1 This was an appeal by Dr. August Wolff GmbH & Co. KG Arzneimittel (the “plaintiff”) against the decision of the learned Principal Assistant Registrar (“PAR”) allowing the application of Combe International Ltd (the “defendant”) for a declaration of invalidity against the registration of Singapore Trade Mark Registration No. T1206670G “VAGISAN” in Class 3 (“Soaps, perfumery, essential oils, cosmetics, hair lotions”) and Class 5 (“Pharmaceutical products, sanitary products for medical purposes; dietetic substances for medical purposes”). The PAR’s reasons are stated in Combe International Ltd v Dr. August Wolff GmbH & Co. KG Arzneimittel [2020] SGIPOS 3 (the “GD”). Having heard and considered the parties’ submissions, I dismissed the appeal. These are my reasons.

2 The plaintiff is part of a family business, the Dr Wolff Group, that was founded in 1905 and named after its founder Dr August Wolff. The plaintiff researches, manufactures and markets dermatological, medicinal and cosmetic products that are distributed in many countries. ¹ In 1994, the Dr Wolff Group conceived the name “VAGISAN” for use in connection with intimate care products. ² According to the plaintiff, “VAGISAN” was conceptualised to be a portmanteau of the English word “vagina” and the Latin word “sanitas” (which means “health”). ³ The plaintiff’s intimate care products, which are now sold in many countries, first bore the “VAGISAN” Mark in 1998. ⁴ The plaintiff registered the “VAGISAN” Mark in Singapore on 19 March 2012. ⁵ Despite intentions to commence sales of its product here, the plaintiff put a halt to the plans pending the determination of the trade mark disputes in several countries, including the present dispute in Singapore. ⁶

3 The defendant is a global personal care company founded in 1949 by its namesake, Ivan Combe. ⁷ Its market strategy focuses on advertising and promoting its in-house brands without use of the company name “Combe”. For example, the defendant has developed and currently markets products such as “JUST FOR MEN” hair care, “SEABOND” denture care, and “AQUA VELVA”, “BRYLCREEM” and “JOHNSON’S FOOT SOAP” men’s grooming products. ⁸

4 “VAGISIL” products for feminine care are among one of the defendant’s key brands. ⁹ The defendant first used the “VAGISIL” Mark in Singapore in 1990. ¹⁰ It is the registered owner of these “VAGISIL” trade marks in Singapore: ¹¹

5 Presently, the defendant’s “VAGISIL” products are available at Singapore retailers such as NTUC FairPrice, Cold Storage, Watson’s, Guardian, Unity Pharmacy, Venus Beauty, Pink Beauty and Tomato. ¹² The products are also available on online retailers such as Lazada, Qoo10, Shopee, Redmart, Amazon Prime Now, Daily Mart, Tan Tock Seng Hospital pharmacy, Splendideals.sg, Cloversoft & Co., EA Mart SG, Corlison and the VAGISIL e-store. ¹³

6 The defendant provided its trade promotion and advertising expenses between 2011 and June 2018 for goods bearing the “VAGISIL” Marks in Singapore: ¹⁴ at [12] of the GD. The defendant also furnished its unit sales in Singapore between 2011 and June 2018 for goods bearing the “VAGISIL” Marks, based on sales by its distributor Corlison to Singapore retailers: ¹⁵ at [13] of the GD. The defendant’s products sold under “VAGISIL” Marks have been referenced in certain television shows, ¹⁶ and the “VAGISIL” Marks have also been referred to in certain literary works: ¹⁷ at [15]–[16] of the GD.

7 On 14 November 2017, the defendant applied to declare invalid the registration of the plaintiff’s “VAGISAN” Mark pursuant to the following three grounds:

(a) Section 23(3)(a)(i) read with s 8(2)(b) of the Trade Marks Act (Cap 332, 2005 Rev Ed) (“the Act”). The first ground provides for the invalidation of the registration of a trade mark which is “similar to an earlier trade mark”, “registered for goods or services identical with or similar to those for which the earlier trade mark is protected”, and “there exists a likelihood of confusion on the part of the public”.

(b) Section 23(3)(b) read with s 8(7)(a) of the Act. The second ground relates to invalidation of the registration of a trade mark to protect rights acquired under “the law of passing off” by any “unregistered trade mark or other sign used in the course of trade”; and

(c) Section 23(3)(a)(iii) read with s 8(4)(b)(i) of the Act. The third ground provides for invalidation of the registration of a trade mark to accord protection to well-known marks.

8 For the first ground, the defendant relied on all its registered trade marks except for Trade Mark No 40201507713Y, which is not a relevant mark given that its application date of 5 December 2014 is after 19 March 2012 ie, the date of registration of the “VAGISAN” Mark. The four other trade marks shall collectively be referred to as “the “VAGISIL Marks””.

9 I pause to observe that the parties are currently embroiled in similar disputes in several jurisdictions around the world. In Australia, the Registrar of Trade Marks refused registration of the “VAGISAN” Mark, but the plaintiff succeeded in its appeal before the Federal Court of Australia: see Dr. August Wolff GmbH & Co. KG Arzneimittel v Combe International Ltd [2020] FCA 39 (“August Wolff (AU)”). The defendant has appealed against this decision. In New Zealand, the defendant successfully opposed the registration of the “VAGISAN” Mark before the New Zealand Intellectual Property Office: see Dr. August Wolff GmbH & Co. KG Arzneimittel v Combe International Ltd [2018] NZIPOTM 10. The plaintiff’s appeal before the New Zealand High Court was dismissed: see Dr. August Wolff GmbH & Co. KG Arzneimittel v Combe International Limited [2020] NZHC 1679 (“August Wolff (NZ)”). In the US, the Trademark Trial and Appeal Board (“TTAB”) dismissed the defendant’s opposition, but a US District Court reversed the TTAB’s decision and ruled in favour of the defendant: see Combe Inc v Dr. Aug Wolff GMBH & Co. 382 F. Supp. 3d 429 (E.D. Va. 2019). In the EU, the Cancellation Division allowed the defendant’s application and cancelled the plaintiff’s “VAGISAN” Mark: see Combe International Ltd. V Dr. August Wolff GmbH & Co. KG Arneimittel Cancellation No 18 101 C (“August Wolff (EU Cancellation)”). The plaintiff’s subsequent appeal to the Board of Appeal was dismissed: see Dr. August Wolff GmbH & Co. KG Arzneimittel v Combe International Ltd. R2459/2019-4 (“August Wolff (EU Appeal)”). The plaintiff has appealed against this decision.

10 Referring to s 101(c)(i) of the Act (which provides that the registration of a person as a proprietor of a registered trade mark shall be “prima facie evidence of … the validity of the original registration”), the PAR noted that the burden of proving the grounds fell on the defendant: at [6] of the GD.

11 The PAR allowed the defendant’s application for invalidation under the first ground set out at [7(a)] above. To summarise, the PAR held that the “VAGISIL” Marks have a normal level of inherent distinctive character: at [32] of the GD. The PAR found that the plaintiff’s “VAGISAN” Mark and the defendant’s “VAGISIL” Marks are visually and aurally similar to an above average degree but conceptually neutral: at [38], [46], [48] and [49] of the GD. Overall, she found that the marks are similar to an above average degree: at [51] of the GD. Also, the PAR found that the plaintiff’s goods are identical or similar to the defendant’s goods in Classes 3 and 5. This, in fact, was not seriously disputed by the plaintiff: at [52]–[54] of the GD. The PAR concluded that there is a likelihood of confusion (especially of what she termed as “indirect confusion”): at [71] of the GD. Therefore, the defendant succeeded on this ground of invalidation: at [72] of the GD.

12 In respect of the second ground set out at [7(b)] above, the PAR found that the plaintiff’s use of the “VAGISAN” Mark amounts to a misrepresentation which creates a likelihood of confusion and a real likelihood of damage to the defendant’s goodwill in the “VAGISIL” Marks. Thus, the defendant also succeeded on this ground of invalidation: at [91]–[98] of the GD.

13 As regards the third ground set out at [7(c)] above, the PAR found that the “VAGISAN” and “VAGISIL” marks are similar: at [74] of the GD. However, as the “VAGISIL” Marks were not well known in Singapore, this ground was not made out: at [83]–[84] of the GD.

14 Before me, the plaintiff contended, in respect of the first ground, that the PAR erred in holding that the marks “VAGISAN” and “VAGISIL” are visually similar to an above average degree, are aurally similar to an above average degree, are similar to an above average degree overall, and that there exists a likelihood of confusion. As for the second ground, the plaintiff argued that the PAR erred in finding that the elements of misrepresentation and damage are made out. While the plaintiff did not challenge the outcome of the third ground (which was in its favour), the plaintiff contended, as a matter of consistency with its arguments in relation to the first two grounds, that the PAR erred in finding that the marks are overall similar. The defendant, on the other hand, submitted that the PAR’s decision should be upheld. I shall expand on the parties’ arguments in my analysis below.

15 For the first ground, the relevant provisions state:

16 Under s 8(2)(b) of the Act, the analysis comprises a “step-by-step” approach. By this, I mean that the court is required to systematically assess the similarity of marks, the similarity of goods or services, and then the likelihood of confusion arising from the two similarities. Only after the court has determined the presence of the first two elements individually can it then ascertain the likelihood of confusion: Staywell Hospitality Group Pty Ltd v Starwood Hotels & Resorts Worldwide, Inc and another and another appeal [2014] 1 SLR 911 (“Staywell”) at [15]. This effectively bifurcates “the issue of resemblance between the competing marks … from the question of the effect of such resemblance”: Staywell at [20].

17 For the first step, the mark-similarity inquiry is “directed at assessing substantive similarity” of the contesting marks “as a whole”. It is a mark-for-mark comparison without consideration of any external matter. While it is not a mechanistic approach, the court considers three aspects of the marks being compared, namely visual, aural and conceptual similarity. The viewpoint is that of “the average consumer who would exercise some care and a measure of good sense in making his or her purchases” but also one who has “imperfect recollection”: Hai Tong Co (Pte) Ltd v Ventree Singapore Pte Ltd and another and another appeal [2013] 2 SLR 941 (“Hai Tong”) at [40].

18 Integrated into the analysis of visual, aural and conceptual similarity is a consideration of whether the earlier mark is distinctive (in both its technical and non-technical sense): Hai Tong at [26]; Staywell at [30]. In the non-technical sense, distinctiveness refers to “what is outstanding and memorable about the mark”. Such components tend to draw the average consumer’s attention and stand out in the average consumer’s imperfect recollection: Staywell at [23]. In the technical sense, distinctiveness ordinarily refers to a mark’s inherent capacity (ie, relating to a mark’s fanciful or descriptive nature) or acquired capacity (ie, based on the duration and nature of the use of the mark) to operate as a badge of origin: Hai Tong at [31]–[33]; Staywell at [24]. A mark which has greater technical distinctiveness enjoys a high threshold before a competing sign will be considered dissimilar to it: Staywell at [25].

19 While mindful that this is not a separate step within the marks-similarity inquiry, the PAR, in her analysis, highlighted this matter separately. Given that distinctiveness features in all aspects of the mark-similarity analysis, as a matter of practicality, I shall also start with this. I then proceed to the issues of visual and aural similarities (as parties accepted that the marks are conceptually neutral), before dealing with the overall similarity of the marks.

20 The plaintiff did not specifically challenge the PAR’s finding that the “VAGISIL” Marks possess a normal level of inherent distinctive character. However, certain aspects of her reasoning were emphasised before the plaintiff submitted on mark similarity.

21 To elaborate, the PAR was of the view that “VAGI” is of weak distinctive character. The distinctiveness of the defendant’s “VAGISIL” Marks lies in the combination of “VAGI” and “SIL”. Normally, invented words are considered to have a very high degree of distinctive character. However, the relevant public would likely recognise “VAGI” as a shortening of the word “vagina”, and the invented word as indicating the intended treatment area of the goods. Therefore, the inherent distinctiveness of the “VAGISIL” Marks is of a normal level. The PAR also noted that the defendant did not rely on acquired distinctiveness, and did not consider the matter: at [31]–[33] of the GD.

22 The plaintiff emphasised the PAR’s conclusion that “VAGI” has weak distinctive character and went as far as to submit that “VAGI” is descriptive. In particular, the plaintiff argued that “VAGI” is a direct reference to the products in question, ie, intimate care products, and “vagina” is a commonly used word in ordinary parlance. The prefix “VAGI” is also commonly used by other third parties in naming their intimate care products, including “VAGISTAT”, “VAGIFEM”, “VAGINNE”, “VAGICAL”, “VAGIKOOL”, “VAGICAINE”, “VAGIFIRM”, “VAGI-PAL”, “VAGI-SITZ”, “VAGI-CLEAR”, “VAGI TIGHT”, “VAGI DOC”, and “VAGI SNUG”. ¹⁸ Notably, products bearing the “VAGISTAT” and “VAGIFEM” marks have been available in the Singapore market since at least as early as 2013 and 2008 respectively. ¹⁹ “VAGINNE”, “VAGILUR”, and “VAGIFEM” are also marks registered in Singapore for products such as “vaginal cleaning liquids for personal sanitary or deodorizing purposes”, “gynaecological preparations”, and “pharmaceutical products”. ²⁰ Moreover, the plaintiff asserted that the PAR’s finding that “VAGI” has weak distinctive character coheres with findings of the courts or Trade Mark Offices in Australia, the EU and New Zealand, all of which have consistently found that “VAGI” is descriptive or a direct reference to intimate care products.

23 The defendant, on the other hand, emphasised the PAR’s findings that “the distinctiveness of the VAGISIL mark lies in the combination of ‘VAGI’ and ‘SIL’ and the fact that it is conjoined to form an invented word”, and that “VAGISIL” has at least a “normal level of inherent distinctive character”. According to the defendant, these findings are consistent with Hai Tong (concerning the “Lady Rose” mark), where the Court of Appeal held at [35] that while the words “Lady” and “Rose” were not distinctive words when used individually, “[i]t is the juxtaposition of the words that engenders distinctiveness, in that when used together, the words do not convey a sensible meaning”. Accordingly, the defendant argued that, notwithstanding that “VAGI” is recognised as a shortened form of “vagina”, the suffix “SIL” is distinctive and thus the “VAGISIL” mark as a whole is inventive and does not convey a sensible meaning. Reliance was also placed on Harrods’ Application (1935) 52 RPC 65 at 70 (cited in David Keeling et al, Kerly’s Law of Trade Marks and Trade Names (Sweet & Maxwell, 15th Ed, 2011) (“Kerly’s”) at para 18-215) that “where […] two marks contain a common element which is also contained in a number of other marks in use in the same market, such a common occurrence in the market tends to cause purchasers to pay more attention to the other features of the respective marks and to distinguish between them by those other features”.

24 The parties did not specifically call into question – and I saw no reason to disturb – the PAR’s finding that the “VAGISIL” Marks bear a normal level of inherent distinctiveness. I agreed with the PAR’s reasoning that the prefix “VAGI” is of weak distinctive character, and the invented word would be understood to denote the intended treatment area of the goods. However, the suffix “SIL” is distinctive. Further, the distinctiveness of the “VAGISIL” Marks must be seen to lie in a combination of the prefix “VAGI” and the suffix “SIL”. Therefore, the PAR was correct in holding that, when taken as a whole, the “VAGISIL” Marks possess a normal degree of distinctiveness. As observed in Kerly’s at para 8-133:

25 An important corollary of the finding that the “VAGISIL” Marks possess an overall normal degree of distinctiveness is that the threshold to surmount before the “VAGISAN” Mark will be regarded as being dissimilar is but a moderate one: see Hai Tong at [30]. With these matters in mind, I considered the similarity of the marks.

26 In concluding that the marks are visually similar to an above average degree, the PAR held that “the marks as wholes … coincide in the majority of the letters, as well as in their structure and number of letters”: at [38] of the GD. The plaintiff argued that the PAR erred in placing significant emphasis on the “VAGI” component when greater weight should have been given to the respective suffixes. Additionally, the plaintiff contended that the PAR’s finding disregarded the descriptiveness of the “VAGI” component and the differences in the distinctive suffixes in each mark. The plaintiff relied on Apple Inc. v Xiaomi Singapore Pte Ltd [2017] SGIPOS 10 (“Xiaomi”), where the Registrar found the “PAD” in the “IPAD” mark to be descriptive. In that case, the Registrar held that the “MI PAD” mark did not capture the “I” feature or the standalone “IPAD” word, both of which are distinctive (at [40]–[41] and [50(iv)]). The plaintiff also cited a series of decisions from other jurisdictions which dealt with marks containing identical prefixes.

27 On the other hand, the defendant contended that trade marks must be compared as wholes without giving special regard to particular elements: ACCUTRON Trade Mark [1966] RPC 152 at 155. In comparing whether “ACCUTRON” for clocks and watches could be opposed by the proprietors of the “ACCURIST” mark for wrist watches and the “ACCULARM” mark for alarm clocks, Buckley J held that, when comparing the visual similarities of the marks, the judge “must look at the marks as a whole and judge them as a whole as invented words”. In any event, the defendant highlighted the PAR’s observation that even “the suffixes of the marks in the instant case also share some similarity – they are three letters long and both begin with the letter ‘S’”: at [37] of the GD.

28 Specifically, the defendant relied on Sarika Connoisseur Cafe Pte Ltd v Ferrero SpA [2013] 1 SLR 531 (“Sarika CA”), where the Court of Appeal held that the “NUTELLA” and “NUTELLO” marks were visually similar because the marks had the same length (ie, seven letters), same structure (ie, a single word), and the same six out of seven letters: at [23]. A similar outcome was arrived at in Hyundai Mobis v Mobil Petroleum Company, Inc [2007] SGIPOS 12 (“Mobil”), where the Registrar found the “MOBIS” and “MOBIL” marks to be similar because when the marks were viewed as wholes, a large part of the marks were similar and only the last letters were different (at [41]). The defendant also referred to Johnson & Johnson v Uni-Charm Kabushiki Kaisha (Uni-Charm Corp) [2007] 1 SLR(R) 1082 (“Uni-Charm”) at [9] (where “CAREFREE” and “CAREREE” were found to be visually similar) and Future Enterprises Pte Ltd v McDonald’s Corp [2006] 4 SLR(R) 629 (“Future Enterprises”) at [9] (where “McCAFÉ” and “MacCoffee” were held to be visually similar).

29 Visual similarity is ascertained by “reference to the overall impressions created by the marks or signs, bearing in mind their distinctive and dominant components”: Hai Tong at [62(b)]. An overall impression may “in certain circumstances, be dominated by one or more of its components”: Bridgestone Corporation and Bridgestone Licensing Services, Inc. v Deestone Limited [2018] SGIPOS 5 at [60]; Hai Tong at [62(c)]. A descriptive element of a complex mark is not likely to be perceived “as the distinctive and dominant element of the overall impression conveyed by that mark”: Ceramiche Caesar SpA v Caesarstone Sdot-Yam Ltd [2017] 2 SLR 308 (“Caesarstone”) at [41]. Nevertheless, “[t]he finding of distinctiveness of the separate components of the mark must ultimately be related back to the impression given by the mark as a whole” because “it is the entire mark, and not only a component of it, that must function as the badge of origin”: Staywell at [29].

30 Here, consistent with my views expressed at [24] above, I agreed with the PAR that the prefix “VAGI” has weak distinctive character and, as such, is not likely to be perceived as the distinctive and dominant elements of the overall impressions conveyed by the marks. Rather, my view is that the respective suffixes of both marks constitute their distinctive components. In this regard, I noted that two out of three letters in “SIL” and “SAN” are different while the first out of the three letters are the same (ie, the letter “S”). Bearing all of the foregoing in mind, I was of the view that, based on a comparison of the overall impressions of the marks, the marks are visually similar. The fact that two out of three letters in the “SIL” and “SAN” components are different does not overshadow the coincidence of five letters (ie, “V”, “A”, “G”, “I” and “S”) in the two marks that comprise only seven letters each. Assessing the marks in terms of their overall visual impressions and as wholes, I found the marks to be visually similar. That being said, the visual similarity here is only to an average degree (as opposed to an above-average degree) owing to the differences in the distinctive components (ie, “SIL” and “SAN”).

31 This conclusion, in my view, is consistent with the cases cited by both parties. In Future Enterprises at [9], the High Court held that the “MacCoffee” mark is visually similar to the earlier “McCAFÉ” mark, notwithstanding the difference of four letters. The High Court held that the diacritical in “É” is likely to go unnoticed, the capital “C” divides both marks, and both suffixes share the common letters “c”, “f” and “e” in the same sequence. Moreover, the Court of Appeal in Sarika CA (“NUTELLA” and “NUTELLO”), the Registry in Mobil (“MOBIL” and “MOBIS”), and the High Court in Uni-Charm (“CAREFREE” and “CAREREE”) all held that the relevant marks are visually similar. However, the marks in those cases differ by only one letter while the marks here differ by two letters. For this reason, I found that the degree of visual similarity in the present case is average and not to an above average or high degree. The case of Xiaomi (“IPAD” and “MI PAD”), on the other hand, is quite different because the Registrar observed at [50(iv)] that the coincidence of the letters “I”, “P”, “A”, and “D” is diminished by the fact that the “I” in “MI PAD” is separated from “PAD”. The compound word marks here do not feature such a physical separation.

32 I turn now to the decisions regarding the same pair of marks, which must be treated with some caution given that our “step-by-step” approach diverges from the test in, for example, the EU (ie, the “global appreciation test”) and Australia (ie, the “deceptive similarity test”). In August Wolff (NZ), the New Zealand High Court held at [38] that, “focusing on the distinct suffixes, in an overall assessment and allowing for imperfect recollection, I consider some visual and aural similarity remains”. Similarly, in August Wolff (EU Appeal), the EU Board of Appeal, in dismissing the appeal against the Cancellation Board’s decision to cancel the “VAGISAN” Mark, held at [54] that: “[t]he differing last letters ‘IL’ and ‘AN’ are not sufficient to counteract the similarities resulting from the majority of the common letters, also taking into account that consumers only rarely have the chance to make a direct comparison”. On the other hand, the Federal Court of Australia held in August Wolff (AU) at [55] that the “VAGISAN” and “VAGISIL” marks are not deceptively similar because the common element of the two marks is descriptive and “should be paid less attention”. On the whole, I agreed with the reasoning of the New Zealand High Court and the EU Board of Appeal given their emphasis on the overall visual impression given by the marks. I should add that the marks-similarity test in New Zealand broadly mirrors our “step-by-step” approach: see August Wolff (NZ) at [63].

33 The plaintiff also cited the following cases: Apogepha Arzneimittel GmbH v Peptonic Medical Abi Opposition No B 2930645 (concerning “VAGIMID” and “VAGIVITAL”); Apogepha Arzneimittel GmbH v Peptonic Medical AB Opposition No B 3028522 (also concerning “VAGIMID” and “VAGIVITAL”); Apogepha Arzneimittel GmbH v Continental Brands Plus LLC Opposition No B 3011189 (concerning “VAGIMID” and “VAGISTAT”); and Apogepha Arzneimittel GmbH v Avia Pharma AB Opposition No B 3036640 (concerning “VAGIMID” and “VAGITIN”) (collectively, the “Vagimid Decisions”). In the Vagimid Decisions, the EU Intellectual Property Office found that the “VAGIMID” mark was dissimilar to “VAGIVITAL”, “VAGISTAT” and “VAGITIN”. However, the cases are distinguishable given that the suffixes in “VAGIVITAL” and “VAGISTAT” have more letters than “VAGIMID”. Furthermore, none of the suffixes in the latter three marks begin with the letter “M”. In contrast, the three-letter suffixes of “SIL” and “SAN” begin with the letter “S”.

34 Finally, I considered the decision of In the Matter of Application No. 2273267 by Pilkington Plc (O-116-04) (“Pilkington”). In that case, the applicant sought to register “OPTITEC” in Classes 9, 19 and 21, and the proprietor of the “OPTIROC” mark subsequently opposed the applicant’s registration. After conducting a visual comparison of the marks, the Registrar at [21] observed that both marks have the same “OPTI” prefix and that “the terminations ie, remainder of the marks, are conspicuously different and I do not believe that these differences would be readily ignored or marginalised in use”. I observed that Pilkington provides some useful guidance given that both Pilkington and the present case involve marks that have the same lengths of prefixes (ie, four letters), lengths of suffixes (ie, three letters), proportions of identical letters in the suffixes (ie, one out of three letters), and lengths of the word marks (ie, seven letters). Nevertheless, I was mindful that it is the last letter in the suffixes of “OPTITEC” and “OPTIROC” which are the same (ie, “C” in “TEC” and “ROC”). This seemed to me to be less visually impactful than the similarity in first letter of the suffixes of the competing marks in question (ie, the “S” in “SIL” and “SAN”).

35 For the reasons set out above, I found that the plaintiff’s “VAGISAN” Mark is visually similar to the “VAGISIL” Marks. However, I departed from the PAR’s view that the similarity is of “an above average degree”. Instead, it seemed to me that such similarity is to an average degree.

36 The PAR held that the compared marks are aurally more similar than dissimilar and that the degree of similarity is above average. She considered “the overall phonetic impression produced by the marks, which is influenced by the number and sequence of its syllables, and the rhythm and intonation of the marks”. In arriving at her conclusion, the PAR also considered that in the aural context, “VAGI” may not be immediately associated with the word “vagina” due to differences in pronunciation: at [46] of the GD.

37 Disagreeing, the plaintiff contended that while the root word “vagina” is pronounced with an “uh” vowel sound, the “VAGI” prefix is pronounced with an “eh” vowel sound when read in the context of most derivative words such as “vaginismus”, “vaginitis”, “vaginoplasty”, “vaginosis”. Thus, the plaintiff argued that the average consumer would still associate “VAGI” (albeit pronounced with an “eh” vowel sound) with the word “vagina”. Consequently, the plaintiff submitted that the average consumer would pay less attention to the “VAGI” component of the mark. Furthermore, the plaintiff contended that where a mark is constituted of multiple components of varying distinctiveness, such as in the present case, the “Dominant Approach” rather than the “Quantitative Approach” would be more appropriate. Applying the “Dominant Approach”, the plaintiff argued that the suffixes “SAN” and “SIL” would be pronounced in full and with equal emphasis on the third and first syllable of each mark. In particular, the “AN” and the “IL” sounds would not be confused with each other. Even if the “Quantitative Approach” were to be applied, the plaintiff stressed that the final syllables “SAN” and “SIL” are dissimilar. As such, the plaintiff argued that the compared marks are dissimilar or, at the most, similar to a low degree.

38 On the other hand, the defendant argued that under either approach, the marks would be aurally similar. Under the “Dominant Approach”, the defendant asserted that the first syllable of the “VAGISAN” and “VAGISIL” marks would be the most dominant and distinctive because consumers would emphasise the first two syllables of both marks when pronouncing them. Additionally, the defendant contended that the first syllable of a word is the most important for the purpose of distinction because persons using English tend to slur the termination of words: Ferrero SPA v Sarika Connoisseur Cafe Pte Ltd [2011] SGHC 176 at [63] (“Sarika HC”). The Court of Appeal in Sarika CA at [29] agreed that the first syllable of the marks would likely be emphasised and that “[i]t would do violence to the natural pronunciation of both words to say … that the ending vowels would be accented”. Under the alternative “Quantitative Approach”, the defendant noted that “VAGISAN” and “VAGISIL” each comprises three syllables, and it argued that weight should be accorded to the fact that the first two syllables of the words are identical. Since first impressions are of importance in determining aural similarity, a consumer with imperfect recollection would find the marks to be aurally similar: see Sarika CA at [31]. Even if the prefix “VAGI” is associated with “vagina” regardless of the difference in pronunciation, the defendant relied on the Court of Appeal’s observation in Staywell at [35] that the aural analysis involves the utterance of the syllables without exploring the composite meaning embodied by the words. As such, the defendant contended that the PAR’s reasoning was sound.

39 Having considered the parties’ arguments, I agreed with the PAR that the marks are aurally similar. The principles governing the aural similarity analysis are well settled. In Staywell, the Court of Appeal set out two approaches for assessing aural similarity (at [31]–[32]). The first is the “Dominant Approach” which requires identification of the dominant component of the marks, and the second is the “Quantitative Approach” which requires a quantitative assessment as to whether the competing marks have more syllables in common than not. When choosing the proper approach to apply, I agreed with the plaintiff that the Dominant Approach would be more suitable for marks containing multiple components, particularly where the components on their face differ significantly in their distinctive character. In Guccio Gucci S.p.A. v Guccitech Industries (Private Ltd) [2018] SGIPOS 1 (“Guccio”), the competing marks were “GUCCI” and . In assessing aural similarity, the Registrar opined at [27]–[28]:

40 Under the Dominant Approach, the court first ascertains whether a particular component has a high degree of distinctiveness: Staywell at [28]. In that case, the Court of Appeal held that “Regis” was the distinctive component of the marks in contention ie, “ST. REGIS” and “PARK REGIS”, because it stood out in the imperfect recollection of the consumer. Thus, the finding below of aural similarity on this basis was affirmed: at [31]. Similarly, the High Court in Formula One Licensing BV v Idea Marketing SA [2015] 5 SLR 1349 at [119], when comparing “F1” and “F1H20” aurally, held that where the first portion of a trade mark is used in a descriptive sense (in that case, “F1” was descriptive), its significance is lessened. On that basis, the court concluded that “F1” was aurally dissimilar to “F1H20”.

41 The parties did not dispute that the natural pronunciation of the “VAGISAN” Mark and the “VAGISIL” Marks are “VAGI-SAN” and “VAGI-SIL” respectively. As such, “VAGI” comprises the first component of both marks, while the suffixes “SAN” and “SIL” constitute the second component of the respective marks. Since both marks comprise multiple components, I found that the Dominant Approach was the more appropriate approach to apply here: see Guccio at [27]–[28]. Consistent with my findings at [24] and [30] above, I found that the “VAGI” prefix possesses weak distinctive character while the respective suffixes possess greater distinctive character. I further agreed with the plaintiff that the “SAN” and “SIL” sounds are striking and simple to pronounce for the relevant public in Singapore. Indeed, the “ih” vowel sound in “SIL” and the “ah” vowel sound in “SAN” are acoustically and phonetically distinct. Notwithstanding the articulation with respect to the “IL” and “AN” components of the respective marks, however, I was equally of the view that the foregoing analysis could not overcome the fact that: (a) two out of three syllables are aurally identical; and (b) even within the respective third syllables, the sibilant “s” sound is the same. Taking the effects of slurring, careless pronunciation and imperfect recollection in the round, I was of the view that the marks are aurally similar, but to an average degree.

42 In my view, this is consonant with the reasoning of the EU Board of Appeal in August Wolff (EU Appeal), which held at [54] that the marks “have the same structure, intonation and rhythm”. Separately, in August Wolff (NZ), the New Zealand High Court held at [38] that the “ear may be drawn to the beginning of the word (aurally New Zealanders have a tendency to clip and slur final syllables)”. On the other hand, the Federal Court of Australia held in August Wolff (AU) at [59] that “SIL” and “SAN” “are quite distinct” and that “the word marks sound quite different from each other”. On balance, I agreed with the reasoning in August Wolff (EU Appeal) and August Wolff (NZ). I note that in Pilkington, which concerned the “OPTITEC” and “OPTIROC” marks, the Registrar at [22] held that “the terminations of the respective marks, TEC and ROC, would sound noticeably different and serve to distinguish the marks in aural use”. However, consistent with my analysis at [34] above, unlike in the present case, it is the terminations (ie, the “C”) and not the beginnings of the third syllables in each mark (ie, the “T” and “R” in the respective marks) that are the same. This appeared to be less aurally impactful, particularly taking into account slurring, careless pronunciation and imperfect recollection.

43 For the reasons set out above, I agreed with the PAR to an extent and found that the plaintiff’s “VAGISAN” Mark is aurally similar to the defendant’s “VAGISIL” Marks, but to an average degree.

44 Overall, the PAR held that the marks are “similar to an above average degree” on the basis that they are visually and aurally similar to an above average degree and conceptually neutral: at [51] of the GD. Given my findings that the marks are visually similar to an average degree and are aurally similar to an average degree (and it was not disputed that the marks are conceptually neutral), I was of the view that, on balance, the marks, taken as wholes, are similar to an average degree.

45 Consistent with its position before the PAR, the plaintiff did not dispute that the goods claimed under the marks in Classes 3 and 5 are similar (or even identical in some instances). Therefore, I need not deal further with this step of the analysis.

46 Turning to the third step, unlike the inquiry into mark-similarity which is directed at the features of the marks in question, the question of confusion focuses on the effect that such similarity is likely to have on the relevant segment of the public: Hai Tong at [76].

47 There are at least two specific aspects to this element of confusion. The first is to mistake one mark for the other. In this aspect, the perception of the relevant segment of the public, therefore, is that the later mark is in fact the same as the registered mark. The relevant segment of the public does not perceive that there are two distinct marks. A second aspect of confusion arises where the relevant public perceives that the contesting marks are different, but nevertheless remains confused as to the origin which each mark signifies. Here, the relevant segment of the public perceives that goods bearing the earlier registered mark and those bearing the later mark emanate from the same source and is led to think that the trade source has chosen to diversify its range of marks for any of a multitude of marketing or promotional reasons or that the two sets of goods emanate from sources that are economically linked or associated: Hai Tong at [74]. However, “mere association” 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 concerned. Put another way, it is not sufficient that the relevant segment of the public would recognise or recollect similarities between the contesting marks if there is no likelihood of confusion as to origin ensuing: Hai Tong at [75].

48 The threshold of confusion required is that within a “substantial portion of the relevant segment of the public” in excess of a “de minimis level”: Hai Tong at [78(e)] and Sarika CA at [57]. Once again, the court considers the perspective of “the average consumer who would exercise some care and a measure of good sense in making his or her purchases, not that of an unthinking person in a hurry”. Also, the average consumer is treated to have “imperfect recollection”: Hai Tong at [40(c) and 40(d)]. Apart from the similarity of marks and goods, “extraneous factors” may also be taken into account in assessing the “likelihood of confusion”: Staywell at [95]–[96]. As summarised by the Court of Appeal in Caesarstone at [56], the factors relevant to the confusion inquiry are:

49 In the GD (see [58]–[70]), the PAR examined the various factors. She then held that the fact that the consumers would pay a fairly high degree of attention to the purchase of the goods concerned would render direct confusion “less likely”. However, she remarked that the possibility of “indirect confusion” remained. She reasoned that the relevant segment of the public might, on recognising that the marks are different, nevertheless assume that there is a connection between the two undertakings, especially given the similarity between the marks. Taking into account all the factors, and in the light of imperfect recollection, she concluded that there is on balance a likelihood of confusion: see [68] and [71] of the GD.

50 The plaintiff argued that there is no likelihood of confusion. In particular, the plaintiff cited: (a) the dissimilarity of the marks; (b) the fact that the “VAGISIL” Marks do not have a particularly famous reputation; (c) the distinctive suffixes that overcome imperfect recollection and slurring; (d) the fact that the parties’ marks have coexisted on trade mark registers in at least 44 jurisdictions; ²¹ and (e) the goods being highly personal self-care products requiring specialist assistance. Also, the plaintiff distinguished the present case from the three examples of indirect confusion set out in Virgin Enterprises Ltd v Virginic LLC [2019] EWHC 672 (Ch) (“Virgin”).

51 In response, the defendant contended that there is a likelihood of confusion. Relying on Hai Tong (at [103]–[107]), it agreed with the PAR’s conclusion on confusion. The reasons include: (a) the similarity and identity of goods claimed; (b) the high degree of mark similarity; (c) the reality of imperfect recollection when only the two final letters of the marks differ; (d) the identical segment of the relevant public that the products target; (e) the average consumer comprising both the public at large and those in the trade; (f) the availability of the products in non-specialist contexts; and (g) the fact that the products do not invite significant due diligence on the part of the relevant segment of the public.

52 I highlight that the PAR used the expression “direct confusion” to refer to the first aspect of the element of confusion, ie, mistaking one mark for the other, and “indirect confusion” to refer to the second aspect of the element ie, recognising the marks to be different but nevertheless assuming an economic link or connection between the goods bearing the competing marks.

53 Having heard the parties, I found that six factors supported a finding of a likelihood of confusion:

(a) First, as I have stated at [44] above, the marks in question are similar to an average degree. Even if the “VAGI” component is regarded to have a descriptive connotation, the “VAGISIL” Marks are distinctive as a whole: see [24] above.

(b) Second, I accepted that the “VAGISIL” Marks have a respectable degree of reputation in Singapore. Products bearing the “VAGISIL” Marks were first available in 1990 and continue to be sold at popular Singapore retailers. ²² They have been advertised and sold in Singapore at creditable volumes between 2011 and 2018. They have also been referred to by prominent third parties in television shows, movies, books and articles. While I agreed that the court ought not to engage in a “reputation-therefore-confusion” analysis (Rovio Entertainment Ltd v Kimanis Food Industries Sdn Bhd [2015] 5 SLR 618 (“Rovio”) at [125]), this was nevertheless a relevant consideration in this inquiry.

(c) Third, as noted at [45] above, the plaintiff conceded that the goods covered by the “VAGISIL” Marks in Classes 3 and 5 are similar. In fact, some of the goods claimed under the respective marks are identical. For example, the plaintiff’s “soaps” and “cosmetics” in Class 3 overlap with the defendant’s “cosmetics and toiletries for feminine use”, while the plaintiff’s “pharmaceutical products” in Class 5 overlap with the defendant’s “pharmaceutical preparations, medicated creams, and vaginal suppositories”.

(e) Fifth, the relevant segment of the public (ie, females in Singapore) would carry in their minds an imperfect recollection of the overall impression of the marks.

(a) First, while the relevant goods are relatively inexpensive, I was mindful that the nature of the goods would tend to command a greater degree of fastidiousness and attention on the part of prospective purchasers. The goods in question are highly personal self-care products for women, which require greater care in selection: Consolidated Artists B.V. v THEFACESHOP Co., Ltd. [2017] SGIPOS 7 (“THEFACESHOP”) at [123]. In concluding that there was no likelihood of confusion, the Registrar at [120] observed that, while the goods in Class 3 for body and face care were not very expensive and were purchased off the shelf, consumers “will be cautious about what they are purchasing”. Furthermore, such a purchasing decision would not be made “lightly” as it was a “choice with possible adverse consequences if the wrong products are used, e.g. allergies or acne breakouts”: see [120]. Similarly, the Federal Court of Australia in August Wolff (AU) opined at [68] that “[t]he proposition that a woman is likely to pay little attention to the particulars of any product that she might put in her vagina or on her vaginal area inherently lacks credibility”. Female consumers in Singapore would likely pay at least a medium level of attention before or during the purchase of the products in question. Accordingly, I agreed with the PAR (at [65] of the GD) that the relevant segment of the public would not simply purchase the goods in question in a hurry without putting in some care and attention into the purchase process.

(b) Second, while a portion of consumers would purchase the relevant products online, it seemed to me that, at present, consumers still normally purchase these goods in brick-and-mortar shops. In THEFACESHOP, the Registrar held at [123] that although the facial and body care products were self-service items, “specialists abound to assist in the purchase, having regard to the deeply personal nature of the same”. Furthermore, the Registrar held at [122] that “consumers will be more particular about the origin or marks (emphasis original) of such goods, as they are likely to trust certain brands more than others”. This applies with equal if not greater force to purchases of intimate care products. Notwithstanding that intimate care products may be self-service items, when purchased from shops, decisions to buy these items are likely to be made with the assistance of specialists.

55 On balance, I was persuaded that, having regard to the extraneous factors – even accounting for the two factors discussed at [54] above – as well as the similarity of the marks and goods in question, there exists a likelihood of confusion on the part of the public. In my judgment, even if some consumers would not regard the marks to be the same, a consideration of the factors at [53]–[54] above indicated that those consumers would nevertheless likely assume the existence of an economic link or connection between the goods bearing the two marks.

56 I make three additional points to address the plaintiff’s arguments. First, the fact that other traders have used the “VAGI” prefix in respect of other vagina-related products (see [22] above) does not necessarily mean that no likelihood of confusion arises. Granted, I accepted that the “VAGI” prefix has, in Singapore, been applied to intimate care products. The plaintiff indeed referred to: (a) printouts from e-commerce websites selling VAGI-prefixed goods; (b) articles referencing “VAGISTAT” and “VAGIFEM”; and (c) extracts of detailed particulars for trade mark registrations in respect of “VAGINNE”, “VAGILUR” and “VAGIFEM”. ²³ However, co-existence of marks with a similar prefix on a register does not necessarily negate the possibility of confusion, but may simply show that the common element is unlikely to be viewed as denoting a specific trade origin: see Digipos Store Solutions Group Ltd v Digi International Inc [2008] EWHC 3371 (Ch) at [61].

57 Second, the fact that both marks in question have coexisted on foreign trade mark registers or have been sold in the marketplace in various other countries is neither here nor there. The plaintiff provided evidence that “VAGISAN” and “VAGISIL” products had been sold in the UK since 2013, in Poland since 2015, and in the Netherlands since 2017. ²⁴ Moreover, the plaintiff adduced evidence that the “VAGISIL” Marks and the “VAGISAN” Mark co-exist on the trade marks registers of 44 other foreign jurisdictions. ²⁵ Nevertheless, the Court of Appeal in Caesarstone at [90] did not accord much weight to the coexistence of competing marks in foreign jurisdictions as, apart from the fact that the trade mark regimes in those jurisdictions might have differed from Singapore’s, the court needed more evidence to show how the coexistence stemmed from the absence of confusion.

58 Third, I turn to the decision of Virgin, which was an appeal against a Hearing Officer’s decision dismissing Virgin Enterprises Ltd’s (the proprietor of the earlier mark “VIRGIN”) opposition to Virginic LLC’s registration of “VIRGINIC” in respect of goods in Class 3. It was observed that instances where the court would find a likelihood of indirect confusion tended to fall into one or more of three categories as follows at [17]:

59 The Hearing Officer concluded that there was no likelihood of either direct or indirect confusion. Arnold J departed from the Hearing Officer’s decision and held at [21] that there was a likelihood of indirect confusion arising from the second category cited in the quoted section above. The learned judge noted that the average consumer would perceive “VIRGINIC” as being the result of adding the “IC” suffix (which in English typically means “of” or “pertaining to”) to “VIRGIN”. Since this would make “VIRGINIC” a “newly-minted adjective ‘of or pertaining to VIRGIN’”, Arnold J opined that the average consumer would likely think that “VIRGINIC was a brand extension of VIRGIN”: at [21].

60 I noted that the three categories of indirect confusion were stated as being mere “[i]nstances where one may expect the average consumer to reach … a conclusion” of indirect confusion: at [17]. The three categories are therefore not exhaustive. As such, I did not find this case particularly useful, and rest my decision on the consideration of the factors discussed at [53]–[54] above. In any event, it seemed to be arguable that the present case fell within the third scenario set out in Virgin. Here, the earlier marks, “VAGISIL”, comprise two main components, “VAGI” and “SIL”. The relevant public would likely regard the suffix of the later mark, “SAN”, to be a “change” from the “SIL” component. This “change” may be perceived to be consistent and logical with “VAGISAN” being a brand extension of “VAGISIL” because, in addition to the identical prefixes, the suffixes (ie, “SIL” and “SAN”) of each: (a) have the same length; (b) begin with the same letter; and (c) comprise of a vowel sandwiched between two consonants.

61 Based on the foregoing, I found that the marks are similar to an average degree, the goods are similar, and there is a likelihood of confusion. Accordingly, I agreed with the PAR that the ground pursuant to s 23(3)(a)(i) read with s 8(2)(b) of the Act was made out.

62 Unlike s 8(2)(b) of the Act, the analysis under s 8(7)(a) requires the court to examine the “VAGISIL” Marks as they have been actually used and sold in Singapore. Sections 8(7)(a) and 23(3)(b) of the Act state:

63 The three elements of passing off are (a) goodwill, (b) misrepresentation, and (c) damage to goodwill: Novelty Pte Ltd v Amanresorts Ltd and another [2009] 3 SLR(R) 216 at [37] and affirmed in Singsung Pte Ltd v LG 26 Electronics Pte Ltd (trading as L S Electrical Trading) [2016] 4 SLR 86 (“Singsung”) at [28]. Goodwill refers to the “trading relationship with [the proprietor’s] customers”: Allergan, Inc and another v Ferlandz Nutra Pte Ltd [2016] 4 SLR 919 at [170]. Misrepresentation is a consumer-centric inquiry as it focuses on the deception occasioned to the relevant public. In the present appeal, the parties were agreed that the defendant’s “VAGISIL” Marks possess goodwill but they disagreed on the second and third elements of the tort.

64 In relation to this element, there is a two-stage inquiry. First, “the issue of distinctiveness is best understood as a threshold inquiry in the context of determining whether the defendant has committed an actionable misrepresentation”. If a mark is not distinctive, the mere fact that the defendant has used something similar or identical in marketing and selling its products would not amount to a misrepresentation that the defendant’s products are the plaintiff’s or are economically linked to the plaintiff: Singsung at [38].

65 Second, the test for a misrepresentation which creates a likelihood of confusion is substantially the same as that for “likelihood of confusion” under s 8(2)(b) of the Act: Sarika HC at [212]. In Hai Tong, the Court of Appeal stated at [115] that, “although in an action for passing off, the court is not constrained in the same way that it would be in a trade mark infringement action in identifying the factors it may take into account”, the court was amply satisfied that the element of misrepresentation had been made out for the same reasons as those relied on to hold that there was a “likelihood of confusion” in an infringement action under s 27(2)(b) of the Act (which mirrors the language of s 8(2)(b) of the Act).

66 For the same reasons relied upon in concluding that there is a likelihood of confusion, the PAR held that the plaintiff’s use of the “VAGISAN” Mark amounted to a misrepresentation that creates a likelihood of confusion: at [95] of the GD. She also held that the “V” device, which appears on the defendant’s products, does not detract from the distinctiveness of the defendant’s “VAGISIL” Marks and that consumers would still distinguish the defendant’s goods by the word mark “VAGISIL” with or without the “V” device: at [94] of the GD.

67 Before me, the plaintiff argued that the “V” device is always used in conjunction with (either to the left or atop) the “VAGISIL” word mark on the defendant’s products sold in Singapore. Repeating its analysis that the marks are dissimilar and that there is no likelihood of confusion, the plaintiff also submitted that there is no misrepresentation that creates a likelihood of confusion. Needless to say, the defendant supported the PAR’s reasoning both with regard to the “V” device as well as her conclusion regarding the element of misrepresentation.

68 For reference, the packaging of the defendant’s products as they are sold and advertised in Singapore are set out below:

With regard to the “V” device, I agreed with the PAR and the defendant that it does not detract from the distinctiveness of the defendant’s “VAGISIL” word marks. In Caesarstone at [38], the Court of Appeal stated that where a device is not of a complicated nature and is simple and does not evoke any particular concept in the eyes of the average consumer, the device is “more likely to be perceived as a decorative element rather than as an element indicating commercial origin”. In my view, the “V” device clearly fell within this description. It is simple. It will more likely be perceived as a decorative element, and not as an indication of the source of the goods. Consumers are likely to pay greater attention to the word “VAGISIL” and, in particular, the “SIL” component of the marks.

69 To reiterate, I agreed with the PAR that the “VAGISIL” Marks possess normal distinctive character for the reasons set out at [24]–[25] above. Therefore, the requisite threshold was crossed. For the reasons I have stated above at [53]–[55], I found that there is a likelihood of confusion under s 8(2)(b) of the Act. For the very same reasons, I found that the use of the plaintiff’s “VAGISAN” Mark would amount to a misrepresentation that creates a likelihood of confusion.

70 The PAR found that there is a real likelihood of damage arising from the diversion of custom, having regard to the similarity of goods sold and the likelihood of confusion: at [97] of the GD. The plaintiff relied on its arguments against a finding of misrepresentation to assert that there is no likelihood of damage. On the other hand, the defendant relied on the similar field of business activity giving rise to direct competition and the similar marks to argue that there is a real likelihood of damage.

71 In Hai Tong at [118], the Court of Appeal held that there would likely be diversion of sales of “Lady Rose” products if consumers were to believe that “Rose Lady” cosmetic products were the goods of Hai Tong or were otherwise associated with or connected with Hai Tong. Similarly, in Uni-Charm at [26]), the High Court, having concluded that there was significant similarity between the “CAREFREE” and “CAREREE” marks and the relevant goods (ie, hygiene products), held that there was a likelihood of damage.

72 Given the parties’ direct competition in the same lines of products, the similarity of the marks, and the existence of a likelihood of confusion, I found that there is a real likelihood of damage to the defendant’s goodwill arising from the diversion of custom if consumers from the relevant segment of the public are led to believe that goods bearing the “VAGISIL” Marks are the goods of or are connected or associated with the plaintiff. I agreed with the PAR that the element of likelihood of damage to the defendant’s goodwill is made out.

73 For these reasons, the ground for declaring the plaintiff’s “VAGISAN” Mark invalid under s 23(3)(b) read with s 8(7)(a) of the Act was established.

74 As stated at [14] above, this appeal did not strictly concern the outcome of the third ground. The plaintiff, however, challenged a finding made by the PAR within this ground. Therefore, I deal with this briefly. Sections 8(4)(b)(i) and 23(3)(a)(iii) of the Act respectively state:

75 Under s 8(4)(b)(i) of the Act, the defendant must establish that the defendant’s “VAGISIL” Marks are well known in Singapore, that the whole or an essential part of the defendant’s “VAGISIL” Marks are either identical or similar to the plaintiff’s “VAGISAN” Mark, that the use of the plaintiff’s “VAGISAN” Mark in relation to the goods claimed would indicate a connection between its goods and the defendant, and that the interests of the defendant are likely to be damaged as a result.

76 While the plaintiff supported the PAR’s finding that the defendant’s “VAGISIL” Marks are not well-known, the plaintiff disagreed with the PAR’s finding at [74] of the GD that the marks are similar. Both parties repeated their submissions in respect of mark similarity here. In Staywell at [120], the Court of Appeal agreed that a finding on mark similarity under s 8(2)(b) of the Act would naturally lead to a finding of mark similarity under s 8(4)(b)(i) of the Act. In a similar vein, the High Court in Rovio at [146] held that there was no difference between the two marks-similarity analyses. Consistent with my finding that the marks in question are similar under s 8(2)(b), I found that the marks are likewise similar under s 8(4)(b)(i) of the Act.

77 By the above, I agreed with the PAR that the registration of the “VAGISAN” Mark should be declared invalid on the first two grounds relied on by the defendant.

78 Accordingly, I dismissed the appeal. I ordered the plaintiff to pay costs of the appeal fixed at $7,500 (inclusive of disbursements) to the defendant.