High Court · Decided 26 July 2011

Ferrero SPA v Sarika Connoisseur Cafe Pte Ltd

Feedback & support

[2011] SGHC 176Case no. Suit No 9 of 2010

What happened, in plain English

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

Ask about this judgment

Grounds of decision

Official text from eLitigation

Chan Seng Onn J:
[1] The Plaintiff is a company incorporated under the laws of Italy. It manufactures and sells confectionary goods, including a cocoa-based hazelnut bread spread under the brand “Nutella”. The Defendant is a company incorporated under the laws of Singapore. It is in the business of retail in food and beverage and the wholesale of coffee, cocoa and tea. It owns and operates the café outlets known as “tcc – Art Boutique Caffe”, now known as “tcc – the connoisseur concerto” (“TCC”).
Factual background
[2] The Plaintiff manufactures and markets world renown brands of confectionery such as “Ferrero Rocher”, “Nutella”, “Confetteria Raffaello”, the “Kinder” brand series (comprising “Kinder Bueno”, “Kinder Chocolate” and “Kinder Surprise”), “Mon Cheri” and “Tic Tac”.
[3] With regard to the “Nutella” brand, the Plaintiff is the registered proprietor of the following marks in Class 30 of the International Classification of Goods and Services (“ICGS”) in Singapore (collectively referred to as the “‘Nutella’ marks”):
[4] For clarity, the trade mark reproduced in the first row above (T73/59592H) will be referred to as “the ‘Nutella’ word mark”, whereas all the three trade marks above will be referred to collectively as “the ‘Nutella’ trade marks”.
[5] On 1 August 2007, the Defendant introduced a new gourmet hot coffee beverage served in a shot glass (the “‘Nutello’ drink”, or the “Defendant’s product”) under the “Nutello” sign (the “‘Nutello’ sign”), prepared at TCC outlets by the Defendant’s baristas. I have been careful to use the word “sign” in describing the term “Nutello”, as the word “sign” has a specifically defined meaning in the Trade Marks Act (Cap 332, 2005 Rev Ed) (“TMA”). Section 2(1) of the TMA defines the term “sign” as:
[6] This is the most generic description available under the TMA, and it is necessary to use such a term so as to distinguish a “sign” from a “trade mark”, the latter having its own statutorily defined meaning of (s 2(1) of the TMA):
[7] It is clear from the definitions of “sign” and “trade mark” reproduced above that all trade marks are signs, but that the converse is not true. Some confusion arose in this case as a result of counsel for the Defendant referring to “Nutello” as a “mark”, when what he appeared to have meant was that “Nutello” was a “sign” (see [37] et seq). To avoid such confusion, I have consistently used the generic term “sign” to describe “Nutello” throughout this judgment.
[8] In the Defendant’s TCC Drinks Gallery Menu, the “Nutello” drink was featured under the “Espresso Specialties” section. The “Nutello” drink is served in a shot glass and consists of, inter alia, espresso, milk foam, cocoa powder and “Nutella” chocolate cream spread. The TCC Drinks Gallery Menu describes “Nutello” as “Espresso with lashings of nutella – perfect for cocoa lovers!” Each “Nutello” drink is priced at S$5.60, excluding government taxes and service charge.
[9] Since the launch of the “Nutello” drink, the “Nutello” name has been used in various representations in the Defendant’s promotional materials:
[10] On 3 December 2009, the Plaintiff issued a cease-and-desist letter to the Defendant objecting to the Defendant’s use of the “Nutello” name and the description “Nutello – Espresso with lashings of nutella – perfect for cocoa lovers!” The Plaintiff demanded, inter alia, payment of S$50,000 as damages.
[11] On 15 December 2009, the Defendant requested for an extension of one month, ie till 17 January 2010, to respond to the Plaintiff’s letter. However, the Plaintiff refused to accede to such a request by way of a letter dated 16 December 2009. The Plaintiff initiated the present action on 6 January 2010.
[12] In July 2010, by reason of an annual overhaul of the entire Drinks Gallery Menu, the Defendant discontinued sales of the “Nutello” drink.
The Plaintiff’s Claims
[13] In the Plaintiff’s statement of claim, the Plaintiff alleged that the Defendant:
(a) Infringed the Plaintiff’s “Nutella” word mark under s 27(2)(b) of the TMA by using a similar sign on identical and/or similar goods (hereinafter, the “trade mark infringement claim”); and/or
(b) Infringed the Plaintiff’s well known “Nutella” trade marks by (hereinafter, the “well known trade mark infringement claim”):
(i) using a similar sign on identical and/or similar goods, which is likely to cause confusion (under s 55(2) of the TMA);
(ii) using a similar sign, in relation to any types of goods, such as to indicate a connection between those goods and the Plaintiff, and is likely to damage the Plaintiff’s interests (under s 55(3)(a) of the TMA);
(iii) using a similar sign, in relation to any types of goods, which would:
(A) cause dilution in an unfair manner of the distinctive character of the Plaintiff’s “Nutella” marks (under s 55(3)(b)(i) of the TMA); or
(B) take unfair advantage of the distinctive character of the Plaintiff’s “Nutella” marks (under s 55(3)(b)(ii) of the TMA).
(c) Committed the tort of passing off (hereinafter, the “passing off claim”).
[14] This judgment proceeds to address the various claims made by the Plaintiff. I would note at the outset that many aspects of the various claims overlap with each other, even if, at times, they do not adopt the same legal parlance. Accordingly, in order to streamline the analysis, extensive use of cross-referencing – even across different heads of claim – will be adopted where appropriate.
[15] It should be noted that while injunctions and damages may be ordered under the trade mark infringement claim and the passing off claim, the only relief with regard to the claim of well known trade mark infringement is limited to a restraining injunction.
[16] I turn first to consider the trade mark infringement claim.
The trade mark infringement claim (s 27(2)(b) of the TMA)
[17] The Plaintiff’s trade mark infringement claim is based only on one of the Plaintiff’s registrations, viz, the “Nutella” word mark.
[18] The issues presented by the trade mark infringement claim are multiple and complex. Accordingly, I shall provide a brief overview on the structure of the analysis adopted in this judgment.
[19] The trade mark infringement claim is premised on s 27(2)(b) of the TMA which provides:
[20] Before infringement under s 27(2)(b) of the TMA can be considered, an important threshold issue is whether the infringing sign must be used as a trade mark in the course of trade (“the threshold issue”). This issue will be dealt with in greater detail below (see [29] et seq).
[21] Assuming the threshold issue has been resolved in favour of the Plaintiff, the Plaintiff must still show three elements in order to establish an infringement under s 27(2)(b) of the TMA. The three elements are, namely (The Polo/Lauren Co, LP v Shop-In Department Store Pte Ltd [2006] 2 SLR(R) 690 (“Polo”) at [8]; cited in City Chain Stores (S) Pte Ltd v Louis Vuitton Malletier [2010] 1 SLR 382 (“City Chain”) at [43]):
(a) The alleged offending sign must be similar to the registered mark (hereinafter, “similarity of marks”);
(b) Both the sign and the mark must be used in relation to similar goods or services (hereinafter, “similarity of goods”); and
(c) On account of the presence of the first two conditions, there exists a likelihood of confusion on the part of the public (hereinafter, “likelihood of confusion on the part of the public”).
[22] These elements will be analysed in greater detail below. However, two points should be noted here. First, although there is some link and overlap in the three conditions, it is clear that they are distinct conditions – the fact that a sign is similar to a registered mark, and/or that it is used on similar goods, does not automatically mean that there will be a likelihood of confusion on the part of the public (Polo at [8]; City Chain at [44]). Second, the third question of whether there exists a likelihood of confusion need not be considered if either of the first two conditions is not satisfied. This is because the “global assessment test” – a test which focuses on the ultimate question of whether there is a likelihood of confusion – has been decisively rejected by the Court of Appeal in Polo (see Polo at [8]; and see City Chain at [44]).
[23] I now consider the threshold issue of whether the allegedly infringing mark must be used in a trade mark sense before the issue of trade mark infringement can be considered.
The threshold issue: Must there be trade mark use?
[24] The Court of Appeal in City Chain considered at length the threshold issue of whether the infringing sign must be used in a trade mark sense (hereinafter, referred to as either “trade mark use” or “used in a trade mark sense”) before a trade mark infringement can be found (City Chain at [16]-[38]).
[25] Before proceeding to consider whether trade mark use is necessary, it is necessary to first consider what amounts to “trade mark use”. The cases cited to me have not provided a clear definition of what “trade mark use” constitutes.
The meaning of “trade mark use”
[26] In unpacking the concept of “trade mark use”, the starting point is to seek recourse to the meaning of “trade mark”. Flowing from this definition of “trade mark” in s 2(1) of the TMA (reproduced above at [6]), “trade mark use” must refer to the use of a mark in a manner that is indicative of the trade origin of goods or services. This is because – as is suggested by the s 2(1) definition of “trade mark” – a trade mark is, in essence, a badge of origin. In this regard, it is apposite to note the judgment of Lord Nicholls in R v Johnstone at [13] (also cited in Nation Fittings (M) Sdn Bhd v Oystertec plc [2006] 1 SLR(R) 712 (“Nation Fittings”) at [64]):
[27] Therefore, where the allegedly infringing sign is used for, inter alia, decorative or descriptive purposes, this would not amount to trade mark use. For instance, in City Chain, the Court of Appeal found that the predominant use of the allegedly infringing sign was for decorative purposes and therefore was not trade mark use (City Chain at [38]). Again, in Hölterhoff v Freiesleben [2002] FSR 52 at [16], the European Court of Justice (“ECJ”) noted that a reference made for purely descriptive purposes to reveal the characteristics of the product offered for sale to the potential customer was not trade mark use.
[28] It should be noted that certain types of signs may not be capable of being used as trade marks. Ex hypothesi – and this flows from the definition of “trade mark” in s 2(1) of the TMA – where the allegedly infringing sign is not capable of indicating the trade origin of goods or services, it cannot be considered to be used as a trade mark. However, for the avoidance of doubt, I emphasise that the allegedly infringing sign does not have to be registrable as a trade mark before its use can constitute “trade mark use”. Registrability is contingent on other requirements as found in s 7 of the TMA, and such requirements are not necessary for showing that there has been “trade mark use”. Instead, for the purposes of demonstrating “trade mark use”, it is sufficient that the allegedly infringing sign is shown to be capable of, and is in fact, used as a badge of origin.
Whether trade mark use is necessary
[29] A controversial issue in the law of trade marks, not just in Singapore but also in other jurisdictions (and especially the European Community), is the issue of whether the alleged infringer must have used the allegedly infringing sign as a trade mark before trade mark infringement can even be considered. In City Chain, the Court of Appeal noted that there were two different approaches to this issue, namely: (1) the traditional approach in which infringing use must be trade mark use (“the traditional approach”); and (2) the broader approach adopted by the ECJ (“the broader Community approach”).
[30] The former approach (viz, the traditional approach) was considered by Andrew Phang Boon Leong J in Nation Fittings. The learned judge, albeit in obiter, dealt with the requirement of use of the infringing sign as a trade mark. He considered, inter alia, the approaches taken by Lord Nicholls of Birkenhead and Lord Walker of Gestingthorpe in R v Johnstone [2003] FSR 42 (“R v Johnstone”), the ECJ’s decision in Arsenal Football Club Plc v Reed [2003] Ch 454 (“Arsenal (ECJ)”) and the decision of Jacob J in British Sugar Plc v James Robertson & Sons Ltd [1996] RPC 281 (see [57], [61], [63], [64], and [67]-[69] of Nation Fittings). After considering inter alia the abovementioned cases, the learned judge was of the view that the interpretation adopted must be consistent with logic and fairness (Nation Fittings at [72]). In the circumstances, he favoured Lord Nicholls’ reasoning in Johnstone (Nation Fittings at [63] and [73]), viz, that before there can be a trade mark infringement, the alleged offending use must be of the nature of a trade mark use (Nation Fittings at [62]).
[31] The latter approach (viz, the broader Community approach) had its genesis in the ECJ decision of Arsenal (ECJ). In brief, and to adopt the interpretation of Arsenal (ECJ) taken by Aldous LJ in Arsenal Football Club Plc v Reed [2003] RPC 39 at [17] (“Arsenal (CA)”), the court in Arsenal (ECJ) held that the registration of a trade mark gave the proprietor a property right, and the relevant consideration regarding the allegedly infringing use is whether it is likely to “affect or jeopardise the guarantee of origin which constitutes the essential function of the mark [which] did not depend on whether the use complained of was trade mark use” (emphasis added). The test is whether the infringing use is liable to affect the functions of the registered trade mark (City Chain at [20], interpreting Arsenal (ECJ)).
[32] Following a detailed exegesis of English and European cases as well as academic writing, the Court of Appeal in City Chain noted at [36] that:
[33] The Court of Appeal noted that, on the facts of the case before it, the “predominant use” of the allegedly infringing mark was for “decorative purposes” rather than “trade mark use” (City Chain at [38]). However, although this finding “should conclude the question of [trade mark infringement under s 27 of the TMA]”, the Court of Appeal recognised that this would not conclude the issue under the broader Community approach (at [38]). The Court of Appeal therefore went on to analyse the three elements of trade mark infringement (set out at [21] above). It is therefore clear that the Court of Appeal in City Chain did not have to finally determine the correct approach to be applied vis-à-vis the threshold issue, because on the facts of that case the same outcome would be reached whichever approach was taken (see City Chain at [36], reproduced above at [32]). In other words, although the Court of Appeal favoured the narrower “trade mark use” approach, this was done in obiter.
[34] In the present case, there is no need for me to decide which approach (viz the traditional approach or the broader Community approach) to apply. This is not because the same outcome would have been engendered whichever approach I took (which was the case in City Chain), but because I find that there was trade mark use in the present case. This is sufficient to pass the threshold test under the more stringent traditional approach, and a fortiori, also pass the test under the broader Community approach. I now elaborate on my reasons for finding that there was trade mark use.
Was there trade mark use in the present case?
[35] As a preliminary matter, the Plaintiff argued that the Defendant had referred to the “Nutello” sign as a “mark” in its pleadings, and had used the same term (viz, “mark”) to refer to the Plaintiff’s “Nutella” trade marks. To my mind, this is not determinative of whether or not “Nutello” was used as a trade mark. Indeed, I note that under the previous edition of the Trade Marks Act (viz, Trade Marks Act (Cap 332, 1998 Ed)), the term “mark” was statutorily defined as:
This definition of “mark” under the previous TMA is similar to the definition of the term “sign” under the present version of the TMA (on which see [5] above). Therefore, the use of the term “mark” would arguably be more closely linked with the use of the term “sign” under current legislation, than with the term “trade mark”. Indeed, as counsel for the Defendant was quick to point out, nowhere in the Defendant’s pleadings did the Defendant refer to the “Nutello” sign as a “trade mark”. I therefore think that the fact that counsel for the Defendant had referred to “Nutello” as a mark did not indicate that the Defendant was making an admission that “Nutello” had been used in a trade mark sense.
[36] That said, however, I do not think that the Defendant should be permitted to argue that “Nutello” was not used in a trade mark sense, because the Defendant had omitted to plead the same. Indeed, as was openly conceded by counsel for the Defendant at trial:
[37] The absence of such a pleading, and the failure to apply to amend the pleadings in this regard, suggests to me that the current contention – that the “Nutello” sign was not used as a trade mark – was an afterthought. It was clear that even during the trial, counsel for the Defendant maintained that the defence was not that the “Nutello” sign was being used in a non-trade mark sense. Parties should be bound by their pleadings, and accordingly, the Defendant should not be permitted to raise, after the conclusion of the trial, an issue that was neither pleaded nor raised at the trial.
[39] In any case, quite apart from the fact that the Defendant had, prior to the closing submissions, run its case as if “Nutello” was used in a trade mark sense, I am of the view that the “Nutello” sign was in fact used as a trade mark. The definition of a “trade mark” in s 2(1) of the TMA (reproduced at [6] above) requires that the allegedly infringing sign must be (a) capable of being represented graphically; and (b) capable of distinguishing goods and services dealt with or provided by a person from goods or services so dealt with or provided by any other person. In the present case, it is clear that the “Nutello” sign is capable of being represented graphically (see the graphical reproduction of the “Nutello” sign in [9] above). Furthermore, being an invented word, the “Nutello” sign is capable of distinguishing goods and services.
[40] With regard to this last-mentioned point (viz, that the “Nutello” sign is capable of distinguishing the Defendant’s products), I find that the “Nutello” sign was not only intended and used to identify the Defendant’s product, it was also perceived by members of the public as providing such identification. I am fortified in this observation for the reasons that follow.
[41] First, although Mr Christopher Tan did deny during cross-examination that “Nutello” was used as a trade mark, his denial should be looked at in context: it should be noted that Mr Christopher Tan also denied that the Defendant’s company had any trade marks, when as a matter of fact the Defendant had registered its trade mark “tcc” in class 30, registration number T0313398G on 5 September 2003. Accordingly, Mr Christopher Tan’s bare denials that the “Nutello” sign was used as a trade mark should be regarded with circumspection since, with respect, he did not appear to understand what a “trade mark” was.
[42] Indeed, it should be noted that notwithstanding the bare denials mentioned above, the views espoused by Mr Christopher Tan during cross-examination suggest that “Nutello” was being used as a trade mark. The relevant extracts of cross-examination are reproduced below for easy reference.
(a) He testified that everybody knew that the “Nutello” sign referred to the particular product in question, and that everybody knew that the product came from the Defendant’s company:
(b) He acknowledged that the “Nutello” sign was used by members of the public to indicate goods of the Defendant’s company:
(c) He confirmed that “Nutello” was the “name”, rather than a “description”, of the Defendant’s product:
[43] Second, it was clear from the Internet blogs adduced in evidence that bloggers had used the “Nutello” sign to refer to the Defendant’s product. In other words, the “Nutello” sign had been perceived by members of the public to designate the particular product produced by the Defendant. It is also evident from the Defendant’s pleadings that the “Nutello” sign was used in the course of trade, in relation to gourmet coffees which were sold in the Defendant’s exclusive TCC cafés.
[44] The Defendant attempted, in the reply submissions, to characterise “Nutello” as being descriptive, relying on a statement by Christopher Tan that the name “Nutello” was coined to indicate that the drinks had included the ingredients of espresso and the “Nutella” spread. However, for the reasons given above this was, in my view, too little too late.
[45] Therefore, I find that the use of the “Nutello” sign by the Defendant to label or mark this special drink at its TCC outlets and in its booklet, drinks menu and promotional materials (see [9]) so as to inform and identify for its customers and other members of the public that the drink labelled or marked as “Nutello” originates from the Defendant would amount to trade mark use. Such a finding concludes the threshold issue on both the traditional approach and the broader Community approach. Having crossed this threshold, I turn now to examine each of the three elements of trade mark infringement under s 27(2) of the TMA (see [21]).
Similarity of marks
[46] Similarity of marks is the first element to be considered. Whether the “Nutello” sign and the “Nutella” word mark are similar for the purposes of s 27(2)(b) of the TMA is a question of fact and degree for the court’s determination (Polo at [8]).
[47] In a recent case, the High Court in Ozone Community Corp v Advance Magazine Publishers Inc [2010] 2 SLR 459 (“Ozone Community”) adopted a two-step approach in analysing the similarity of marks, namely:
(a) Whether there is visual, aural and conceptual similarity between the allegedly infringing sign and the registered trade mark; and
(b) Whether the registered trade mark is distinctive.
[48] With regard to the former, it is trite that the court will consider whether there is visual, aural and conceptual similarity between the marks (MediaCorp News Pte Ltd v Astro All Asia Networks plc [2009] 4 SLR(R) 496 at [32] (“MediaCorp”)). It is also well established that the law does not require all three similarities to be made out before the finding that the marks are similar can be made (ibid). On the other hand, the fact that one similarity is made out does not necessarily mean that the marks are similar (ibid). The relative importance of each type of similarity will depend on the circumstances at hand, including the goods and types of marks involved (ibid).
[49] With regard to the latter, it is noted that the Ozone Community decision was made in the context of determining similarity for the purposes of registrability of trade marks, viz whether a sign was (un)registrable because it was similar to an earlier trade mark under s 8 of the TMA. However, it should be noted that the court in Ozone Community cited (at [45]) the Court of Appeal decision of Polo (a case dealing with trade mark infringement under s 27(2) of the TMA) in fleshing out the proposition that the distinctiveness of the registered trade mark is an important factor in the inquiry. Accordingly, the factor of distinctiveness of the registered trade mark should be considered in the present case as well.
[50] Before turning to undertake the two-step approach just mentioned, several legal principles guiding such a determination should be emphasised:
(a) First, the court considers the two signs/marks “as a whole” (Polo at [8]; City Chain at [47], [50]). However, the court would not take into account “any external added matter or circumstances” because the comparison is “mark for mark” (MediaCorp at [33], citing Caterpillar Inc v Ong Eng Peng [2006] 2 SLR(R) 669 at [55] (“Caterpillar”)).
(b) Second, the signs/marks are considered from the viewpoint of the average consumer – not an unthinking person in a hurry, but rather, a person who would exercise some care and good sense in making his purchases (Polo at [34]).
(c) Third, when comparing the two signs/marks, what is relevant is the “imperfect recollection” of the customer (MediaCorp at [33], citing Nautical Concept Pte Ltd v Jeffery Mark Richard [2007] 1 SLR(R) 1071 at [30]). The court will not compare the two marks side by side and examine them in detail, because “the person who is confused often makes comparison from memory removed in time and space from the marks” (MediaCorp at [33], citing Caterpillar at [55]).
Visual, Aural and Conceptual Similarity
(1) Visual similarity
[51] In the case of word marks, a determination of visual similarity typically involves looking at the (a) length of the marks; (b) structure of the marks (ie, whether there are the same number of words); and (c) whether the same letters are used in the marks (Ozone Community at [49], citing Bently & Sherman, Intellectual Property Law (Oxford University Press, 3rd Ed, 2009) at 865 (“Bently & Sherman”)).
[52] With respect to visual similarity, it is noted that both the “Nutella” mark and “Nutello” sign comprise 7 letters each, with the first 6 letters being identical and only the last being different. The type of similarity here alleged was considered in the case of Hyundai Mobis v Mobil Petroleum Company Inc [2007] SGIPOS 12 (Mobil (SGIPOS)), where the marks in question were as follows:
In that case, the Principal Assistant Registrar (“PAR”) held at [41]:
[53] On appeal, the High Court in Mobil Petroleum Co, Inc v Hyundai Mobis [2008] 4 SLR(R) 427 (“Mobil (HC)”) did not disagree with the decision of the PAR (Mobil (HC) at [18]). Indeed, the High Court agreed that there was visual similarity between the marks, although it noted that the degree of resemblance was limited to some extent by the differences in font, capitalisation and colours (ibid at [32]).
[54] The reasoning of the PAR, reproduced above at [52], applies to the present case. Here, both the “Nutella” word mark and “Nutello” sign are of identical length (viz, 7 letters), the first 6 letters of which are identical, with only the last letter (“a” contra “o”) being different. I note that the “Nutella” word mark consists of bold, capitalised, sans-serif letters, whereas the “Nutello” sign is stylised and in sentence case. However, as in the Mobil case, such differences may not be sufficient to render the marks visually dissimilar. Furthermore, taking into consideration the consumer’s “imperfect recollection” and that the marks are not to be compared side-by-side (MediaCorp at [34]), the marks are visually similar when viewed holistically.
[55] At this juncture, I address the Defendant’s argument that the “Nutella” word mark and “Nutello” sign bear different font, typeface and design. According to the Defendant, the “Nutella” word mark consists of plain letters in bold and capital letters, whereas the “Nutello” sign is stylised and in sentence case. This, the Defendant submits, constitutes visual dissimilarity between the two.
[56] This argument is misconceived. It is trite law that word-only registered trade marks render infringing any offending use in any font or stylisation. As stated by the learned authors of Kerly’s Law of Trade Marks and Trade Names (Sweet & Maxwell, 14th Ed, 2005) (“Kerly’s”) at para 2-036:
[57] The above passage from Kerly’s was cited with approval in Richemont International SA v Da Vinci Collections Pte Ltd [2006] 4 SLR(R) 369 (“Richemont v Da Vinci”)). The learned judge in Richemont v Da Vinci went on to quote from the English Court of Appeal case of Morny Ld’s Trade Marks, in the Matter of (1951) 68 RPC 131 (“Morny”), a passage which, for its relevance to the present case, justifies reproduction in full (Richemont v Da Vinci at [21]):
The learned judge went on to state (Richemont v Da Vinci at [32]):
The learned judge went on to hold that the use of the words “Da Vinci” in cursive form constituted an infringement of the registered trade mark “Da Vinci” in block capitals.
[58] The reasoning in Richemont v Da Vinci, Morny and Kerly’s is compelling. A block letter word mark is registered without any additional distinctiveness by virtue of stylisation by way of font type or graphical device. It would be illogical that the protection granted to a registered proprietor’s trade mark may be circumscribed simply by tweaking the font or style of the lettering of the word mark, especially where the registered proprietor had deliberately registered the word mark in block letters to represent that the distinctiveness of the mark lies in the word itself. It would also be impractical for the proprietor of a trade mark to have to register every possible stylistic permutation for his word mark in order to protect his trade mark, just as it would be impractical for him to have to make a fresh registration every time he wishes to use his word mark in a different font.
[59] Following from the above, the Defendant’s argument that the cursive font of the “Nutello” sign renders it visually dissimilar from the block capital registration of the “Nutella” word mark fails. In this regard, I would further note that besides the use of the cursive font used in the Defendant’s menu, the Defendant had also used the “Nutello” sign on its website and a booklet; these “Nutello” signs were not distinguished by the use of any special font whatsoever (see [9] above).
[60] In the final analysis, therefore, the “Nutello” sign is visually similar to the “Nutella” word mark.
(2) Aural similarity
[61] In the present case, the Plaintiff’s mark is “Nutella” and the Defendant’s mark is “Nutello”. The Plaintiff has confirmed that the correct pronunciation of “Nutella” is “nootella”, based on the testimony of the Plaintiff’s representative Mr Malatesta as well as the Plaintiff’s video advertisements aired in Singapore. This pronunciation is borne out of the Italian pronunciation of the word “nut”. To the credit of counsel for the Defendant, he pointed out that an article by Adnkronos Multimedia, featured on the Italian Institute for Foreign Trade’s website asserted that the name is pronounced “nutella” (emphasis added). The Defendant nonetheless argued that – relying on the Italian pronunciation put forward by the Plaintiff – the pronunciation of the Plaintiff’s mark is aurally dissimilar to the pronunciation of “Nutello” (which, the Plaintiff contends, begins with “nut” and stresses the last two syllables in the word “nutello”).
[62] The aural difference emphasised by the Defendant, however, is more apparent than real. First, it is important to consider the way that the average consumer in Singapore would pronounce the marks. In this regard, I refer to the Mobil case, where the learned PAR held that “Mobil” and “Mobis” were aurally similar. Comparing both marks, only the last consonant sound (at the ending of the second syllable) was different (Mobil (SGIPOS) at [39]). The PAR noted that although in the UK the same marks were considered not to be aurally similar, such a finding was based on the normal pronunciation of the marks in the UK, which did not necessarily apply to Singapore (Mobil (SGIPOS) at [39]). On appeal, the High Court in Mobil (HC) did not disagree that there was aural similarity in the two marks. Returning to the facts of the present case, two of the three syllables (“nut-te-[la/lo]”) will be pronounced the same way by the average consumer in Singapore. In this regard, notwithstanding the Plaintiff’s evidence that “Nutella” is more precisely pronounced (at least, by Italians) as “nootella”, it should be noted that in the course of cross-examination, Mr Christopher Tan read “Nutella” as “nutella” (as opposed to “nootella”). Mr Christopher Tan later confirmed that both he and his customers pronounced the first syllable as “nut”. It is therefore clear that the first two syllables of the word “Nutella” are pronounced in a manner identical to the first two syllables of the word “Nutello”. “Nutella” and “Nutello” are accordingly aurally similar.
[63] Second, and subject to the caveat above that pronunciation may differ between UK and Singapore, it is generally accepted that the first syllable of a word is usually the most important for the purpose of distinction between words because there is a tendency of persons using the English language to slur the termination of words, although this principle should not be applied in a “blanket fashion” (Valentino Globe BV v Pacific Rim Industries Inc [2009] 4 SLR(R) 577 at [34], citing London Lubricants (1920) Ltd’s Application (1925) 42 RPC 264 at 279). Accordingly, where the words compared terminate with vowels rather than consonants, there is arguably greater aural similarity between the words (Opposition by The Coca-Cola Company to the registration of trade mark application 731335 in the name of The Big Australian (Australia) Pty Limited (“Coca-Cola”) at 9). In the present case, where the difference between “Nutella” and “Nutello” is the final vowel sound (“a” contra “o”), the Hearing Officer’s observation in Coca-Cola is especially apposite:
I note that the beginning syllable sounds of marks may not always be more important than the ending syllables – there may be some cases where emphasis may be placed on some other syllable. For instance, in Mobil (SGIPOS), the PAR found that “Mobil” and “Mobis” were pronounced with emphasis on the second syllable (Mobil (SGIPOS) at [40]). Similarly, in Opposition by MCT UNILABELS SA to Trade Mark Application No 462838 in the Name of PETER KATHOLOS, the Hearing Officer suggested that the invented word “cappé” may be pronounced with an emphasis on the second syllable. In my view, however, “Nutella” and “Nutello” are probably pronounced with an emphasis on the first syllable, and the last syllable – being a vowel sound – is not sufficiently different to render the two words aurally dissimilar.
[64] Third, as was exhorted in Future Enterprises Pte Ltd v McDonald’s Corp [2006] 4 SLR(R) 629 (“Future Enterprises (HC)”) at [12], aural similarity must be considered in light of “imperfect recollection” and “careless pronunciation and speech”, rather than as if the marks were being pronounced carefully by someone such as a “teacher of elocution”. In Future Enterprises (HC), the court found that “MacCoffee” (pronounced “mac-kof-fee”) was aurally similar to “McCafe” (pronounced “mac-caf-fay” or “mc-kah-fay”). A fortiori, “Nutella” (pronounced “nut-te-la”) is clearly aurally similar to “Nutello” (pronounced “nut-te-lo”).
[65] I therefore find that the Defendant’s sign, “Nutello”, is aurally similar to the Plaintiff’s “Nutella” word mark.
(3) Conceptual similarity
[66] In considering whether there is conceptual similarity between marks, it is necessary to consider the ideas that lie behind or inform the earlier mark (Festina Lotus SA v Romanson Co Ltd [2010] 4 SLR 552 (“Festina”) at [38], citing Bently & Sherman at p 866).
[67] In Festina, the High Court considered several cases dealing with conceptual similarity. The illustrations provide some guidance on how conceptual similarity is analysed. The relevant paragraphs in Festina are reproduced here (Festina at [40]-[42]):
[68] On the facts of Festina itself, the court considered that (at [45]):
[69] Considering the three cases analysed in the quotation from Festina (reproduced at [67] above), the present case most closely resembles the facts in Festina itself: the “Nutella/Nutello” comparison is closer to the “Festina/Jestina” comparison, than it is with the “Bass/Pash”, “Came/Sweetcamel” and “Glamour/Hysteric Glamour” comparisons. Conceptually, both “Nutella” and “Nutello” are invented words, in that they are not English words found in dictionaries. The imperfect recollection of an average consumer, who does not juxtapose the two marks side-by-side, will probably consider the marks to be conceptually similar. Indeed, it should also be noted that Mr Christopher Tan admitted under cross-examination that the word “Nutello” was derived from the words “Nutella” and “espresso”; having been so derived, and with only a minor change being made to the word “Nutella”, it is difficult to see how “Nutella” and “Nutello” are conceptually dissimilar.
Distinctiveness
[70] As alluded to above, the relevance of the distinctiveness of the registered trade mark was discussed in Polo. The Court of Appeal held that the distinctiveness in the registered trade mark (ie, the “Nutella” word mark in this case) is an important factor in the inquiry into the similarity between the marks, stating that (at [23]):
The Court of Appeal went on to find that the word “polo” was an “ordinary English word meaning a game played on horseback”, and was accordingly “not an inventive word and could claim no inherent distinctiveness” (Polo at [23]).
[71] The factor of distinctiveness is important because it affects the question of whether marks are similar. As was noted in Bently & Sherman at pp 866-868 (and cited in Ozone Community at [44]):
[72] The policy motivating these principles was alluded to by the High Court in The Polo/Lauren Co, LP v Shop In Department Store Pte Ltd [2005] 4 SLR(R) 816 at [27] (“Polo (HC)”):
[73] In Festina, in the context of registrability of trade marks under s 8 of the TMA, the High Court noted that invented words such as “FESTINA” possess inherent distinctiveness, and such distinctive quality would be taken into account in the overall assessment of similarity between the marks (Festina at [58]). The court further held that for such invented words with inherent distinctiveness, a “modification of at least a substantial degree has to be done for the Respondent’s mark to be found to be dissimilar to it” (Festina at [60], emphasis added). The court went on to find that the respondent’s mark (“J.ESTINA”) was similar to the registered “FESTINA” mark as the latter was an inherently distinctive invented word.
[74] It is undisputed that the word “Nutella” has no dictionary meaning. It is an invented word which is inherently distinctive. Merely changing the last letter of the mark from an “a” to an “o” is insufficient, as such a change does not amount to “modification of at least a substantial degree” (Festina at [60]).
Concluding analysis: the marks are similar
[75] As demonstrated above, there is visual, aural and conceptual similarity in the marks “Nutella” and “Nutello”. The “Nutella” mark, being an invented word, is inherently distinctive, and the Defendant has not made sufficient modification to it such as to render “Nutello” dissimilar to “Nutella”. In the final analysis, I find that the marks are similar, and accordingly, that the first element of s 27(2)(b) of the TMA is made out. I would here add that although the similarity of the marks is an objective inquiry to be undertaken by the court (rather than by witnesses), I am fortified in my view that the marks are similar by the fact that at least two of the Defendant’s witnesses, viz Mr Christopher Tan and Mr Robin Ng, agreed to the same under cross-examination.
Identity or Similarity of goods
[76] In order for the requirement of identity or similarity of goods to be fulfilled, the Plaintiff must show that the “Nutello” sign is used in relation to goods that are identical or similar to those for which the “Nutella” word mark is registered (see s 27(2)(b) of the TMA, reproduced at [19] above). In other words, the Plaintiff must show that the “Nutello” sign is used in relation to goods identical or similar to “[c]onfectionery, baking powder, chocolate products, cream comprising cocoa with or without other ingredients” (see [3] above).
[77] The identity or similarity of goods is to be determined by taking into account all relevant factors, including the nature of the goods, end users, method of use and whether they are in competition or complementary (Canon Kabushiki Kaisha v Metro Goldwyn Mayer Inc [1999] ETMR 1 at [23]).
Are the goods identical?
[78] In British Sugar plc v James Robertson & Sons Ltd [1996] RPC 281 (“British Sugar”), Jacob J formulated a set of guidelines to assist the court in assessing the similarity of goods (the guidelines are set out in [83] below). For the avoidance of doubt, it should be emphasised that these factors are considered only in situations where the goods are being analysed for similarity. In other words, once it has been determined that the goods are identical from a comparison of the goods’ specifications, there is no need to consider the said guidelines (Richemont International SA v Goldlion Enterprise (Singapore) Pte Ltd [2006] 1 SLR(R) 401 (“Richemont v Goldlion”) at [18]).
[79] The Plaintiff argued in closing submissions that the “Nutello” sign was used on goods (viz, the Defendant’s “Nutello” drink) which fall within the broad specification of “chocolate products” (for which the “Nutella” mark is registered – see [3] above). Therefore, the Plaintiff argues that the “Nutello” sign is used in relation to goods identical to those for which the “Nutella” mark is registered.
[80] This argument depends on the Plaintiff being able to show that the Defendant’s product fell within the specification of “chocolate products”. To this end, the Plaintiff argued that the Defendant’s product was a chocolate product for the following reasons:
(a) First, the main ingredient in the Defendant’s “Nutello” drink is the use of the Plaintiff’s “Nutella” chocolate spread. The “Nutello” drink was promoted as being “[e]spresso with lashings of Nutella” (emphasis added), the word “lashings” being defined as “a copious amount of something, especially food or drink” (as defined by The New Oxford Dictionary of English (Oxford University Press, 1998) at p 1038);
(b) Second, the product is marketed by the Defendant in the Defendant’s menu as being “[p]erfect for cocoa lovers” (emphasis added). It should be noted that Mr Christopher Tan agreed under cross-examination that the Defendant attempted to attract chocolate lovers when promoting the “Nutello” drink.
(c) Third, the manner in which the “Nutello” drink is served shows that it is a chocolate product. In cross-examination, Mr Christopher Tan confirmed that the Defendant’s customers could, when drinking the “Nutello” drink, either scoop up the portions of the “Nutella” chocolate spread which was not dissolved in the espresso, or scoop out the “Nutella” chocolate spread before it dissolved so as to consume the chocolate spread directly. The manner of consumption of the “Nutello” drink, viz where the consumer could actually consume the chocolate spread directly, was confirmed by the Plaintiff’s witness Mr Ramakuddy in oral testimony. It was also confirmed by other consumers of the “Nutello” drink in their Internet blogs. For example, one customer noted that “[T]hey provide a small spoon for you to nibble and scrape the chocolate…”, another noted “[w]hen you put your spoon in, you will haul out a spoonful of gooey Nutella mixed with rich coffee”, and yet another noted that the drink was “a single e[s]presso shot with lotsa [sic] [N]utella in it”.
(d) Fourth, the Defendant’s customers perceive the “Nutello” drink to be a chocolate product. The Plaintiff attempted to substantiate this contention from two different sources.
(i) In the market survey conducted by the Defendant, when survey respondents were asked regarding the most likely reason for them to purchase the “Nutello” drink, most (viz, 30%) of the respondents indicated that they liked “the taste of Nutella chocolate” (emphasis added). A further 24.8% responded that they wanted to “try new drinks/new flavour/new taste of Nutella flavour” (emphasis added), while 16% said that they would purchase the “Nutello” drink because they “[l]ike[d] the taste of chocolate” (emphasis added). It is notable that, of the respondents who would not purchase the “Nutello” drink, the largest percentage (viz, 5.2%) indicated that they would not purchase the drink because they “do not drink chocolate drinks” (emphasis added). Analysing the survey response, the survey report concluded that with regard to the “Nutello” drink, most of the surveyed respondents provided “mostly positive [feedback] with people being attracted to chocolate and specifically the Nutella ingredient. Some would reject the drink for being too sweet or for not liking chocolate drinks” (emphasis added). It should be noted that most of the survey respondents were attracted by the chocolate character of the “Nutello” drink. It is also significant that most of the survey respondents who would not purchase the drink were repelled by the drink’s chocolate character.
(ii) In addition, perusal of Internet blogs (which were, incidentally, adduced in evidence by the Defendant rather than the Plaintiff) reveals that many of the consumers of the “Nutello” drink emphasised the chocolate content of the drink. For example, one customer noted that “Nutello” was a shot of espresso with “nutella smeared all over… I[f] you are a chocolate lover, this is the drink for you…” (emphasis added), another noted “[i]f you like chocolate, try Nutello!” (emphasis added) and yet another complimented the “thick layer of Nutella… [with] fragrant espresso and a layer of whipped cream, finally dusted with cocoa powder” (emphasis added).
[81] Despite the rigorous arguments just outlined, I am ultimately not convinced that the “Nutello” drink is a “chocolate product”.
(a) First, although the “Nutello” drink was recommended for “chocolate lovers” (see [80(b)] above), it was advertised under the “Espresso Specialties” section of the Defendant’s menu.
(b) Second, the fact that one of the ingredients in the “Nutello” drink is a chocolate spread does not per se make the drink a chocolate product. I note that it is possible for a product to fall within one or more categories. However, I do not think that the “Nutello” drink is a “chocolate product” simply by virtue of the fact that there was “Nutella” chocolate spread in it. To do so would be akin to describing coffee drinks as milk products (by virtue of milk being one of the ingredients) and describing certain teas as fruit products (by virtue of dried fruits being one of the ingredients).
(c) Third, the Defendant rightly pointed out that Class 30 of the International Classification of Goods and Services (under which the Plaintiff’s “Nutella” mark was registered) comprises a list of specific goods indications, which include categories of goods such as “Beverages (Chocolate-based)”, “Beverages (Cocoa-based)”, “Chocolate beverages with milk”, “Chocolate-based beverages”, etc. The Plaintiff could have, but did not or chose not to, include any beverage-related specifications in the registration of the “Nutella” mark. The specification of “chocolate products” thus does not appear to me to have been intended to cover all beverages with some chocolate content.
[82] I turn next to consider whether the “Nutello” sign is used in relation to goods similar to those for which the “Nutella” mark is registered.
Are the goods similar?
[83] As alluded to above at [78], Jacob J in British Sugar formulated a set of guidelines to assist the court in assessing the similarity of goods. The factors laid out in British Sugar were adopted by Singapore courts (see, eg, Caterpillar at [57]; MediaCorp at [41] and Ozone Community at [75]). The factors are as follows (British Sugar at 296-297):
[84] Relying on the above factors, the Defendant argued that:
(a) First, the uses of the respective goods are different. The Plaintiff’s “Nutella” product is not a beverage but a bread spread, is not usually consumed by itself but in conjunction with bread, and is usually consumed during breakfast. In contrast, the Defendant’s “Nutello” drink is used in relation to liquid gourmet coffee beverage, and is used for purposes distinctly different from those of the Plaintiff’s semi-solid, semi-liquid viscous cream spread.
(b) Second, the users of the respective goods are different. The users of the Plaintiff’s “Nutella” product are mainly ordinary shoppers looking for bread spreads for breakfast for their families, while the users of the Defendant’s “Nutello” drink are mainly discerning and well-informed professionals, managers, executives and business people.
(c) Third, the physical nature of the respective goods is different. The Defendant’s “Nutello” drink is a concoction comprising a number of ingredients, and is served in a shot glass rather than a jar. On the other hand, the Plaintiff’s product is a dark brown, viscous cream spread presented in a jar.
(d) Fourth, the trade channels of the respective goods are different. The Plaintiff’s “Nutella” product is sold in supermarkets, convenience stores and neighbourhood shops, on shelves alongside jams and breads. In contrast, the Defendant’s “Nutello” drink is promoted and sold exclusively at TCC outlets in a restaurant setting.
(e) Fifth, the respective goods are neither competitive with, nor substitutes for, one another. The Plaintiff’s goods compete with other brands of bread spreads, while the Defendant’s goods compete with other coffee companies and cafés.
[85] In addition, the Defendant pointed out that in British Sugar, Jacob J also held that in construing a word used in a trade mark specification (in the present case, “chocolate products”), the court is (British Sugar at 289):
Relying on this observation by Jacob J, the Defendant argued that a product specification must be construed in accordance with its use in trade. Accordingly, the Defendant argues that the expression “chocolate products” (in the product specification under which the “Nutella” word mark is registered) must be construed in accordance with the Plaintiff’s products as used in practice, viz, as a sweet bread spread rather than a beverage.
[86] The arguments raised by the Defendant have some weight, and do appear to suggest that the respective goods are not similar under the British Sugar guidelines. However, in the final analysis, I find that the Defendant’s “Nutello” drink is similar to the “chocolate products” specification for which the “Nutella” word mark is registered. As a matter of law, in determining the similarity of goods, the comparison is made not only with the actual product in use (viz the Plaintiff’s sweet bread spread) but with the type of products for which the Plaintiff has registered its word mark “Nutella”. In this case, the comparison has to be with the possible range of “chocolate products” as used in the trade and not only with the Plaintiff’s sweet bread spread as such.
[87] It is also important to keep in mind that the British Sugar factors “must not be regarded as requirements that must all be satisfied before the goods can be treated as similar” (Johnson & Johnson v Uni-Charm Kabushiki Kaisha (Uni-Charm Corp) [2007] 1 SLR(R) 1082 at [18]; Mobil (HC) at [25]). Indeed, Jacob J himself cautioned that he did not think that the list “can provide other than general guidance” (British Sugar at 297).
[88] Second, it should be noted that in British Sugar the goods in question fell into different classes of product specification – the proprietor’s mark was registered in Class 30, while the alleged infringer’s products fell into Class 29. As was noted in Polo (HC) at [33]:
In other words, the British Sugar guidelines may not have as much weight (if any at all) in a situation where the respective products fall within the same class of products. On the present facts, the Defendant had pointed out (as alluded to earlier in [81(c)] above), and accordingly must be taken to concede, that the Defendant’s “Nutello” drink was likely to fall within the categories of “beverages”. Such beverage-related specifications are still within the Class 30 classification. I also take cognisance of the arguments addressed in [80] above, albeit in the context of assessing whether the Defendant’s “Nutello” drink was a “chocolate product”.
[89] Third, the High Court in Mobil (HC) noted at [25] that an “important” factor to be considered in determining whether the respective goods are similar is “whether such goods will be considered by consumers to be similar goods”. In this regard, the point made in [80(d)] above demonstrates that the consumers do indeed consider the respective goods to be similar.
[90] For the reasons stated, I find that the “Nutello” drink is similar to a chocolate product although it may not be regarded as a chocolate product. Accordingly, the second element of s 27(2)(b) of the TMA is made out.
Whether there exists a likelihood of confusion on the part of the public
[91] The third element of s 27(2)(b) of the TMA is the issue of the likelihood of confusion on the part of the public. Before proceeding to apply the law to the facts of the present case, I will summarise the relevant legal principles governing this particular issue.
The law on “likelihood of confusion”
[92] The issue of likelihood of confusion is to be determined as at the time when the alleged infringing use of the sign commenced (City Chain at [51]), the test being whether a substantial portion of the relevant public will be confused. The three italicised portions indicate three different aspects of the test. I shall address them in the following order: (a) the meaning and nature of the “confusion” required by s 27(2)(b); (b) the meaning and nature of the “relevant public”; and (c) the requirement of a “substantial portion” of the relevant public being confused.
[93] With regard to the meaning and nature of the “confusion” required, the Court of Appeal in City Chain noted that in determining confusion, one should be cognisant of the risk that the public might believe that the goods come from the same source or economically-linked sources (see City Chain at [52]). In other words, the relevant confusion is not limited to the “classic confusion” scenario where consumers believe that the goods or services emanate from a particular origin (when they in fact come from another origin). There is instead a “broader” kind of confusion stemming from the consumer’s (incorrect) assumption of some kind of economic connection between the users of the marks (eg that the goods are being provided by a subsidiary or licensee of the trade mark proprietor) (see Bently & Sherman at p 872).
[94] It should further be noted that, by virtue of the wording of s 27(2) of the TMA, the likelihood of confusion must arise because of the similarities in the respective marks and goods concerned. It is insufficient that the confusion arises for any other reason. For instance, it would clearly not be sufficient if a member of the public appears confused as to the origin of the “Nutello” drink, such confusion stemming from his incapability of understanding the survey questions directed at him (eg his inability to understand what it means to “produce” a drink, or to “licence / authorise” a drink, etc). This has important implications on the framing of survey questions, and will be dealt with in greater detail at [136] et seq.
[95] With regard to the meaning and nature of the “relevant public”, it is trite that the issue of likelihood of confusion is to be determined by reference to average consumers of the goods and services in question. Such consumers are reasonably well-informed, reasonably observant and circumspect, and will exercise ordinary care and intelligence. They are also literate, educated, “constantly exposed to the world, either through travel or media” and are unlikely to be “easily deceived or hoodwinked” (McDonald’s Corp v Future Enterprises Pte Ltd [2005] 1 SLR(R) 177 at [64]).
[96] With regard to the requirement that a “substantial portion” of the relevant public must be confused, it is insufficient that only a “single member of the public”, or “a very small unobservant section of society”, or “a moron in a hurry”, would be confused (Mobil Petroleum Co Inc v Hyundai Mobis [2010] 1 SLR 512 (“Mobil (CA)”) at [79], citing Tong Guan Food Products Pte Ltd v Hoe Huat Hng Foodstuff Pte Ltd [1991] 1 SLR(R) 903 (“Tong Guan”) at [24]). The degree of confusion must be taken above a “de minimis level”, and while there does not have to be a “majority” of the relevant public that is confused, there must be a “not insubstantial number” (Mobil (CA) at [77]-[78]).
[97] I turn now to other relevant principles vis-à-vis the test for likelihood of confusion. First, it is now well-settled that the issue of likelihood of confusion is to be addressed globally, taking into account all the circumstances of the case. This would include, inter alia, the closeness of the goods, the impression given by the marks, the possibility of imperfect recollection and the risk that the public might believe that the goods come from the same source or economically-linked sources (City Chain at [52], Polo (CA) at [28]).
[98] Second, following from the phrasing of s 27(2)(b) of the TMA, the requirement of confusion is not automatically made out just because the marks and the respective goods on which they are used are similar (Polo (CA) at [25]). In other words, there is no presumption that confusion arises where the marks and goods are similar (Polo (CA) at [8]). However, it is also clear from s 27(2)(b) that the likelihood of confusion which is relevant is the likelihood of confusion which is caused by the similarity of the respective marks and goods to each other. Accordingly, it stands to reason that where there is greater similarity in the marks and the goods, it would be easier to find a likelihood of confusion (Mobil (CA)); Future Enterprises (HC) at [22]). For the same reason, where the earlier mark is distinctive, it is more likely that a court would find that a likelihood of confusion exists (see Sabel BV v Puma AG, Rudolf Dassler Sport [1998] RPC 199 at 224).
[99] Third, it should be noted that steps taken by the alleged infringer to differentiate his goods from those of the registered proprietor are also relevant (City Chain at [53], Polo (CA) at [28]). Where the alleged infringer has taken pains to distinguish his products from those of the registered proprietor, the likelihood of confusion may turn out to be “merely hypothetical or speculative” (City Chain at [53], citing Polo (HC) at [21]).
[100] Finally, the “mere association” of the public between the two marks based on their similar use is not in itself a sufficient basis for concluding that there is a likelihood of confusion in the absence of any possibility of misapprehension as to the origin of the goods or services (City Chain at [58]). I emphasise that this holding is not inconsistent with the point made in [93] above with regard to the “broader” kind of confusion stemming from the consumer’s (incorrect) assumption of some kind of economic connection between the users of the marks. The “economic connection” point relates to the origin of the goods or services, viz, the (erroneous) belief that the goods are originating from a licensee of the registered proprietor’s goods. This must be contrasted with the “mere association” point, which relates to – as the phrase itself suggests – a mere association, without any confusion as to origin.
[101] Before turning to apply the law to the facts of the present case, it should be noted that the question as to whether consumers are likely to be confused is an issue to be determined by the court. The court does not, and cannot, abdicate this decision to witnesses (Payton & Co, Limited v Snelling, Lampard & Co Limited [1901] AC 308 at 311, Electrolux Limited v Electrix Limited (1954) 71 RPC 23 at 31 per Evershed MR).
Application to the facts
[102] For the purposes of the present case, the relevant date is 1 August 2007, which was the time when sale of the “Nutello” drink in Singapore commenced (see also [5] and [92] above, as well as [199] below).
[103] In many of the cases cited before me, the issue of likelihood of confusion was often determined without the help of direct survey evidence on the point (see, for instance, City Chain at [57]). In the present case, however, separate market surveys were done by the Plaintiff and the Defendant. There are, however, notorious difficulties with the probative value of market surveys, and I pause to note that this case raised interesting questions about the design of surveys in ascertaining whether there is a likelihood of confusion on the part of the public. At appropriate junctures in the analysis that follows, I will evaluate the design of the said surveys. I shall consider the Defendant’s survey first, followed by the Plaintiff’s, before analysing whether the element of likelihood of confusion is made out in the present case.
(1) The Defendant’s survey
[104] The Defendant engaged a market research company, Kadence International Pte Ltd, to conduct a survey on the question of whether the use of the “Nutello” sign in relation to the Defendant’s product would cause a likelihood of confusion. The Defendant’s expert witness, Mr Piers Lee, testified as to the truth and accuracy of the survey. The survey was an exit survey conducted with the Defendant’s customers at 20 different TCC outlets in Singapore. The sample size of the survey was 500. The results of the Defendant’s survey showed that 27% of the Defendant’s consumers who were surveyed believed that the “Nutello” drink was produced by the manufacturer of the “Nutella” chocolate spread (viz, the Plaintiff), and that another 3% were of the opinion that “Nutello” was produced by both the Defendant and the Plaintiff. An additional 12% of the respondents answered that they “can’t say” who produced the “Nutello” drink, while 58% thought that the Defendant produced the “Nutello” drink.
[105] According to the Defendant’s survey, therefore, at least 30% of the consumers would be confused into thinking that the “Nutello” product was either produced by the Plaintiff, or the Plaintiff in association with the Defendant. This is, the Plaintiff alleges, a substantial number of the relevant public who would be confused as to the origin of “Nutello” product.
[106] Indeed, the Plaintiff goes on to argue that had the Defendant’s survey been carried out properly, it would have shown that more than 30% of the relevant public would be confused. Five arguments are raised to substantiate this claim.
(a) First, the sequence in which the questions were asked and the manner in which the questions were framed attempted to “teach” or “lead” the survey respondents to the desired answer that it must be the Defendant which produced the “Nutello” drink at its TCC outlet and not the Plaintiff.
(b) Second, in the most crucial question where the survey respondents were asked regarding who they thought the producer of the “Nutello” drink was, the respondents were forced to make a choice between the Plaintiff and the Defendant, without any option of indicating that both the Plaintiff and the Defendant produced the drink (see Question 8, reproduced in [109] below).
(c) Third, the relevant question is not whether the survey respondents thought that the Plaintiff produced the “Nutello” drink; instead, the relevant question is whether the respondents thought that the Plaintiff produced or licensed the “Nutello” drink.
(d) Fourth, the Defendant’s expert admitted that if more facts or time had been given, some of the 12% who indicated that they “can’t say” could have said that the “Nutello” drink was either produced by the Plaintiff, or the Plaintiff in association with the Defendant.
(e) Fifth, the Defendant’s survey was conducted on the Defendant’s customers as an exit survey, viz, after they had patronised the Defendant’s outlet and as they were leaving it. The Defendant had consistently characterised its customers as being well-educated and discerning. It is therefore possible that had the survey been conducted on the general public, the proportion of the persons who would have been confused might have been higher.
[107] Although the Defendant attempted, in Reply Submissions, to dismiss the abovementioned arguments as being merely “speculative”, there is, in my view, much merit in the Plaintiff’s arguments. I limit my elaboration to the points raised in sub-paragraphs (a), (b) and (c).
[108] With regard to sub-paragraph (a) (viz, the framing of the questions to “teach” or “lead” the respondents to the correct answer), I am reminded of the caution expressed by Whitford J in Imperial Group Ltd v Philip Morris & Co [1984] RPC 293 (“Imperial Group”) at 303, that “[g]reat importance inevitably attaches to the way in which the questions are cast” (emphasis added). The framing of questions, however, is not (and should not be taken to be) an easy task. Indeed, Whitford J opined that it was “very difficult in connection with [the particular market survey in Imperial Group] to think of questions which, even if they are free from the objection of being leading, are not in fact going to direct the person answering the question into a field of speculation upon which that person would never have embarked had the question not been put” (emphasis added).
[109] Unfortunately, and with respect, the Defendant’s survey fell foul of Whitford J’s specific exhortation against leading questions. It is useful to illustrate this with reference to the actual survey questions used in the Defendant’s survey. This should be taken as a message that utmost care should be taken in framing survey questions, so as to avoid “teaching” or “leading” the survey respondents which may affect the objectivity of the survey and render the survey results less reliable. In brief, there are two questions (viz, Questions 6 and 7) leading to the crucial question (viz Question 8) which addresses the issue of who – in the mind of the survey respondent – produced the “Nutello” drink. All three questions are set out in full, in the order in which they were asked:
[110] Question 6, which purports to determine whether the survey respondents are aware that coffee chains buy ingredients of “their” (emphasis in original) drinks from manufacturers, is not relevant at all to the issue of whether the relevant public would be likely to be confused vis-à-vis the origin of the “Nutello” drink. In my view, it appears that the leading nature of Question 6 laid the ground work and injected new information into the minds of the survey respondents, preparing them for the questions to follow. It subtly teaches the survey respondents that coffee chains do buy ingredients of “their” drinks from the manufacturers of the ingredients. Furthermore, the emphasis on the possessive term “their” may have created the impression that the drinks were produced by the coffee chain – a subtle but nonetheless possibly significant suggestion that the Defendant, being the “coffee chain”, produced their own drinks.
[111] Although Question 7 is framed as a hypothetical, nevertheless it does inject new information into the minds of the respondents. Not only does it repeat the point that coffee chains buy the ingredients of their drinks from “other sources” (emphasis in original), it attempts to clarify that coffee chains still “make[] the drink in their coffee chain” using the ingredients they had bought (emphasis added). In cross-examination, the Defendant’s expert witness, Mr Piers Lee, admitted that Question 7 had indeed injected these two pieces of information. Once again, it appears to be a subtle but nonetheless possibly significant suggestion that the Defendant, being the “coffee chain”, made the drinks and not the manufacturer of the ingredients.

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

Ferrero SPA v Sarika Connoisseur Cafe Pte Ltd

Grounds of decision

Factual background

The Plaintiff’s Claims

The trade mark infringement claim (s 27(2)(b) of the TMA)

The threshold issue: Must there be trade mark use?

The meaning of “trade mark use”

Whether trade mark use is necessary

Was there trade mark use in the present case?

Similarity of marks

Visual, Aural and Conceptual Similarity

(1) Visual similarity

(2) Aural similarity

(3) Conceptual similarity

Distinctiveness

Concluding analysis: the marks are similar

Identity or Similarity of goods

Are the goods identical?

Are the goods similar?

Whether there exists a likelihood of confusion on the part of the public

The law on “likelihood of confusion”

Application to the facts

(1) The Defendant’s survey