1 Much of the food that is consumed in Singapore is produced abroad. Generally, the consumer has little interest in the provenance of the produce that ends up on the dinner table. This, however, is not always the case. Many would say that the limestone-permeated water of Ipoh lends a special quality to the kway teow produced there. Also famous is the pork that comes from the Black Iberian pig which has been raised on the Iberian Peninsula and fed a diet of acorns grown there. In both these examples the special qualities of the landscape (or terroir to use the French phrase) have given the food product a unique flavour or quality which consumers actively seek out.

2 How then do consumers identify those food products which bear a special connection with the terroir? This is where the notion of Geographical Indications (“GIs”) which identify and connect foods with a particular region comes into play. It should be noted that there are two main approaches to this notion. The first is what is known as the terroir approach. This refers to the objectively verifiable causal connection between a product’s qualities and its region of origin: Irene Calboli, “In Territorio Veritas: Bringing Geographical Coherence in the Definition of Geographical Indications of Origin under TRIPs” (2014) 6 WIPO J 57 at 60–61.

3 The origins of the terroir approach can be traced back to the attempts to prevent fraud in the marketplace for wines in France. Given the prevalence of counterfeits and adulterated products in the late 1800s, France enacted a wine labelling law in 1905 which prohibited the misuse of wine names. This law was revised in 1919 and again in 1935. To justify protecting geographical names, French laws wove, into the normative framework of protection, the “notion that location, and more precisely the terroir – a deep connection between the products and the land … where the vine was grown and the wines were made, was a ‘key ingredient in differentiating between wines by indicating a distinct origin.’”: Irene Calboli, “Geographical Indications between Trade, Development, Culture, and Marketing: Framing a Fair(er) System of Protection in the Global Economy?” in Geographical Indications at the Crossroads of Trade, Development, and Culture (Irene Calboli and Ng-Loy Wee Loon (eds)) (Cambridge University Press, 2017) (“GIs at the Crossroads of Trade, Development, and Culture”) at p 11.

4 This marked a move towards “explicitly protecting ‘geographical origin’” which was a change from the previously existing laws on unfair competition and the “protection granted to ‘indications of origin’ in the 1883 Paris Convention for the Protection of Industrial Property”: GIs at the Crossroads of Trade, Development, and Culture at pp 10–11. GIs can therefore be said to denote the unique relationship between the products and the terroir: GIs at the Crossroads of Trade, Development, and Culture at p 23. This connection between the terroir and the GI-denominated products means that GIs offer accurate information to consumers about the geographical origins of the product in question.

5 The second approach to GI protection is connected to unfair competition. Here, the paramount consideration is whether a valuable reputation exists in the marketplace for a regional product. The existence of such a reputation can be established through evidence such as sales figures, advertisements, or consumer surveys. Crucially, this reputation does not have to be essentially or causally attributable to geographical origin in the manner that the terroir approach requires. This approach traces its roots to other European jurisdictions where GIs were protected under unfair competition law (a term used to refer to a cluster of tortious actions (or their equivalents in delict), an individual tort by itself or the basis for a specific statutory regime). The central theme, however, is the emphasis on misconduct at the marketplace. For example, a trader can claim that her goods are better than those of a rival, or claim that her rival’s goods are inferior, but she cannot lie to customers about the source or quality of her goods: Dev S. Gangjee, “From Geography to History: Geographical Indications and the Reputational Link” in Geographical Indications at the Crossroads of Trade, Development, and Culture (Irene Calboli and Ng-Loy Wee Loon (eds)) (Cambridge University Press, 2017) at pp 45–46.

6 Many countries around the world have, to meet their obligations under various international treaties, implemented legislative frameworks for the protection of GIs. Singapore is one such country.

7 The present appeal is the first time this court has had to consider the operation and interpretation of various provisions under the Geographical Indications Act 2014 (No 19 of 2014) (the “GIA”) which came into force on 1 April 2019. The issue specifically raised in this appeal concerns the operation of s 41(1)(f) of the GIA which provides that a GI, which contains the name of a plant variety or an animal breed and is likely to mislead the consumer as to the true origin of the product, shall not be registered as a GI.

8 The Appellant in this case is the Consorzio di Tutela della Denominazione di Origine Controllata Prosecco (the “Consorzio”). The Consorzio is a trade body established and organised under the laws of Italy. It is responsible in Italy for protecting, promoting, marketing and generally overseeing the use of the term “Prosecco”.

9 The Respondent is Australian Grape and Wine Incorporated (“AGWI”). It is the representative body for grape growers and winemakers in Australia.

10 On 3 May 2019, the Consorzio applied to register “Prosecco” as a GI in respect of wines in Singapore (the “Application GI”). The claimed geographical area for the production of “Prosecco” wines was the “North East region of Italy”, and included the entire territory of Belluno, Gorizia, Padova, Pordenone, Treviso, Trieste, Udine, Venice and Vicenza (the “Specified Region”). The Application GI was accepted and published in the Geographical Indications Journal under Geographical Indication No 50201900088S on 21 June 2019.

11 On 9 September 2019, AGWI filed a notice of opposition against the registration of the Application GI. It relied on two grounds of opposition. First, under s 41(1)(f) of the GIA, that the Application GI contained the name of a plant variety and was likely to mislead the consumer as to the true origin of the product. Secondly, under s 41(1)(a) of the GIA, that the Application GI did not fall within the meaning of a “geographical indication” as defined in s 2(1) of the GIA.

12 AGWI’s opposition was dismissed by the Principal Assistant Registrar of Geographical Indications (“PAR”) who ruled that neither ground of opposition had been made out: Consorzio di Tutela della Denominazione di Origine Controllata Prosecco v Australian Grape and Wine Incorporated [2021] SGIPOS 9. In her decision, the PAR noted (at [49]) the length of time that Italian “Prosecco” wine had been sold in Singapore (since 2011) and the amount sold over eight years (387,100 litres in 2018 alone) and considered that the likelihood of consumers being misled was small in view of the popularity, reputation and renown of these wines. Australian “Prosecco” wine, on the other hand, had only been introduced into the Singapore market in 2015 and the quantity of such wine sold here was significantly lower (9,657 litres in 2018).

13  Dissatisfied with the result, AGWI appealed against the PAR’s decision. The appeal was heard by a judge in the General Division of the High Court (the “Judge”). The Judge ruled that AGWI’s opposition under s 41(1)(f) of the GIA succeeded but dismissed AGWI’s opposition under s 41(1)(a) of the GIA: Australian Grape and Wine Inc v Consorzio di Tutela della Denominazione di Origine Controllata Prosecco [2022] SGHC 33 (the “GD”).

14 The Judge ruled, in relation to AGWI’s ground of opposition under s 41(1)(f) of the GIA, that the Application GI did, objectively, contain the name of a plant variety, to wit the Prosecco grape. It followed, the Judge reasoned, that the Application GI would be likely to mislead the consumer if “Prosecco” grapes and “Prosecco” wines had been cultivated or produced in significant quantities outside the Specified Region. Given the evidence that “Prosecco” wines were being produced in commercial quantities in countries such as Australia, the Judge concluded that the Application GI was likely to mislead the consumer in Singapore and thus allowed AGWI’s opposition under s 41(1)(f) of the GIA.

15 As to AGWI’s second ground of opposition under s 41(1)(a), the Judge ruled that s 2(1) of the GIA simply required that a GI be any indication used in trade to identify goods as originating from a place. The Judge concluded from the evidence that “Prosecco” was an indication used in trade to identify wine originating from the Specified Region. The crucial question was whether there was a given quality, reputation or other characteristic of the goods (ie, Prosecco wine) which was essentially attributable to the Specified Region. In making the Application GI, the Consorzio had adduced evidence to establish that the relevant characteristics of “Prosecco” wine were essentially attributable to the Specified Region. AGWI disputed the adequacy of the Consorzio’s evidence. But AWGI was unable to adduce expert factual evidence to prove its assertion that the specified characteristics of the wine, Prosecco, did not attach to the entirety of the Specified Region or its assertion that the “Prosecco” wine could be made outside the Specified Region. Therefore, it failed in its opposition on this ground.

16 The Consorzio now appeals against the Judge’s decision in relation to the ground of opposition under s 41(1)(f) of the GIA. There is no cross-appeal by AGWI against the Judge’s decision to dismiss its ground of opposition under s 41(1)(a) of the GIA.

17 In the hearing of this appeal, we were greatly assisted by the views of Professor Ng-Loy Wee Loon (“Prof Ng-Loy”), who was appointed as Independent Counsel.

18 There are two main issues raised by this appeal:

(a) What is the proper approach to be taken under s 41(1)(f) of the GIA (“Issue 1”)?

(b) Applying this approach, does AGWI’s ground of opposition under s 41(1)(f) of the GIA succeed (“Issue 2”)?

19 We begin by tracing the history of the GIA.

20 The first piece of legislation implementing protection for GIs was the Geographical Indications Act 1998 (“GIA 1998”). The GIA 1998 was enacted for two reasons. The first reason was to enhance the level of protection for GIs and the second reason was to enable Singapore to comply with its obligations under Arts 22 to 24 of the World Trade Organisation Agreement on Trade Related Aspects of Intellectual Property Rights (the “TRIPs Agreement”): Singapore Parliamentary Debates, Official Report (26 November 1998) vol 69 at col 1704 (Assoc Prof Ho Peng Kee, Minister of State for Law).

21 The GIA 1998 was subsequently repealed and replaced by the GIA. This legislation was enacted in 2014 to enable Singapore to comply with its obligations under the Free Trade Agreement between the European Union and the Republic of Singapore (the “EUSFTA”): Singapore Parliamentary Debates, Official Report (14 April 2014) vol 91 (Ms Indranee Rajah, Senior Minister of State for Education and Law (“Ms Indranee Rajah”)) (“Second Reading of the Geographical Indications Bill 2014”).

22 While, like the GIA 1998, the GIA conferred the level of protection mandated by the TRIPs Agreement on a GI, without the need to register the GI, the GIA provided for enhanced protection by establishing a system for the registration of GIs. A Registry of Geographical Indications was set up by the GIA. Its role is to examine applications for GI registration in respect of: (a) wines and spirits; and (b) selected categories of agricultural products and foodstuffs such as cheese, meat and seafood. There are three main parts to the GI registration process:

(a) First, an application to register the GI has to be made. Applicants must specify the quality, reputation or characteristics of the product in question and how that is attributable to the geographical origin.

(b) Second, the application would be examined in detail to ensure that certain requirements have been fulfilled.

(c) Third, the application would be made public for a period to allow third parties to object to the registration of the proposed GI.

23 Registration of a GI is valid for ten years, after which it can be renewed for further periods of ten years. GI registration is also based on a “first in time, first in right” principle – a new application for GI registration may not invalidate a prior conflicting GI or trade mark which already exists. Registering a GI also confers the following benefits: increased certainty of recognition, an enhanced level of protection above that afforded to unregistered GIs and improved border enforcement measures: Second Reading of the Geographical Indications Bill 2014. For instance, pursuant to s 56 of the GIA, a party interested in goods identified by a registered GI can file a request with the Director-General of Customs to seize goods which bear a term identical with that of the registered GI, but which do not originate from the place indicated by that registered GI.

24 While registered GIs are afforded protection under the GIA, it is important to reiterate that unregistered GIs are also protected. This protection finds its basis in s 4 of the GIA which states:

25 In particular, ss 4(2)(c) and 4(2)(d) of the GIA set out the “special non-confusion based protection reserved for GIs identifying wines or spirits”: Ng-Loy Wee Loon, Law of Intellectual Property of Singapore (Sweet & Maxwell, 3rd Ed, 2021) (“Law of Intellectual Property of Singapore”) at [28.3.5]. Unregistered GIs which identify wines or spirits enjoy this enhanced level of protection because of Singapore’s obligations under the TRIPs Agreement (see Art 23 of the TRIPs Agreement) as well as the EUSFTA (see Art 10.19 of the EUSFTA). This enhanced protection accorded to GIs identifying wines or spirits is extended to registered GIs listed in a schedule: s 4(6) of the GIA. As explained by Ms Indranee Rajah at the Second Reading of the Geographical Indications Bill 2014:

26 The GIA was recently amended with the introduction of the Geographical Indications (Amendment) Act 2020 (No 5 of 2020), which entered into force on 15 August 2020. The legislative changes effected by the amendments have no bearing on the present appeal.

27 The PAR, as well as the Judge, ruled that the inquiry under s 41(1)(f) of the GIA had two conjunctive requirements. First, that the Application GI contained the name of a plant variety, and secondly, that the Application GI was likely to mislead the consumer as to the true origin of the product.

28 Neither the Consorzio, nor AGWI, disputes the ruling that s 41(1)(f) of the GIA contains these two elements that need to be satisfied for the section to apply. Rather, the crux of their dispute is how the inquiry as to the first element or requirement is to be carried out. The Consorzio argues that the Judge erred in ruling that this inquiry as to whether a GI contains the name of a plant variety or an animal breed is a question of objective fact. According to the Consorzio, this assessment should also take into account the perspective of the Singapore consumer. In other words, if the Singapore consumer does not know that a word in a GI is the name of a plant or an animal breed, that ignorance would defeat any opposition brought under the sub-section. This interpretation would, the Consorzio argues, be consistent with both the express wording and a purposive interpretation of s 41(1)(f).

29 In response, AGWI argues that the Judge was correct in finding that the first element only entailed an objective factual assessment of whether the Application GI contained the name of a plant variety or an animal breed. The perspective of the consumer should not be taken into account in this assessment as doing so would change the plain and ordinary meaning of the provision, and there is no evidence of any legislative intention that the perspective of the consumer should have a bearing on whether a name is the name of a plant variety or an animal breed.

30 We agree that when an objection is raised to the registration of a GI under s 41(1)(f) of the GIA, the first thing that needs to be established is that the proposed GI contains the name of a plant variety or an animal breed. This is, in a sense, a threshold requirement because if the GI does not contain such a name there will be no need at all to consider the question of confusion. In most cases, whether the name requirement is fulfilled will not be a disputed point requiring proof because many plant varieties and animal breeds are well known either generally or within the specific industry to which the proposed GI relates. In the unusual case where the name is exotic or has been changed or for some other reason is disputed, we agree with the Judge that the fulfilment of the requirement is to be assessed objectively. There must be evidence demonstrating that the name in question is indeed the name of a plant variety or an animal breed. Assuming this determination is made, the enquiry then turns to consider whether the Application GI, which contains the name of a plant variety or an animal breed, is likely to mislead the consumer. In the paragraphs that follow, we give our reasons for agreeing with the Judge’s interpretation of the section and rejecting the Consorzio’s argument.

31 It is established law that when it comes to statutory interpretation, the purposive approach is to be applied and the text of the statute must necessarily be the court’s first port of call: Tan Cheng Bock v Attorney-General [2017] 2 SLR 850 (“Tan Cheng Bock”) at [37]. We thus begin our analysis with the text of s 41(1)(f) of the GIA:

32 We note that AGWI argues that Art 6(2) of the Regulation on Quality Schemes for Agricultural Products and Foodstuffs, EU Council Regulation 1151/2012, [2012] OJ L 343/1 (the “EU Regulation 1151”) could inform the interpretation of s 41(1)(f) of the GIA. This is because s 41(1)(f) of the GIA was adapted from Art 6(2) of the EU Regulation 1151, as indicated in the draft of the proposed legislative amendments to the GIA annexed to the Intellectual Property Office of Singapore, Public Consultation on Changes to be made to the Geographical Indications Act and Trade Marks Act to enhance Singapore’s Regime for the Protection of Geographical Indications (1 November 2013). While the Consorzio accepts that s 41(1)(f) of the GIA was adapted from Art 6(2) of the EU Regulation 1151, it makes the point that the legislative frameworks in European Union (“EU”) and Singapore are completely different such that there is limited scope to apply EU principles to the interpretation of the GIA. For clarity, the text of Art 6(2) is as follows:

33 We do not consider that the principles relating to the interpretation of Art 6(2) of the EU Regulation 1151 are useful in arriving at the proper interpretation of s 41(1)(f) of the GIA. It is evident to us that there are marked differences between the Singapore model of GI protection and the EU model. As Prof Ng-Loy pointed out, EU GI law is found in more than one Regulation, including:

(a) the EU Regulation 1151 which governs GIs for agricultural products other than wines and spirits;

(b) Regulation 1308/2013 which governs GIs for wines (“Wine Regulation”); and

(c) Regulation 2019/787 which governs GIs for spirits.

The Wine Regulation contains no equivalent of Art 6(2) of the EU Regulation 1151. Instead, a further regulation was passed specifying which grape varieties could be used as names of wines by particular European Union countries.

34 It is thus clear that the EU regime is different from the Singapore one. The interpretation of Art 6(2) of the EU Regulation 1151 will necessarily have to take into consideration the GI regime in the EU. Similarly, in the Singapore context, how s 41(1)(f) of the GIA is to be interpreted must necessarily be considered in the context of the GIA. What is more relevant and pertinent to this exercise of statutory interpretation is the legislative intention behind the GI protection framework. We therefore turn to consider the parliamentary intention behind s 41(1)(f). In doing so, we consider all the relevant materials surrounding the promulgation of s 41(1)(f): Attorney-General v Ting Choon Meng and another appeal [2017] 1 SLR 373 at [18].

35 As we have noted above (at [21]), Singapore implemented the GIA to comply with its obligations under the EUSFTA. The provisions of the EUSFTA are therefore useful in our exercise of interpreting s 41(1)(f) of the GIA. There is, as Prof Ng-Loy pointed out, a provision in the EUSFTA which explicitly preserves the right to protect the name of a plant variety or an animal breed as a GI. This is Art 10.22(8) which states:

36 In this connection, Prof Ng-Loy also drew our attention to the side letter of the EUSFTA which recorded the following agreement reached between Singapore and the EU on the registration of plant varieties as a GI:

37 We agree with Prof Ng-Loy’s view that this agreement in the side letter is reflected in the legislation which allows the registration of GIs which contain the name of a plant variety, provided that such registration shall not prevent the use of the name as a plant variety. This is expressed in s 15(b) of the GIA which states:

38 What s 15(b) of the GIA does is to carve out an exception. It provides that the use of the name of a plant variety or an animal breed, that also happens to be a registered GI, in the course of trade shall not constitute infringement, as provided for by s 4 of the GIA. As a side point, the recognition of the need to protect the use of such names in the course of trade indicates that the query as to whether the name is one of a plant variety or an animal breed must be an objective one, not connected with the subjective beliefs of consumers.

39 It is also clear to us, and we agree with Prof Ng-Loy, that when one reads s 15(b) and s 41(1)(f) of the GIA, in the context of what was agreed in the side letter to the EUSFTA, it is the confusion requirement that serves as gatekeeper when it comes to the registration of GIs that contain the name of plant varieties or animal breeds. The GIA essentially recognises that what is objectively the name of a plant variety or an animal breed can, apart from being used in the course of trade, also be used in another capacity: that of a GI. There is, however, one potential problem with allowing the names of plant varieties and animal breeds to be used in this manner – that is, the consumer may well be misled as to the true origin of the GI-denominated goods. This could happen if the plant variety or animal breed is cultivated in large quantities outside of the defined area for which the GI is registered. Take, for example, a vodka made using basmati rice grown only in the state of Punjab in India. A GI “Basmati” is registered, in respect of this vodka, to denote that it is made from basmati rice grown in the Punjab region. Let us assume that it can be demonstrated that the consumer knows that basmati is a type of rice that is grown, not only in several states in India, but also in Pakistan. The end result is that the GI “Basmati” would not clearly indicate to the consumer that the specific vodka has a connection with the terroir because it is made from basmati rice grown only in Punjab. Consumers could well think, upon seeing the GI “Basmati” that the vodka also originates from Pakistan or other states in India, when in fact, the vodka only has a connection with the state of Punjab.

40 It is due to the need to guard against such deception, that the GIA only allows the registration of GIs that contain the names of plant varieties or animal breeds where such registration will not conduce to consumer confusion as to the true origin of the product. Indeed, consumer protection is one of the key policies underlying the GIA. As stated by Ms Indranee Rajah during the Second Reading of the Geographical Indications Bill 2014:

41 This relates to the function of a GI which is defined in s 2(1) of the GIA:

42 In other words, GIs are signs or marks used, in relation to goods, to identify that the goods originate from a particular region or territory, where a given quality, reputation or other characteristic of the goods is essentially attributable to their geographical origin: Law of Intellectual Property of Singapore at para 28.0.1. In this sense, a GI is similar to a trade mark in that both serve to denote the source of goods. By guaranteeing that the goods originate from a particular source or geographical area, both GIs and trade marks serve a consumer protection role.

43 While there are similarities between a GI and a trade mark, some of the key differences between the two bear noting. One such difference is that a trade mark belongs to a particular trader whereas a GI belongs to all traders whose products possess the relevant characteristics as indicated by the GI.

44 Another difference lies in their historical development. Where GIs are concerned, the existing legal frameworks for GI protection evolved from unfair competition laws, as well as from the attempts to prevent fraud in the marketplace for wines in France.

45 The historical development of trade mark law could not be more different. Antecedents to the modern concept of trade marks include the merchant’s mark as well as the craftsman’s mark. The merchant’s mark was a mark that was voluntarily affixed to the goods by their owner. Its purpose was to establish ownership of the goods. Craftsman marks, on the other hand, were more personal in nature as they identified the work of a single craftsman. Statutes, administrative orders, or regulations of a guild or municipality often made it compulsory to affix a craftsman’s mark. Legal recognition and protection of trade marks originally arose from the adjudication of claims in deceit. The notion of a property right in trade marks was developed in later cases, and given “particular impetus in a series of judgments by Lord Westbury, who expressed the view that there was a property right in trade marks, whereby the proprietor could dispose of the right and enforce it even as against innocent infringers”: Neil J. Wilkof and Daniel Burkitt, Trade Mark Licensing (Sweet & Maxwell, 2nd Ed, 2005) at paras 2-05–2-06 and 2-09–2-10.

46 Given these differences between a GI and a trade mark, we would be slow to import principles of trade mark law into the law of GIs. It is for this reason that we do not find persuasive the Consorzio’s reference to s 7 of the Trade Marks Act 1998 (2020 Rev Ed) (the “TMA”), in relation to whether the perspective of the consumer should be taken account of in the assessment of whether the Application GI contained the name of a plant variety or an animal breed. The Consorzio’s argument that the perspective of the Singapore consumer which has been read into provisions in the Trade Marks Act should similarly be read into the GIA finds its roots in the following passage from Susanna H.S. Leong, “European Union-Singapore Free Trade Agreement: A New Chapter for Geographical Indications in Singapore” in Geographical Indications at the Crossroads of Trade, Development, and Culture (Irene Calboli and Ng-Loy Wee Loon (eds)) (Cambridge University Press, 2017) at pp 257–258:

47 Fundamentally, trade marks and GIs are, as Prof Ng-Loy pointed out, distinct species of intellectual property. While it is true that there are similarities between GIs and trade marks, it cannot be simply assumed that the principles governing the refusal of registration of a trade mark under s 7 of the TMA can be wholly imported into the GIA. We agree with Prof Ng-Loy that, in so far as the analysis of s 41(1)(f) of the GIA is concerned, this provision must be examined within the context of the GIA and not the TMA, especially given the specific policy considerations raised by use of the names of plant varieties or animal breeds as GIs.

48 Returning to our analysis, we have thus far concluded that consumer protection is one of the key policy considerations underlying the GIA. We have also concluded that the GIA envisions that the names of plant varieties and animal breeds can be used as GIs, without unduly restricting their use in the course of trade. The use of a name of a plant variety or an animal breed in this manner, however, could well mislead the consumer as to product origination – and so the GIA provides that GIs which contain the name of a plant variety or an animal breed shall not be registered unless it can be shown that the consumer is not likely to be misled.

49  In this regard, although it may appear obvious, we would like to emphasise that all references to “the consumer” in s 41(1)(f) of the GIA are plainly references to the consumer in Singapore which means Singapore citizens and residents and not those who are merely passing through. This is not a contentious point – the common understanding is derived from the fact that the GIA is a Singapore statute and thus, unless it expressly provides otherwise, has only a territorial application. Any protection offered to consumers by the statute must be understood to apply to persons residing in Singapore on a more than transient basis, whether citizens or not, and references in this judgment to the “Singapore consumer” must be understood in that way.

50 One difficulty that may arise when deciding if a GI contains the name of a plant variety or an animal breed is where that name has been changed by some or most of its users. For example, in the present case, the Consorzio had argued before the Judge that the Application GI did not contain the name of a plant variety or an animal breed because Regulation on amending and correcting Commission Regulation (EC) No 606/2009 laying down certain detailed rules for implementing Council Regulation (EC) No 479/2008 as regards the categories of grapevine products, oenological practices and the applicable restrictions, EC Commission Regulation 1166/2009, [2009] OJ L 314/27, had renamed the grape variety, “Prosecco”, as “Glera” so that in Europe the grape is better known by the latter name. However, the change in the scientific or legal name of a plant variety or an animal breed may not immediately result in a change in its use amongst all consumers. In such a case, however, if the evidence is that in Singapore the old name holds sway, it is our view that an objective determination would hold that a GI containing the old name would be a GI containing the name of a plant variety or an animal breed notwithstanding the change in nomenclature by any particular country or producer. Objectively, a plant variety or an animal breed may be known by different names by significant numbers of people. This would mean that the opposition on the grounds of s 41(1)(f) of the GIA could succeed, even if there is evidence to show that many people use the new name exclusively and consider the old one obsolete.

51 Indeed, in framing the nature of the GI as one of the elements to be considered in relation to s 41(1)(f) of the GIA, we would state that this is not a high threshold to be crossed. All that would need to be shown to demonstrate that the GI sought to be registered contains the name of a plant variety or an animal breed, is that the name in question is indeed recognised as the name of a plant variety or an animal breed by a not insignificant population of people. Evidence of this could come from sources such as reputable scientific journals, or legal registers of plant varieties or from the general usage of the term as denoting a plant variety or an animal breed among a body of consumers or producers.

52 Once this element is satisfied, the analysis then turns to consider whether the consumer in Singapore is likely to be misled by the GI that is sought to be registered. As a preliminary point, it is important to be clear as to what exactly the Singapore consumer is being misled about. In the hearing before the PAR, the Consorzio had argued that the phrase “true origin” as used in s 41(1)(f) of the GIA refers to the “product’s true plant (or animal) origin” instead of the “true geographical origin” of the product. The PAR rejected this argument and ruled that the words “true origin” referred to the “true geographical origin of the product”.

53 We agree with the PAR’s reasoning that the nature of this confusion relates to the true geographical origin of the product rather than the true plant origin of the product. As we have discussed above, the function of a GI is to indicate the geographical origin of the product. The GIA recognises, for example, that “Prosecco” can not only be used as a registered GI, but also as the name of a grape variety in the course of trade. The confusion inquiry in s 41(1)(f) of the GIA thus examines whether “Prosecco” can still continue to function qua GI, notwithstanding its other, permissible, use in the course of trade.

54 There is one other reason as to why the phrase “true origin” in s 41(1)(f) must be taken as referring to the “true geographical origin of the product”. This was raised by Prof Ng-Loy in her written submissions. She pointed out various instances where the draftsman of the GIA had regarded the term “true origin” as having the same meaning as “true geographical origin” and this was evident from how Singapore had implemented its international obligations under the TRIPs Agreement and the EUSFTA. To cite an example, in relation to a registered GI identifying an agricultural product or foodstuff other than a wine or a spirit, Art 10.19(3)(a) of the EUSFTA requires Singapore to prohibit the use of the GI in relation to a like product which does not originate from the place indicated by the GI even where the “true origin of the good is indicated”. This finds expression in s 4(6)(a) of the GIA which prohibits, in relation to a registered GI identifying an agricultural product or foodstuff other than a wine or a spirit, the use of such GI in relation to any agricultural product or foodstuff in the same category which does not originate from the place indicated by the GI, whether or not the “true geographical origin of those goods” is used together with the registered GI.

55 The focus of the inquiry under s 41(1)(f) of the GIA is thus whether the Singapore consumer is likely to be misled, by the GI sought to be registered, as to the true geographical origin of the goods. The relevant question to ask, as the Judge framed it, is whether the Application GI is likely to mislead consumers into thinking that “Prosecco” wine could only originate from the Specified Region when, in fact, their true origin could be other geographical locations where the “Prosecco” grape variety is used to make wine.

56 We agree with the way in which the Judge framed the inquiry. The function of a GI is, after all, to indicate that the product originates from a specified region. Such a GI, which contains the name of a plant variety or an animal breed, would be likely to mislead the consumer if the plant variety or animal breed is cultivated in areas outside of the relevant specified region. This GI would be an instrument of deception because it represents to consumers that the product, made from the animal breed or plant variety of the same name, only comes from the relevant specified region when, in truth, it could also come from areas outside that region. Having said this, whether the GI is an instrument of deception will turn very much on what the Singapore consumer is aware of. The way in which the consumer understands the GI, and what it represents about the product’s true geographical origin, will depend on general knowledge since we are concerned about the average consumer and not someone with a specialist knowledge of wines. In this vein, there are at least three factors that must be taken into account in determining whether such a GI is likely to mislead the Singapore consumer as to the true geographical origin of the product.

57 The first is whether the average consumer here is even aware that the name in question is indeed the name of a plant variety or an animal breed. As a matter of logic, it stands to reason that if the consumer does not even perceive or believe the name in question to be that of a plant variety or an animal breed, then it is unlikely that any operative deception would arise. For instance, if the Singapore consumer does not know that “Prosecco” is also the name of a plant variety (or more specifically, a grape), then all that a consumer in Singapore is likely to understand, when she sees the GI, “Prosecco”, is that it refers to a wine originating from the Specified Region. The Singapore consumer would not associate “Prosecco” with a type of grape that is cultivated in commercial quantities outside the Specified Region.

58 The second factor is whether the Singapore consumer is aware that the plant variety or animal breed in question is involved in the production of the product over which GI protection is sought. If the consumer does not associate the plant variety or animal breed with the product in question, then it is unlikely that she will be misled as to the true geographical origin of the product. To use the present case as an example, it is relevant to ask whether the consumer knows that the “Prosecco” variety of grape is used in the production of the wine, “Prosecco”. If the Singapore consumer is shown to be ignorant as to the type of grape used to produce “Prosecco”, it cannot be said that she will be misled as to the true geographical origin of “Prosecco” wine.

59 The third factor concerns the GI that is sought to be registered. A GI that is identical with the name of the plant variety or animal breed would convey a very different message to the Singapore consumer as opposed to a GI that contains other words in addition to the name of the plant variety or animal breed. As Prof Ng-Loy pointed out, using the example of the Bardsey variety of apple tree which was first discovered on Bardsey Island off the coast of Wales, England, the message conveyed by the GI “Bardsey” for cider, is likely to be different from the message conveyed by the GI “Wales Bardsey” for cider. Similarly, in the present case, we are only concerned with the message conveyed by the Application GI, “Prosecco”, and not a GI named “Italian Prosecco”. The latter clearly conveys a very different message from the former.

60 While we have listed three factors which we think are relevant to the inquiry under s 41(1)(f), we would reiterate that this is not a closed list of the factors that the court can take into consideration. The inquiry is, ultimately, one that must turn on the facts, and the factors we have mentioned above merely serve as guidance to the issues which the court would consider in determining whether the Singapore consumer is likely to be misled.

61 To summarise, we find that the inquiry under s 41(1)(f) of the GIA is to be approached in the following manner. One must first consider whether the name of the GI sought to be registered contains the name of a plant variety or an animal breed. This is to be done on an objective basis. It is only when this element is satisfied that one can move on to consider whether the GI sought to be registered is likely to mislead consumers into thinking that the product could only originate from the specified region when, in fact, its true origin could be other geographical locations where the plant variety or animal breed used to make the product is found. In answering this question, it is necessary to focus on matters which the Singapore consumer is aware of given that such awareness naturally affects whether the Singapore consumer is likely to be misled by the GI that is sought to be registered.

62 We first deal with a preliminary point as to which party bears the legal burden of proof of establishing that the ground of opposition under s 41(1)(f) of the GIA has been made out. Our view is that it is the party opposing the registration of the Application GI who bears this burden. As set out in r 27 of the Geographical Indications Rules 2019 (the “GIR”), the following must be produced by a party seeking to oppose the registration of a GI:

63 It is clear from r 27 of the GIR that the party who opposes the registration of the GI must produce evidence to support its opposition, barring which the notice of opposition will be treated as though it had not been filed: r 27(3) of the GIR.

64 In this case, that party is AGWI. It must therefore demonstrate that s 41(1)(f) applies because the Application GI contains the name of a plant variety and the Singapore consumer is likely to be misled by the Application GI.

65 To support its contention that “Prosecco” is, objectively, the name of a plant variety, AGWI points to the following pieces of evidence:

(a) There have been documented historical references to “Prosecco” as the name of a grape variety. In particular, AGWI cites two sources of considerable vintage:

(i) Il Roccolo, Ditirambo by Aureliano Acanti, published in 1754, which included “Prosecco” in a list of grape varieties.

(ii) In 1868, Count Marco Giulio Balbi Valier had identified a type of “Prosecco” vine that was better than the others.

(b) Authorities and official bodies from various countries had recognised “Prosecco” to be the name of a grape variety. For instance, countries such as Argentina, Australia, Bosnia-Herzegovina, Bulgaria, Croatia and Slovenia had requested that “Prosecco” be included as the name of a grape variety in the 2013 Edition of the “International list of vine varieties and their synonyms” published by the International Organisation of Vine and Wine.

(c) Various legislative decrees, international treaties and agreements have recognised “Prosecco” as the name of a grape variety. For example, in the 1994 Agreement between the European Community and Australia on trade in wine, protection was claimed for the GI, “Montello e Colli Asolani” accompanied by the names of various vine varieties which included “Prosecco”.

66 While before the Judge the Consorzio argued that “Prosecco” was not objectively the name of a grape variety, before us it concentrated on contending that a subjective approach must be taken to this issue. It therefore contends that, in assessing whether “Prosecco” is indeed the name of a grape variety, the perspective of the Singapore consumer has to be taken into account. In this connection, its stand is that the various pieces of evidence which AGWI had adduced, did not establish what the Singapore consumer knew. Further, such evidence ought to be given little weight because it comprises legal or technical documents, and so the Singapore consumer would not be aware of the information contained therein. As we have explained (at [38]), we do not accept this as a valid approach.

67 In our view, AGWI has crossed the threshold of showing that “Prosecco” is, objectively, the name of a plant variety. It is, in our view, material that before 1 August 2009, when the variety was renamed “Glera”, “Prosecco” was officially recognised as the name of a vine/grape variety in the EU. This showed that, at the very least, prior to 1 August 2009, “Prosecco” had indeed been recognised in the EU as the name of a grape variety. Also relevant is a publication by Vitis Rauscedo, one of the largest vine nurseries in the world, in the 2007 edition of its General Catalogue. We reproduce an extract of that publication below:

This extract traces the history of the Prosecco vine, and is, in our view, additional evidence demonstrating that “Prosecco” is, objectively, the name of a grape or vine variety notwithstanding that in Europe that variety is now called “Glera”.

68 However, in order for its ground of opposition to succeed, AGWI must also establish that the Singapore consumer was likely to be misled by the Application GI at the time the application was made. This is where, in our view, AGWI’s opposition flounders. The evidence which AGWI has adduced simply does not establish that the Singapore consumer was likely to be misled by the Application GI at that time. We note that AGWI did not produce any evidence of consumer surveys and instead chose to rely on advertising materials as well as statistics showing the increase in the volume of Australian “Prosecco” imported into Singapore. Advertising materials may be useful insofar as they provide some evidence as to how the product for which the GI is being registered has been marketed to the consumer in Singapore. In our view, however, consumer surveys would have been a more direct way of demonstrating whether the Singapore consumer would have been misled by the Application GI.

69 Although we have emphasised the importance of consumer surveys in the inquiry of whether the Singapore consumer is likely to be deceived, merely adducing such surveys showing that the consumer is likely to be deceived is not, in and of itself, determinative of the issue. Consumer surveys can, as one knows, be skewed to reach a certain desired result. Parties should therefore, when adducing evidence of such consumer surveys, also place before the court evidence of how such surveys were conducted. For example, it would be relevant to provide evidence of what questions were asked in the surveys, the demographics of those surveyed and how the survey was conducted. This would allow the court to assess whether the consumer survey can indeed be accepted as evidence that the consumer is likely to be misled by the GI that is sought to be registered.

70 The second piece of evidence which AGWI relies on to show that the Application GI was likely to mislead the Singaporean consumer comprises statistics showing the increase in the volume of Australian “Prosecco” imported into Singapore. This found weight with the Judge below, who reasoned that evidence of “significant or commercial quantities of production outside the Specified Region would cross the relevant threshold for the Application GI to be likely to mislead”: the GD at [37]. With respect, we do not agree with the Judge on this point. As we have pointed out above, the inquiry into whether the Application GI is likely to mislead depends very much on what the Singapore consumer is aware of. Figures showing an increase in the volume of Australian “Prosecco” imported into Singapore show, at best, that local wine sellers are managing to sell more Australian “Prosecco” than previously, and that there is growing local demand. But such evidence does not shed light on the material inquiry, ie, whether, for example, Singapore consumers might be aware that “Prosecco” is also the name of a grape variety used to make wine of the same name.

71 During the hearing, counsel for the Consorzio, Mr Sivagnanaratnam Sivananthan (“Mr Sivananthan”), highlighted that when the Consorzio had pointed out, via a Statutory Declaration made by its President, Mr Stefano Zanette, that the grape variety used in the Australian “Prosecco” sold by Straits Wine was “Glera”, Mr Patrick Sng, the Group Chief Executive Officer of Straits Wine Company Pte Ltd, a company in the business of wine distribution and retail in Singapore had explained, in reply, that this was an error because Straits Wine had not previously placed much emphasis on its online business. The point Mr Sivananthan was making was that when AGWI filed the opposition on 9 September 2019, their own evidence showed that what had been marketed to the Singapore consumer was that “Glera” was the variety of grape used to produce the wine “Prosecco”.

72 It was clear to us from the advertising materials which AGWI had adduced, that very often what had been marketed to the Singapore consumer was that “Prosecco” wine was produced with a variety of grape called “Glera”. We attach a few examples of such marketing:

73 To be fair, some of the listings did also specify that the name of the grape variety used was “Glera (Prosecco)” or “Prosecco”. But even this was, in our view, insufficient to establish that the Singapore consumer would be misled as to the origin of the wine “Prosecco”. The evidence which AGWI had adduced was advertising material. To establish that the Singapore consumer would be misled, it must be assumed that the consumer would peruse the advertising material and notice that the grape variety “Prosecco” had been used to make the wine of the same name. No evidence was before the court, however, which demonstrated the extent of user engagement with such advertising material. It is, therefore, impossible to infer the sort of knowledge which the Singaporean consumer would have had of the name of the grape variety used to produce the wine “Prosecco”.

74 The conclusion which we ultimately reach is that AGWI has failed to establish that the Application GI is likely to mislead the Singapore consumer as to the true geographical origin of “Prosecco”.

75 We therefore allow the appeal. The order of the Judge below is set aside. The result is that the Application GI must be allowed to proceed to registration.

76 The Consorzio, having succeeded in its appeal, is entitled to costs here and below, as well as before the PAR. As for the quantum of such costs (including disbursements), the Consorzio proposes the figure of $150,804.09. AGWI submits that a fairer sum would be $77,000. Given the complexity of the issues raised in this appeal and below, we fix the costs of this appeal together with the hearings before the Judge and the PAR at $120,000 (inclusive of disbursements) to be paid by AGWI to the Consorzio.

77 Finally, we record our deep appreciation to Prof Ng-Loy for her clear and cogent submissions, which greatly assisted us in navigating the issues raised in this appeal. We would also commend counsel for both parties for their well-written and researched submissions, although we did not necessarily agree with every point that had been made.