1 Linking the origin of a product to its quality has long-established historical roots. Léo Loubère, in The Wine Revolution in France – The Twentieth Century (Princeton University Press, 1990) at p 114, traces the practice to ancient Greek and Roman efforts at wine regulation, and remarks that the connection between geography and quality had become “a widely accepted belief, if not to say myth,” by the nineteenth century. Dev Gangjee, in Relocating the Law of Geographical Indications (Cambridge University Press, 2012) at pp 93–96, attributes the prompt for the modern French system of regulation to the crisis experienced in French vineyards in the late nineteenth century, brought on first by fungal diseases, and then a pandemic of phylloxera. Following the dramatic destruction of French vineyards, counterfeits flourished in tandem with the decrease in the production of wine. France enacted a law in 1905 prohibiting the misuse of wine names to address fraud in the wine marketplace. What began as a tradition to mandate quality-related criteria, such as the controlled planting and cultivation of vines and their yields, grew into a basis for delimiting the products whose quality was attributable to their specific region of production. Over time, different jurisdictions used a variety of methods to ensure the accuracy of geographical source as an indicium of specific quality or reputation for a wide range of products. Such attributions are deeply historical. Not all geographical locations serve to indicate specific reputational characteristics of a product. Conversely, not all geographical indications need be names of geographical locations; non-geographical names may, through tradition, nevertheless function to identify a product’s geographical source.

2 This perception of origin as a proxy for quality was used to economic advantage not only by canny merchants, but also by nation-states in their relations with each other. In the eighteenth and nineteenth centuries, reports Paul Duguid in “Networks and Knowledge: The Beginning and End of the Port Community Chain, 1703-1860” (2005) 79(3) Business History Review 453, British customs officials were careful to verify the origin of Port wine because such wines from Portugal were favoured with preferential duties in England.

3 The modern global economy has accentuated the value of such reputational links between products and their sources. At the same time, the nature, conceptual approach, scope and institutional forms of protection available are diverse across jurisdictions. As a result, provisions to protect geographical indications feature within many international trade agreements. Susanna Leong recounts in Intellectual Property Law of Singapore (Academy Publishing, 2013) at para 34.005 that the history of these international agreements, from the Paris Convention 1883, to the Madrid Agreement (Indications of Source) 1891, the Lisbon Agreement 1958, and the World Trade Organization (“WTO”)’s Agreement on Trade-Related Aspects of Intellectual Property Rights (“TRIPS”) in 1994, “tells the story of a relentless campaign for more than a century by wine producers (French in particular) to secure comprehensive international protection against misappropriation and misuse of GIs”.

4 These negotiations recognise that within each jurisdiction, protection depends upon the specific legal framework in place. Whether a sign functions as a geographical indication is a matter of national law. In 1998, Singapore enacted its first Geographical Indications Act – the Geographical Indications Act 1998 (Act 44 of 1998), later replaced by the Geographical Indications Act (Cap 117B, 1999 Rev Ed) (“the 1999 GIA”) – to comply with its obligations under TRIPS. This 1999 GIA did not set up a registration system; it granted the owner of any indication that fulfilled the definition of a geographical indication (“GI”) the ability to file suit for protection. In 2014, in compliance with Singapore’s obligations under the European Union – Singapore Free Trade Agreement, the 1999 GIA was repealed and replaced by the Geographical Indications Act 2014 (Act 19 of 2014) (“the GIA”). This GIA further enhanced the protection of GIs. As part of this enhanced protection, a registration system was set up at the Intellectual Property Office of Singapore.

5 The present appeal arises out of an application (“the Application”) made by the Consorzio di Tutela della Denominazione di Origine Controllata Prosecco (“the Consorzio”) to register “Prosecco” as a GI denoting wines originating from a specified region in Northern Italy (“the Application GI”). ¹

6 The Consorzio is a consortium established and organised under the laws of Italy, ² and is responsible for protecting, promoting, marketing and generally overseeing the use of the term “Prosecco”. ³ On 3 May 2019, it applied to register “Prosecco” as a GI in respect of wines in Singapore. The claimed geographical area for the production of “Prosecco” wines was “the North East region of Italy, and include[d] the entire territory of Belluno, Gorizia, Padova, Pordenone, Treviso, Trieste, Udine, Venice and Vicenza” (“the Specified Region”). ⁴ The Application GI was accepted and published in the Geographical Indications Journal under Geographical Indication No 50201900088S on 21 June 2019.

7 Australian Grape and Wine Incorporated (“AGWI”) is the representative body for grape growers and winemakers in Australia. ⁵ On 9 September 2019, AGWI filed a notice of opposition against the registration of the Application GI. ⁶ It is not disputed that Italian “Prosecco” wines have a longer history of local distribution. Australian “Prosecco” wines have been available in Singapore, albeit in smaller quantities, since at least 2015. The Consorzio asserts that “Prosecco” has been used in Singapore as a geographical indication for the Specified Region since at least 2010.

8 AGWI relied on two grounds in its opposition to the registration: first, s 41(1)(a) of the GIA, that the Application GI did not fall within the meaning of “geographical indication” as defined in s 2(1) of the GIA; and second, 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. The relevant provisions in s 41(1) of the GIA read as follows:

9 On 4 May 2021, the Principal Assistant Registrar (“the PAR”), disagreeing on both grounds, dismissed AGWI’s opposition and ordered that the Application GI should proceed to registration. ⁷ Her Full Grounds of Decision (“the PAR’s Full Grounds”) were issued on 12 August 2021.

10 On 6 September 2021, AGWI applied for the PAR’s decision disallowing its opposition to the Application GI to be reversed. On 12 November 2021, I allowed AGWI’s appeal on the basis of s 41(1)(f) of the GIA but not s 41(1)(a) of the GIA. These are my reasons for doing so.

11 I deal first with s 41(1)(f), which contains two conjunctive requirements:

12 The PAR found that “Prosecco” was the name of a grape variety as at 3 May 2019 when the Application was filed (“the Relevant Date”) and held that the first requirement was satisfied. At the same time, she held that the Application GI was not likely to mislead the consumer as to the true origin of the product. As a result, the opposition was not made out. AGWI supported her first finding but not the second. The Consorzio, on the other hand, disagreed with her first finding but agreed with her second.

13 AGWI submitted that the PAR was correct to find that the term “Prosecco” was the name of a grape variety as at the Relevant Date. AGWI relied on the following evidence which, it argued, established the incontrovertible fact that “Prosecco” was the name of a grape variety: ⁸

14 The Consorzio submitted that the PAR erred in finding that “Prosecco” was the name of a grape variety because, as at the Relevant Date, “Prosecco” was not recognised by consumers in Singapore as the name of a grape variety. ¹¹ The Consorzio disagreed with the PAR’s interpretation that the first limb of s 41(1)(f) of the GIA was satisfied so long as some or any countries in the world used the term “Prosecco” as the name of a grape variety. Instead, the Consorzio submitted that this required AGWI to show how the average end-consumer in Singapore perceived the term when used in relation to wines, having regard to evidence in Singapore. ¹² Applying this approach to s 41(1)(f) of the GIA, the Consorzio argued that the Singapore consumer recognised the term “Prosecco” to be a GI designating wines from Italy only. The correct term of reference for the grape variety in issue was “Glera”, as EU Regulation 1166 had renamed the grape variety “Prosecco” as “Glera” and provided that these amendments were to be applied retroactively as of 1 August 2009.

15 Arising from the parties’ arguments, the question of whether “Prosecco” was a grape variety comprised two issues:

I deal with each in turn.

16 The Consorzio contended that the relevant approach was to ask whether the term “Prosecco” was recognised as the name of a grape variety by the average consumer in Singapore, for four reasons: ¹³

(a) First, the essential purpose of GIs lay in safeguarding the interest of consumers through providing assurances to consumers of the geographical origin, quality and characteristics of the product in question. Flowing from this, where the registration of a GI was sought in Singapore, it was the perspective of the Singapore consumer that would be relevant in assessing the registrability of potential GIs under the grounds set out in s 41(1) of the GIA.

(b) Second, the language of s 41(1)(f) of the GIA – which requires the court to consider whether the GI which contains the name of a plant variety is likely to mislead the consumer – strongly implied that whether “Prosecco” was the name of a grape variety should be determined from the perspective of the average consumer in Singapore. The Consorzio substantiated this by stating that if the average consumer in Singapore did not recognise the term as the name of a grape variety, then there could be “no meaningful analysis regarding… whether the term [was] likely to mislead”. ¹⁴

(c) Third, the PAR’s approach was arbitrary, particularly where there was clear disagreement as to whether “Prosecco” was the name of a grape variety, as it was not clear how a Singapore court should decide which country or region to follow in determining whether “Prosecco” was recognised as the name of a grape variety.

(d) Fourth, a purposive interpretation of s 41(1)(f) of the GIA favoured an approach where the question of whether a term was the name of a plant variety was assessed from the perspective of consumers in Singapore.

17 The plain words of the section did not lend weight to the Consorzio’s emphasis. The view of the consumer is not expressed as part of the first limb. The emphasis of the first limb is factual: whether the GI “contains the name of a plant variety or an animal breed” [emphasis added]. The Consorzio’s approach would add extra words, and there was no basis in the language of s 41(1)(f) of the GIA for the additional requirement to be implied. Instead, the perspective of the Singapore consumer was relevant only to the second limb of s 41(1)(f) of the GIA, in ascertaining whether the GI was “likely to mislead the consumer as to the true origin of the product” [emphasis added]. The interests of the consumer are safeguarded through the second limb. A statutory provision ought to be given its plain and ordinary meaning. A purposive construction becomes particularly relevant only if there are two or more possible interpretations of a given legislative provision: see Tan Cheng Bock v Attorney-General [2017] 2 SLR 850 at [36].

18 A further argument made by the Consorzio was that its interpretation of s 41(1)(f) of the GIA was necessary to make sense of s 15(b) of the GIA. Section 4 of the GIA allows an interested party of goods identified by a GI to bring an action against another for using the GI in one of the ways specified in s 4(2) of the GIA. Division 2 of the GIA, which starts with s 15 of the GIA, lists exceptions applicable to registered GIs. Section 15(b) of the GIA is as follows:

19 The Consorzio submitted that the phrase “the name of a plant variety or an animal breed” should bear the same meaning in both s 41(1)(f) and s 15(b) of the GIA. Flowing from this, the Consorzio asserted that it would lead to incongruous and unreasonable outcomes if the first limb of s 41(1)(f) of the GIA were satisfied simply because a GI contained the name of a plant variety or animal breed in some or any country. The Consorzio argued that this would render the registration of such GIs otiose. This was because, despite the registration, any interested party would not be able to enforce the rights in the registered GI against third parties in Singapore who used the GI in the course of trade in some or any country which recognised the term as the name of a plant variety or animal breed, even if the significance of that term was not recognised or understood as such by Singapore consumers. Instead, under the Consorzio’s proposed interpretation, the exceptions to a registered GI holder’s rights under s 15(b) of the GIA would only apply where the significance of the term was recognised and understood as such in Singapore. ¹⁵

20 I agreed that the phrase “the name of a plant variety or an animal breed” should bear the same meaning in both s 41(1)(f) and s 15(b) of the GIA. This, however, did not lead to statutory absurdity, contrary to the Consorzio’s submission. The GIA is consistent in excluding appropriate uses of names of plant varieties from its regulatory regime. This is consonant with the object of GIs to make goods attributable to their geographical origin, and to guard against the wrongful use of indicators of geographical origin. The logic of s 15(b) of the GIA may be seen in applying the section to the example of the registered GI, “Conegliano Valdobbiadene – Prosecco”. Such a GI would not, for example, bar the use of the word “Prosecco” in the course of trade as a plant variety. It is the same policy intent behind the need in s 41(1)(f) of the GIA to ensure that any registration of a plant variety is not misleading to the consumer. The object of the statutory scheme is to ensure that the registration of a GI is not misleading, not only at the point of registration but also post-registration. As such, s 15(b) of the GIA permits any use of registered GIs in the course of trade as a plant variety, as such use would be appropriate use, and not misuse. Therefore, contrary to the Consorzio’s assumption, even if “Prosecco” were registered as a GI, s 15(b) of the GIA serves to limit the Consorzio’s ability to use the remedies provided for under Part II of the GIA where “Prosecco” is used in the course of trade as a plant variety. In my view, this is not surprising once the historical context I highlighted at [1]–[4] is understood. GIs may be drawn from a rich diversity of names and the GIA takes cognisance of the potential that a name of a plant variety could function as a GI. In such a case, the imperative remains that it must guarantee the accuracy of the relevant geographical-reputational attribution. In this respect, I draw a distinction with other forms of intellectual property that protect distinctive marks or names that serve as indications of commercial origin, rather than of location origin. The Consorzio in its submissions placed emphasis on a Singapore-registered mark which contained the words “Prosecco PDO” within a round shape. ¹⁶ The existence of this registered mark did not add to the Consorzio’s argument because the object of a GI is not to protect the use of a distinctive mark or name. GIs are necessarily descriptive: they serve as geographical attributions which accurately link the origins of a product with its distinctive characteristics.

21 Therefore, whether the term “Prosecco” was the name of a plant variety was simply a matter of objective fact. When the relevant inquiry was viewed in this light, the Consorzio’s argument that it was not clear how a Singapore court should decide which country or region to follow in determining whether “Prosecco” was recognised as the name of a grape variety lost any cogency. The issue was simply one of fact to be decided on the evidence presented. I turn then to the salient query.

22 In my view, the PAR was correct to find that “Prosecco” was the name of a grape variety as at the Relevant Date, in view of the extensive evidence adduced by AGWI in support of this proposition (as summarised at [13] above). ¹⁷ Although much of the material AGWI relied on came from sources outside Singapore, this material nevertheless provided evidence that “Prosecco” was, as a matter of fact, the name of a grape variety as at the Relevant Date.

23 The Consorzio submitted that the correct term of reference for the grape variety from which “Prosecco” wines are made was “Glera”, as EU Regulation 1166 had renamed the grape variety “Prosecco” as “Glera” and provided that these amendments were to be applied retroactively as of 1 August 2009. ¹⁸

24 Nevertheless, this name change from “Prosecco” to Glera applied only to the EU. ¹⁹ The impetus for the name change was to prevent confusion between the name of the vine variety “Prosecco” and the name of the Protected Designation of Origin (“PDO”) “Prosecco” in EU law (see Recital (2) to EU Regulation 1166). ²⁰ Contrary to the Consorzio’s submission, EU Regulation 1166 provided clear evidence that “Prosecco” was, as a matter of fact, the name of a grape variety in Europe until its legislative renaming on 30 November 2009. The Italian Ministerial Decree of 17 July 2009 referred to “the synonym ‘Glera’ for the vine variety of ‘Prosecco’”. Thus, the renaming of the grape to “Glera” was neither contradictory nor mutually exclusive with the grape being known as “Prosecco”. The same grape could be, and was, known by more than one name. There was a wealth of evidence which showed that “Prosecco” remained the name of a grape variety outside the EU as at the Relevant Date. In particular, wine merchants in Singapore referred to “Prosecco” as a grape variety. On the evidence before me, I accepted that “Prosecco” was indeed the name of a grape variety at the Relevant Date.

25 The Consorzio’s contention to the contrary was premised on the assertion that that “Prosecco” was not recognised by the Singapore consumer as a grape variety, but instead as a GI designating wines from Italy only. I have explained at [17]–[21] why an objective approach, and not a subjective one, was to be used to decide this factual issue. In any event, the Consorzio’s assertion here was not supported by evidence. As counsel for the Consorzio acknowledged during the hearing, no market test was carried out to ascertain whether Singapore consumers recognised “Prosecco” to be a grape variety, and there was no concrete evidence of Singapore consumers perceiving “Prosecco” as a GI designating wines from Italy only. Instead, counsel for the Consorzio essentially urged me to infer consumer perception from the presence of Italian “Prosecco” from 2010 in the Singapore market, and wine traders’ trade advertisements and labelling of their products. ²¹ This was not sufficient to substantiate the Consorzio’s assertion that the evidence from outside Singapore of “Prosecco” being the name of a grape variety ought to be disregarded because consumers in Singapore did not recognise it as such. Indeed, there was evidence to the contrary, which I highlight at [37] and [41] below.

26 I therefore found that AGWI had proven that “Prosecco” was the name of a grape variety as at the Relevant Date, such that the Application GI contained the name of a plant variety for the purposes of the first limb of s 41(1)(f) of the GIA.

27 I turn then to the second limb of s 41(1)(f) of the GIA, which requires the court to consider whether the Application GI is “a geographical indication which … is likely to mislead the consumer as to the true origin of the product”.

28 The PAR was not satisfied that there was a likelihood that the consumer would be misled as to the true geographical origin of the product. In her decision dated 4 May 2021, the PAR acknowledged that, although the grape variety “Prosecco” may have originated from Italy, it had since “left its cradle of origin” and was now cultivated in commercial quantities in other countries, such as Australia. ²² Australian wines made from the “Prosecco” grape variety had been exported to Singapore since 2015. Nevertheless, the PAR made the following findings in her Full Grounds:

(a) AGWI’s evidence was insufficient to prove that, on the Relevant Date, the Singapore consumer was aware that “Prosecco” wines could originate from both Australia and Italy. For a period of at least four years before the Relevant Date, consumers in Singapore had been exposed to “Prosecco” wines from Australia as well as from the Specified Region in Italy, but (based on export volume) this exposure did not appear to be very intensive (PAR’s Full Grounds at [51]–[55]). ²³

(b) There was no evidence to support a finding that the Application GI was likely to mislead the consumer as to the true geographical origin of the product (PAR’s Full Grounds at [62]–[65]): ²⁴ First, no evidence had been lodged to show that consumers had actually been misled, although Australian “Prosecco” wines had been sold alongside Italian “Prosecco” wines from the Specified Region for at least four years in Singapore before the Relevant Date. Second, because the distinctions made between the goods concerned (wines to be imbibed) were often matters of consumer preference, consumers were likely to pay a relatively high degree of attention to the purchase of such goods, and this reduced the likelihood of consumers being misled as to their origin. Third, it was industry practice in Singapore for the wines concerned to be marketed with accompanying descriptions of the relevant grape variety and region of production. The PAR further found that consumers were unlikely to be misled into thinking that a bottle of Italian “Prosecco” came from Australia or vice versa (PAR’s Full Grounds at [64]). ²⁵ Thus, as framed by the PAR, the essential question was whether the Application GI was likely to mislead consumers into thinking that “Prosecco” wines from Australia could originate from the Specified Region in Italy.

29 AGWI submitted that the PAR erred in finding that the Application GI was not likely to mislead the consumer as to the true origin of the product, and that the PAR applied the wrong interpretation of the phrase “likely to mislead the consumer as to the true origin of the product”. ²⁶

30 First, AGWI contended that the PAR failed to consider the relevant EU authorities which shed light on the legislative history, purpose and proper interpretation of s 41(1)(f) of the GIA (which was derived from Art 6(2) of the EU’s Regulation on Quality Schemes for Agricultural Products and Foodstuffs, EU Council Regulation 1151/2012, [2012] OJ L 343/1 at p 9 (“EU Regulation 1151”)). ²⁷ Based on various EU authorities, AGWI argued that there was a paramount public interest in protecting the freedom of producers and traders to use the names of plant varieties and animal breeds to legitimately describe their products, such that the registration of the names of plant varieties and animal breeds as GIs should only be allowed in the exceptional case where the variety or breed in question has remained exclusively confined to its territory of origin. AGWI contended that the name of a plant variety would be misleading (such that it could not be registered as a GI) where it had been produced or cultivated in commercial quantities outside the claimed geographical area. ²⁸

31 On this basis, AGWI argued that the mischief targeted by s 41(1)(f) of the GIA was the registration of a GI containing the name of a plant variety when the said plant variety was being cultivated and produced in commercial quantities outside the relevant region claimed by the GI. Such a GI would inevitably mislead the consumer into thinking that the relevant goods could only originate from the region claimed by the GI, when that was in fact untrue. AGWI contended that the PAR erred in asking whether consumers were likely to be misled into thinking that a bottle of Italian “Prosecco” came from Australia or vice versa ²⁹ (PAR’s Full Grounds at [64]). ³⁰

32 AGWI submitted that, on the facts of the present case, “Prosecco” functioned as a guarantee to consumers that “Prosecco” wines originated exclusively from the Specified Region in Italy. However, as at the Relevant Date, the “Prosecco” grape variety had left the Specified Region and was being cultivated in significant and commercial quantities in other countries (such as Australia) for wine-making purposes. As “Prosecco” was also the name of a plant variety, and since the “Prosecco” plant variety was being cultivated and produced in commercial quantities outside the Specified Region, the Application GI was likely to mislead consumers who relied upon it as a badge of geographical origin into thinking that the true origin of “Prosecco” wines could only be the Specified Region, when this was in fact untrue. ³¹

33 The Consorzio submitted that the PAR was correct to find that the Application GI was not likely to mislead the consumer as to the true origin of the product, such that the PAR’s dismissal of this ground of opposition should be affirmed. This was because s 41(1)(f) of the GIA required the court to assess whether the application GI, and not the registration of the application GI, was likely to mislead the consumer as to the true origin of the product (which here referred to the product designated by the GI in question, and not the entire category of goods). In this case, the consumer in Singapore was not likely to be misled as to the true origin of the product for two reasons: first, even if “Prosecco” was the name of a grape variety, the consumer in Singapore did not associate the grape variety with any other country or geographical place, and was not aware that the grape variety (and wines made therefrom) could originate from other countries like Australia. Second, given the degree of recognition in Singapore of the Application GI and the reputation of the goods offered thereunder, as well as their nature (it being not uncommon for consumers of wines to pay great attention to their country of origin), the consumer in Singapore was not likely to be misled as to the true geographical origin of the product (ie, Italy). ³²

34 In my view, the relevant question was whether the Application GI was likely to mislead consumers into thinking that “Prosecco” wines could only originate from the Specified Region, when in fact their true origin could be other geographical locations where the “Prosecco” grape variety was used to make wines. This approach cohered with the “essential function” of a GI, which was “to guarantee to consumers the geographical origin of the goods and the specific qualities inherent in them”: The Tea Board v European Union Intellectual Property Office (EUIPO) (Delta Lingerie intervening) [2018] Bus LR 1095 at [56]. The very definition of a GI in s 2(1) of the GIA requires that the indication is “used in trade to identify goods as originating from a place”. Similarly, as explained during the Second Reading of the Geographical Indications Bill (Bill No 13 of 2014), GIs are “terms which we use to inform consumers that a product comes from a particular place” (Singapore Parliamentary Debates, Official Report (14 April 2014) vol 91 (Indranee Rajah, Senior Minister of State for Education and Law)). A GI would therefore be misleading, and as such unregistrable, if the product could in fact come from a different place. This would be the case if the term contained the name of a plant variety or animal breed which was used to produce and describe the relevant product outside of the place denoted in the GI.

35 This was consistent with the position adopted in EU law, which was relevant to the interpretation of the GIA because s 41(1)(f) of the GIA was adapted from Art 6(2) of EU Regulation 1151 (as was expressly indicated in the draft of the proposed legislative amendments to the GIA annexed to the Intellectual Property Office of Singapore’s consultation paper dated 1 November 2013). ³³ Article 6(2) of EU Regulation 1151 is in pari materia with s 41(1)(f) of the GIA, save that the former refers to the name of a GI which “conflicts” with the name of a plant variety or animal breed whereas the latter refers to a GI which “contains” the name of a plant variety or an animal breed. ³⁴ It has been recognised, for the purposes of EU law, that “names of plant varieties and animal breeds should remain available to breeders and producers in all locations”. To achieve this, EU Regulation 1151 excludes from registration not only names that are identical to the names of plant varieties or animal breeds, but all similar names that are likely to mislead the consumer as to the true origin of the product: see Justine Pila & Paul L C Torremans, European Intellectual Property Law (Oxford University Press, 2016) at p 473. ³⁵ Where the names of plant varieties or animal breeds conflicted with potential GIs, the GI might nevertheless be registered as a PDO in EU law on the basis that no confusion for consumers had been established. For example, where “it was shown that the product in question had been traditionally produced exclusively in the specific geographical area, [it would be] clear that the use of the name as a PDO would not confuse the consumer” [emphasis added]: Research Handbook on EU Agriculture Law (Joseph A McMahon & Michael N Cardwell eds) (Edward Elgar Publishing, 2015) at p 292. ³⁶

36 In my view, this illustrated the general public interest in protecting the freedom of producers and traders to use the names of plant varieties and animal breeds to legitimately describe products made from these sources. ³⁷ Further, the EU material shed light on the correct approach to the inquiry into whether a GI was likely to mislead the consumer as to the true origin of the product, by indicating that the relevant question was whether consumers were misled into equating products made from a particular plant variety or animal breed with products originating from a particular geographical location.

37 In the present case, I agreed with AGWI that the Application GI, containing as it did the name of a grape variety, would be misleading if “Prosecco” grapes had been cultivated and “Prosecco” wines produced in significant or commercial quantities outside the Specified Region. While I was able to accept, as a matter of principle, the Consorzio’s argument that a GI would not automatically become misleading once the relevant product left its “cradle of origin”, ³⁸ I was of the view that significant or commercial quantities of production outside the Specified Region would cross the relevant threshold for the Application GI to be likely to mislead. Furthermore, I was satisfied that as at the Relevant Date, “Prosecco” grapes were being cultivated and “Prosecco” wines were being produced in commercial quantities in countries such as Australia, as found by the PAR. ³⁹ This was clearly demonstrated by the evidence adduced by AGWI of the tenfold increase in the export volumes of Australian “Prosecco” wines to Singapore from 2015 to 2018, with 9,657 litres of “Prosecco” wine from Australian wine producers being exported to Singapore in 2018, ⁴⁰ and of more than 25 different brands of Australian “Prosecco” wines being exported to Singapore. ⁴¹ The Consorzio sought to argue that Australian “Prosecco” wines were not produced in such commercial quantities as to be liable to mislead, ⁴² but I was unable to accept this assertion. Even if Singapore consumers had a greater degree of exposure over a longer period of time to Italian “Prosecco” wines compared to Australian “Prosecco” wines, or might more commonly associate “Prosecco” wines with Italy, this did not mean that the Application GI was not likely to mislead consumers as to the true origin of “Prosecco” wines. Consumers were still likely to be misled by the Application GI into thinking that all “Prosecco” wines originated from the Specified Region in Italy, when in fact some “Prosecco” wines originated from Australia.

38 In this regard, the Application GI ought to be contrasted with the indications “Conegliano Valdobbiadene – Prosecco”, “Conegliano – Prosecco” and “Valdobbiadene – Prosecco”, which are registered as GIs. Conegliano and Valdobbiadene are both towns in the Veneto region of Italy. These indications were not liable to mislead consumers as to the true origin of the products they were used to designate, as (for example) “Conegliano – Prosecco” could indeed only originate from the town of Conegliano in Italy.

39 A related question was whether the “product” referred to in s 41(1)(f) of the GIA should be interpreted as the particular class of products specified in the GI application (in this case, “Prosecco” wines originating from Italy), or the broader class of products to which the term sought to be registered as a GI was applied (in this case, all “Prosecco” wines). The Consorzio argued that the correct interpretation was the former. ⁴³ A plain and ordinary reading of the subsection indicated otherwise. It reads, “a geographical indication which … is likely to mislead the consumer as to the true origin of the product”: in context, the last word “product” is referred to directly as that bearing the “geographical indication” at the beginning of the sentence. The Consorzio’s reading put an additional gloss on the section. Further, it was a circular argument. On such a reading, the Application GI (which was sought to be registered in respect of wines originating from Italy), by definition, could not mislead the consumer as to the true origin of “Prosecco” wines originating from Italy. Instead, in my view, the “product” as to whose true origin consumers might be misled under s 41(1)(f) of the GIA had to refer to the entire class of products to which the GI was to be applied, ie, all “Prosecco” wines.

40 In its oral submissions, the Consorzio maintained its position on the basis that s 41(1)(f) of the GIA referred to a “geographical indication”, which was defined in s 2(1) of the GIA. That argument, too, was circular. The issues for consideration within s 41(1)(f) of the GIA were framed on the assumption that the indication being opposed under that same subsection was indeed a GI. The correct ground for contesting whether any GI fulfilled s 2(1) of the GIA was s 41(1)(a) of the GIA, which specifically mentions this ground of challenge (and which is dealt with below, at [43]–[56]). The relevant question framed by s 41(1)(f) of the GIA was therefore whether, if the Application GI was not refused registration as a GI, the Application GI was likely to mislead the consumer as to the true origin of any wine marked as “Prosecco”. This was the fundamental question if a geographical attribution was to have its proper meaning.

41 The difficulty in the Consorzio’s argument is reflected in the PAR’s Full Grounds (at [64]–[65]), where, in assuming the same position as the Consorzio, she found that there was no confusion in the sale of Australian and Italian “Prosecco” because of consumer awareness and local industry practice in marking the two types of wines with accompanying descriptions of the relevant grape variety and region of production. However, the concurrent sale of Australian and Italian “Prosecco” defeats the purpose of “Prosecco” as a geographical indication. The very reason wine merchants use the differentiating markings, and consumers pay attention to such descriptions, is to avoid confusion as to product origin. In each case the merchants or customers realise the word “Prosecco” could refer to a grape variety, or more casually, as akin to a varietal wine. Concurrent use of the same GI in respect of different locations is not envisaged under the GIA. To the contrary, s 42 of the GIA permits the Registrar to impose various practical conditions when registering a geographical indication that is homonymous in relation to an earlier GI, so as to differentiate the homonymous GI from the earlier GI. And the protection afforded to GI-owners for wine extends further, beyond prohibiting confusion as to location of origin. Within both the 1999 GIA and the GIA, the use of geographical indications for wine is prohibited even in circumstances where the true geographical origin is clearly marked, used in translation, or accompanied by clarifying words such as “kind”, “style”, or “imitation”: see s 3(2)(c) of the 1999 GIA, and s 4(2)(c) of the GIA.

42 Thus, in my judgment, the second limb of s 41(1)(f) of the GIA was also satisfied. Accordingly, the registration of the Application GI ought to be refused under s 41(1)(f) of the GIA.

43 I turn to the second main issue of the appeal, namely, whether the ground for refusal of registration in s 41(1)(a) of the GIA was established. I did not agree with AGWI on this ground. I deal with it briefly, as it was relevant to the issue of costs.

44 Section 41(1)(a) of the GIA provides that indications which do not fall within the meaning of “geographical indication” as defined in s 2(1) of the GIA shall not be registered. Section 2(1) of the GIA reads as follows:

45 There was no dispute that the Specified Region in the Application GI was that of a region in a “qualifying country”. Italy has been a member of the WTO since 1 January 1995 and a party to the Paris Convention since 20 March 1883. AGWI’s opposition rested on two grounds: first, that the Application GI was not used in the course of trade to identify goods as originating from a place; and second, that the quality and reputation of the goods was not attributable to that place.

46 The PAR found that the Application GI fell within the meaning of “geographical indication” under s 2(1) of the GIA. It had been used in trade to identify goods as originating from the Specified Region (PAR’s Full Grounds at [74]). ⁴⁴ Further, the PAR was not satisfied that AGWI had substantiated its claim that “Prosecco” wines did not have any qualities, reputation or other characteristics that were essentially attributable to the Specified Region. AGWI had relied, in particular, on a 2019 research report published by the Faculty of Law of Monash University, titled “The European Union’s Attempts to Limit the Use of the Term ‘Prosecco’” (“the Monash Report”). ⁴⁵ The PAR observed that AGWI’s case rested only on a single piece of evidence, and declined to attach too much weight to the Monash Report for two reasons: first, it had been prepared at AGWI’s request and could be biased; and second, the writers of the Monash Report did not themselves make the statutory declaration in which it was exhibited (PAR’s Full Grounds at [86]–[87]). ⁴⁶

47 AGWI submitted that the Application GI did not fulfil the definition of a GI in s 2(1) of the GIA.

(a) First, AGWI contended that the phrase “used in trade to identify goods as originating from a place” in s 2(1) of the GIA required the Application GI to identify and guarantee to consumers that they were obtaining wines from the Specified Region in Italy and no other location. AGWI argued that, in light of its evidence of the growing demand for Australian Prosecco in Singapore, the Consorzio did not establish that consumers perceived “Prosecco” as denoting wines coming exclusively from the specified region. Therefore, AGWI took issue with the PAR’s finding ⁴⁷ that the definition of a GI in s 2(1) “is not concerned with how the indication is perceived by the consumers”, contending that the PAR’s view “adopts an extremely lax standard which will result in completely undeserving indications being accorded GI protection”. ⁴⁸

(b) Next, turning to limb (b) of the s 2(1) definition of a GI, AGWI argued that “Prosecco” wines did not have any qualities, reputation or other characteristics that were essentially attributable to the Specified Region (which occupied a large geographical area, and within which the climate and wine-making methods differed greatly). AGWI contended that it had adduced unrebutted evidence that the qualities and other characteristics of “Prosecco” wines were, instead, owed to the underlying “Prosecco” grape variety. Further, AGWI argued that the PAR erred in finding that the Monash Report was insufficient for AGWI to prove its case on this point. ⁴⁹

(c) Further, the Consorzio had not applied for the registration of the Application GI on the basis that “Prosecco” wines had a reputation essentially attributable to the Specified Region, and that there was nothing in the Consorzio’s application which specified that “Prosecco” wines had a particular reputation essentially attributable to the Specified Region. Accordingly, AGWI was not required to prove that “Prosecco” wines did not have any given reputation that was essentially attributable to the Specified Region. This was because the application for GI registration had to set out the precise basis for seeking registration and the contents thereof constituted the subject matter of challenge in opposition proceedings. In any event, the Consorzio had not filed any evidence to substantiate its claim that “Prosecco” wines enjoyed a particular reputation that was essentially attributable to the Specified Region, as opposed to the underlying grape variety. ⁵⁰

48 The Consorzio submitted that this ground of opposition should fail as the Application GI fell within the meaning of “geographical indication” as defined under s 2(1) of the GIA, in that “Prosecco” was an indication used in trade to identify wines as originating from the Specified Region, and the quality, reputation and/or other characteristics of “Prosecco” wines were essentially attributable to a combination of natural environmental, historical and human factors in the Specified Region. ⁵¹ The Consorzio further argued that no weight should be accorded to the Monash Report and that, contrary to AGWI’s assertions, it had adduced evidence to rebut the claims made in the Monash Report. ⁵² The Consorzio submitted that the correctness of its position was affirmed by the strong and widespread recognition of “Prosecco” as a GI around the world, with Australia being the exception rather than the norm. ⁵³

49 Moreover, the Consorzio argued, the “Prosecco” indication had already been protected in Singapore as an unregistered GI, and registration of the Application GI would correctly reflect what the position was prior to 1 April 2019 (being the date on which the registration regime was implemented). ⁵⁴

50 AGWI argued that the correct interpretation of the first head of s 41(1)(a) required that “Prosecco” identified wines originating exclusively from the Specified Region, drawing upon the ECJ’s decision in Tea Board at [56] that the “essential function of a geographical indication is to guarantee to consumers the geographical origin of the goods”. AGWI’s interpretation of this head of s 2(1) of the GIA put an additional gloss on the words of the section. There was no reason to add the word “exclusively” into s 2(1), or to introduce into the section the perception of the Singaporean consumer. The sentence quoted from Tea Board at [56] was taken out of context. That portion of the judgment did not concern the EU Regulation which was similar to s 2(1) of the GIA, did not deal with the relevance of consumer perception in the definition of a GI, and does not aid statutory interpretation of s 2(1) of the GIA.

51 Returning to the statute, a plain reading of this first limb of 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”. On the available evidence, I agreed that “Prosecco” was an indication used in trade to identify wines originating from the Specified Region. The name “Prosecco” was protected as a GI as a matter of law within Italy and the EU. Its application also contained evidence that linked this GI to the Specified Region. Within the EU, this GI was used to verify its geographical source. Outside the EU, it was also valid as a GI in a range of other jurisdictions, either through trade agreements or through national registration. Therefore, it could not be said that the Application GI was not an “indication used in trade to identify goods as originating from a place”.

52 Relevant to this issue was the Consorzio’s submission that sought to take this indication further in Singapore. The Consorzio contended that, in Singapore, prior to the implementation of the registration regime, the Application GI had been afforded protection as an unregistered GI, and the registration of the Application GI would correctly reflect the position prior to the date the registration regime was implemented, 1 April 2019. The Consorzio had not, however, sought remedies under the 1999 GIA. Whether the Application GI had been afforded protection as a GI prior to 1 April 2019 was therefore an untested question. There was no basis for this submission. The Consorzio did not dispute that the date that was relevant for the purposes of assessing the merits of AGWI’s opposition was the date of the filing of its application to register the Application GI. ⁵⁵ While the Consorzio also relied on its registration of a Singapore trade mark with “Prosecco PDO” within its mark, as I mention at [20], this protected a form of intellectual property that was distinct from a GI, and did not aid it in respect of its application to protect the Application GI.

53 The crucial issue in this case regarding s 41(1)(a) of the GIA, in my view, was whether “a given quality, reputation or other characteristic of the goods is essentially attributable to that place”, as required in limb (b) of the definition of a “geographical indication” in s 2(1) of the GIA. In making its application, the Consorzio adduced evidence to establish that the relevant characteristics of “Prosecco” wines were essentially attributable to the Specified Region in Northern Italy. AGWI contended that there was insufficient evidence in the Consorzio’s Application. A perusal of the relevant part of the Application included information on the special notes of the wine; linked its quality and reputation to requirements for the cultivation of the vine and making of the particular wine within the Specified Region; and detailed the cause-and-effect relationship of the grape variety with environmental factors such as climate and the alluvial soil of the Specified Region. In my view, these specified details were sufficient to sustain the Consorzio’s Application.

54 This did not mean, nevertheless, that the Application was not open to challenge. The relevant point was that AGWI bore the burden of proof in its opposition. Limb (b) of the definition of a “geographical indication” alludes to the European concept of terroir, where the assertion is that it is the particular climate, soil and terrain that is responsible for specific unique characteristics of the wine, and it is this that guarantees the special quality of the geographical attribution. Implicit in this criterion, and the Application, are at least two assertions that an opponent could challenge. These are, first, that the specified characteristics attach to the entirety of the Specified Region; and second, that the resulting product is not found elsewhere.

55 AGWI engaged with both ideas. AGWI contended that the Specified Region was too large a region. This submission was not sufficient on its own to prove the wider substantive point, without more, as the size of the stipulated terrain was not definitive. Rather, the specific question was whether, as a factual matter, the specified qualities were present throughout the Specified Region. While as a practical matter, regulation of a smaller area is easier than regulation of a larger one, as the Consorzio pointed out, “Champagne” covers a larger territory than the Specified Region. AGWI’s further contention was that the specified characteristics were only the result of the grape variety and not the result of the interaction of the soil, climate and terrain of the Specified Region. The Monash Report made the assertion that a “breadth of different growing conditions arose with “differences in terrain and climate”, resulting in wines where “the common factor is the grape variety grown, and not the geographic location or method”. ⁵⁶ Relying on this, AGWI pointed to the lack of consistency in the terrain, climate and production methods in the Specified Region, and contended that the common factor to the product specified was solely attributable to the grape variety used. As part of its arguments, AGWI pointed generally to the significant quantities of Australian “Prosecco” exported to the Singapore market.

56 The fundamental weakness in AGWI’s case was that its arguments on this limb required expert factual evidence. Its case comprised various legal submissions and an article by legal academics, rather than specific factual evidence by wine experts as to the interaction of the “Prosecco” grape variety with elements of the Specified Region’s, or even Australian, terroir. Nor was there any expert analysis of the taste of the various wines resulting from the different locales. The success of exports of Australian “Prosecco” did not lead to any specific conclusion regarding any comparison of “Prosecco” from different terrain. Thus, there was no factual evidence to support AGWI’s assertion that the sole unifying factor in the wines in the market marked as “Prosecco” was their grape variety. Therefore, in my judgment, AGWI did not produce sufficient evidence to satisfy its burden of proof on this limb.

57 I therefore allowed the appeal on the basis of s 41(1)(f) of the GIA. Sections 41(1)(f) and 41(1)(a) of the GIA are disjunctive and independent grounds for the refusal of registration. Having regard to the specific circumstances, I agreed with the Consorzio that I should exercise my discretion to adopt an issues-based approach to costs. The portion of the costs of the appeal and the hearing before the PAR specific to s 41(1)(f) of the GIA were awarded to AGWI and fixed at $21,000. This total quantum included disbursements and took into consideration the amount that the parties previously agreed between

themselves in respect of the costs of appearance before the PAR.