be sensible in situations where the goods involved have a naturally occurring shape, like flowers or fish. In a sense, these shapes are also functional from the perspective of the flower or the fish, but in trade mark law, we do not concern ourselves with the needs of plants or marine life (except insofar as they further our own ends). In such cases the Nature Exception can be applied quite straightforwardly.

190. [117]

     [117] I illustrate this by referring to the written Opinions of the Advocate General for Hauck, where Mr Maciej Szpunar gave a number of examples which would be excluded from registration under the Nature Exception (at paras 46, 49 and 59):

191. (a) natural goods, eg, the shape of a banana for bananas (being a natural good);

192. (b) standardised shapes, eg, a rugby ball;

193. (c) a shape consisting exclusively of characteristics ordinarily attributed to the goods in question, eg, the shape of a parallelepiped for a brick, the shape of a container with a spout, a lid and a handle for a teapot, or the shape of fork tines for a fork; and

194. (d) shapes which essential characteristics derive from the function of the goods in question, eg, feet attached to a horizontal plate with regard to a table.

195. [118]

     [118] It is apparent that the Nature Exception does apply to man-made goods of which the shape is not pre-determined and which may have alternative shapes. Take the example of the table, which Mr Szpunar described as a horizontal plate with legs. There may be other possible shapes, for example, a horizontal plate that is suspended from the ceiling. Such a legless shape may still function as a table, but that does not mean that it is not natural for a table to have legs.

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