N V SUMATRA TOBACCO TRADING COMPANY V BRITISH AMERICAN TOBACCO (BRANDS) INCORPORATED CA761/2008

N V SUMATRA TOBACCO TRADING COMPANY V BRITISH AMERICAN TOBACCO (BRANDS) INCORPORATED CA761/2008 24 February 2010IN THE COURT OF APPEAL OF NEW ZEALAND CA761/2008 [2010] NZCA 24BETWEEN N V SUMATRA TOBACCO TRADING COMPANY Appellant AND BRITISH AMERICAN TOBACCO (BRANDS) INCORPORATED Respondent Hearing: 1 October 2009 Court: Glazebrook, O'Regan and Baragwanath JJ Counsel: D L Marriott for Appellant A H Brown QC and B J Shone for Respondent Judgment: 24 February 2010 at 4.00 pmJUDGMENT OF THE COURT A The appeal is dismissed. B The cross-appeal is allowed. C The appellant must pay the respondent costs for a standard appeal on a Band A basis plus usual disbursements. We certify for second counsel.____________________________________________________________________REASONS OF THE COURT(Given by Glazebrook J)Table of ContentsPara NoIntroduction [1]Background [6]N V Sumatra's position [6]LUCKY STRIKE trade marks [10]LUCKIES trade mark [11]The appeal: s 17 [13]The law [13]N V Sumatra's contentions [19]The LUCKIES mark [21]Look and sound [24]Conceptual similarity [26]Whole of the marks [30]Significance of the SFE Act [34]Co-existence in other countries and other classes [40]Luxury items [43]Reasonable possibility or reasonable likelihood [45]Comparison with other cases [52]The cross-appeal: s 16 [66]The law [66]Clifford J's decision [69]BAT's submissions [71]N V Sumatra's submissions [75]Our assessment [77]Result and costs [83]Introduction[1] British American Tobacco (Brands) Incorporated (BAT) owns the word mark LUCKY STRIKE, which has been a registered mark in New Zealand since 1913. In addition, BAT is the registered proprietor of a number of logo marks incorporating the words LUCKY STRIKE and of the word mark LUCKIES. [2] N V Sumatra Tobacco Trading Company (N V Sumatra) has applied for the registration of the word marks LUCKY DRAW and LUCKY DREAM in respect of cigarettes and other cigarette related products, in Class 34. The effective date of N V Sumatra's application was 21 July 2003. This means that the Trade Marks Act 1953 (TMA) is the applicable legislation.[3] BAT opposed N V Sumatra's application but it was granted by the Assistant Commissioner of Trade Marks.1 BAT appealed to the High Court against that decision and it was overturned by Clifford J in a judgment of 11 November 2008. 2[4] Clifford J held that N V Sumatra had failed to discharge its onus of establishing that there was not a reasonable likelihood that the fair notional use of the marks LUCKY DREAM and LUCKY DRAW could cause deception or confusion when compared with the notional fair use of the mark LUCKY STRIKE. He therefore considered that BAT had made out its opposition under s 17(1) of the TMA. Even with the further evidence that had been called in the High Court, Clifford J held, however, that BAT's objection under s 16(1) failed because he considered that BAT had not established the required "necessary substantial reputation" in New Zealand. [5] N V Sumatra appeals against Clifford J's decision on s 17(1) and BAT cross- appeals against Clifford J's finding in respect of s 16(1). We first set out the background in more detail before moving to the appeal and then the cross-appeal.BackgroundN V Sumatra's position[6] N V Sumatra is the proprietor of registrations for the trade marks LUCKY DREAM and LUCKY DRAW for its tobacco products in a number of countries throughout the world, including Belize, Burundi, Cambodia, Hong Kong, Indonesia, Japan, Lesotho, Liberia, OAPI, Nepal and Puerto Rico. In each of those countries N V Sumatra's trade marks co-exist with BAT's trade marks. The registrations are currently being opposed in Australia. Another Indonesian company associated with N V Sumatra has registrations in New Zealand for LUCKY DREAM and LUCKY DRAW in respect of food and beverage products in classes 29, 30 and 32.1 Intellectual Property Office of New Zealand T41/2007, 28 November 2007.2 British American Tobacco (Brands) Inc v NV Sumatra Tobacco Trading Company HC Wellington CIV 2007-485-2814, 11 November 2008.LUCKY STRIKE trade marks[7] In the Interbrand publication of 1996 edited by Nicholas Kochan, The World's Greatest Brands, 3 LUCKY STRIKE is listed as the third most famous tobacco brand in the world. The brand has been (and continues to be) extensively advertised in the United States and worldwide where local legislation allows. [8] Evidence of the annual sales of LUCKY STRIKE cigarettes in New Zealand dated back to 1969 when 20 million LUCKY STRIKE cigarettes were shown as having been sold that year. The sales remained at around the 20 million mark until 1973. In the period 1973 to 1980 they dropped to about five million. From 1980 to 1996 they dropped to a low of 236,000 in 1991 before increasing again. The evidence of sales from 1997 through to 2003 shows cigarette sales of LUCKY STRIKE in New Zealand from eight million in 1997 to over three million in 2003. It is also significant that the sales of LUCKY STRIKE cigarettes were nationwide. [9] The LUCKY STRIKE trade mark is used in sponsorship of motorbike racing and also for the BAT racing team in the Formula One Grand Prix competition. Evidence showed that approximately 100,000 to 300,000 people in New Zealand viewed the Australian Grand Prix from 1999 to 2003 (when N V Sumatra filed its application) which featured BAT's STRIKE car. [10] BAT has used the trade mark LUCKY STRIKE in relation to its tobacco products since the 1850s and in relation to cigarettes since 1916. Its registered trade marks incorporating LUCKY STRIKE have been registered in New Zealand progressively since 1913. As at the date of application for registration of the word mark LUCKY STRIKE (15 March 1913), it was being used by the then applicant.LUCKIES trade mark[11] The affidavit evidence filed on behalf of BAT confirmed that LUCKY STRIKE has commonly been abbreviated to LUCKIES by consumers worldwide.3 Nicholas Kochan (ed) The World's Greatest Brands (Palgrave Macmillan, 1996)BAT therefore, in 1964, successfully applied to register the word mark LUCKIES in New Zealand. Two of the registered logo marks at issue in this appeal also include on the label the word LUCKIES as well as LUCKY STRIKE. [12] At the hearing before the Assistant Commissioner, BAT sought to rely upon its registration of the word mark LUCKIES. The Assistant Commissioner refused permission to rely upon that additional registration in relation to s 17(1) of the TMA on the basis that it did not form part of the case pleaded and amendment at that late stage would be unduly prejudicial to N V Sumatra. The Assistant Commissioner did, however, allow evidence of use of LUCKIES to be relied upon in relation to the s 16 ground of opposition.The appeal: s 17The law[13] Section 17(1) of the TMA provides as follows:Prohibition of identical and similar trade marks(1) Subject to subsection (5) of this section, no trade mark shall be registered in respect of any goods if it is identical with or similar to a trade mark belonging to a different proprietor and already on the register in respect of – (a) the same goods; or (b) similar goods; or (c) services that are similar to such goods – if use of the first-mentioned trade mark is likely to deceive or cause confusion.[14] Both parties agree that the onus was on N V Sumatra to satisfy the Court that there will not be a reasonable likelihood of deception or confusion amongst a substantial number of persons if N V Sumatra uses the proposed trade marks LUCKY DRAW and LUCKY DREAM normally and fairly in respect of any goods covered by the proposed registration, assuming the notional use by BAT of its registered trade marks for LUCKY STRIKE in a normal and fair manner for any ofthe goods covered by the registration of these marks: see the test in Smith Hayden & Co Ltd's Application4 as applied by the Privy Council in Hannaford & Burton Limited v Polaroid Corporation. 5[15] Mr Marriott, for N V Sumatra, submits that there are three elements to the enquiry under s 17(1): (a) Are the marks that are applied for by the applicant similar to the trade mark registrations of the opponent? (b) Are the applicant's marks to be registered in respect of the same or similar goods or services as the goods covered by the opponent's trade mark registrations? (c) Is use of the applicant's mark likely to deceive or confuse? [16] In this case, N V Sumatra accepts that the second question in relation to the similarity of goods must be answered in the affirmative. Mr Marriott submits, however, that the question of degree of similarity and the likelihood of deception and confusion must be considered separately, albeit it is recognised that the questions are inter-dependent in the sense that it must be determined whether the degree of similarity is such as to give rise to the reasonable likelihood of deception or confusion. In Mr Marriott's submission, it is only if both questions are answered affirmatively that registration can be refused. [17] Mr Brown QC, for BAT, submits that it is necessary for the Court to conduct a global assessment, which would cover the distinctive nature of the already registered trade marks, the similarity of the goods and the rules of comparison contained in Re Pianotist Co's Application.6 Pianotist was cited with approval by this Court in Jamieson & Co Limited v J and J Abel Limited7 and also in Polaroid Corporation v Hammerford & Burton Limited.8 We note that the Privy Council in4 Smith Hayden & Co Ltd's Application (1946) 63 RPC 97 at 101.5 Hannaford & Burton Limited v Polaroid Corporation [1976] 2 NZLR 14 (PC) at 18.6 Re Pianotist Co's Application (1906) 23 RPC 744 at 777.7 Jamieson & Co Limited v J and J Abel Limited [1926] NZLR 556 (CA) at 584.8 Polaroid Corporation v Hammerford & Burton Limited [1975] 1 NZLR 566 (CA) at 570.Polaroid, while not citing it, analysed the issue consistently with Pianotist.9Pianotist requires an analysis of the likelihood of confusion or deception amongst a substantial number of persons, the circumstances of purchase, the reason the applicant has chosen these marks and a comparison with other cases. [18] In our view the differences in approach between the parties really amount to semantics. There will be an overall global assessment of the inter-dependent questions set out above at [14]. However, it is likely that the questions (as well as the other factors referred to by Mr Brown insofar as they are relevant) would usually be considered and answered separately in the course of the making of the global assessment.N V Sumatra's contentions[19] On behalf of N V Sumatra, Mr Marriott submits that the High Court made a number of errors in finding that registration of N V Sumatra's LUCKY DREAM and LUCKY DRAW marks should be refused under s 17(1) of the TMA. In his submission, Clifford J erred: (a) When he took into account BAT's trade mark LUCKIES in the s 17 analysis; (b) In giving inadequate weight to the fact that the marks look and sound different; (c) In overstating their conceptual similarity; (d) By wrongly concentrating on the first word as against the whole of the marks; (e) In failing to perceive the full significance of the Smoke Free Environments Act 1990 (SFE Act);9 At 19.(f) In failing to take proper account of the marks' co-existence in other countries and in different classes; (g) In failing to treat cigarettes as luxury items; (h) In applying the lesser test of reasonable possibility rather than reasonable likelihood; and (i) In failing properly to apply relevant case law. [20] We examine each of these contentions in turn.The LUCKIES mark[21] In the High Court Clifford J held that, if the brand LUCKIES was relevant for the purpose of s 16(1) (discussed on the cross-appeal from [66] below), it was difficult to conclude that any prejudice would arise if it was considered similarly in the context of the analysis under s 17(1). [22] Mr Marriott submits that the rationale for this conclusion is flawed and Clifford J was wrong to have regard to the un-pleaded trade mark. In terms of s 17(1), Mr Marriott submitted that N V Sumatra only had to prepare a case and argument to meet the opposition as pleaded. It cannot be obliged to conduct its own search of the trade marks register to determine whether there are any other potentially conflicting trade marks. He also submits that it cannot be assumed that LUCKIES is a form of notional fair use of the trade mark LUCKY STRIKE. [23] We accept Mr Brown's submission that Clifford J was relying on the use of the abbreviation, rather than on the word mark LUCKIES itself. There was an evidential foundation for the Judge's reliance on the use of "luckies" as an abbreviation (see at [11] above). It was legitimate for Clifford J to place reliance on that factor in his s 17(1) analysis.Look and sound[24] Mr Marriott submits that, when comparing the respective trade marks, the High Court erred by giving no or insufficient weight to the starkly different and distinctive elements of the marks being compared, including its own finding that the marks were dissimilar when judged by reference to their look and sound. [25] We do not accept this submission. In fact, Clifford J acknowleged that by their look and sound alone there was some possibility of deception or confusion between the marks: at [71]. We consider that the Judge probably understated the similarity. The use of the word "lucky" followed by a single syllable word in our view creates a reasonably significant risk of confusion even though neither "draw" nor "dream" sounds like "strike".Conceptual similarity[26] The next submission on behalf of N V Sumatra was that the Court gave undue weight to the perceived conceptual similarity between the respective marks; conceptual similarity is a secondary factor only. Mr Marriott submits further that a finding of conceptual similarity was unsupported by the evidence in any event. [27] We consider first Mr Marriott's contention that there was no evidence as to the ideas that consumers actually associated with any of the marks. As pointed out by Mr Brown, this is irrelevant. The decision is one for the judge, taking into account any relevant evidence. Turner J in New Zealand Breweries Limited v Heineken's Bier Browerij Maatschappij NV stated: 10It is for the Court to decide the question of fact as a matter of impression, having due regard, of course, to any relevant evidence which has been produced. But the question is not one to be decided on the opinion of the witnesses. It is a matter for the Judge. He looks at the exhibits before him, and, while he must pay due regard to any relevant evidence produced, the matter remains one of personal impression, visual or phonetic.10 New Zealand Breweries Limited v Heineken's Bier Browerij Maatschappij NV [1964] NZLR 115 (CA) at 139.[28] We agree with Clifford J's view that there is a conceptual similarity in the idea behind the respective marks. The Shorter Oxford Dictionary defines "Strike Lucky" as meaning "have a lucky success". This has the same idea as LUCKY DRAW. Further, Clifford J was correct in our view to hold that LUCKY DREAM also carries a connotation of a lucky outcome. [29] The idea of a mark was considered by the Privy Council in Polaroid and by this Court in Stichting Lodestar v Austin Nichols11 to be clearly a legitimate consideration. The weight to be accorded to it will of course depend on the circumstances. We do not consider that the Judge gave too much weight to conceptual similarity in this case, given his finding of some possibility of confusion or deception from look and sound alone.Whole of the marks[30] Mr Marriott's next submission is that the High Court erred in concentrating on the first word of the mark and thus in failing to limit its comparison to the whole of the respective marks as they appear on the trade marks register. In his submission, that was tantamount to recognising that BAT had a monopoly on the word lucky and BAT's registration did not entitle it to such a broad monopoly. [31] We do not accept that submission. As Mr Brown submits, when making a s 17 comparison, while the focus is on the totality of the mark applied for, the Court can properly look at a common or essential feature in both marks as being key. InHeineken, North P stated: 12But while it is true that it is necessary to have regard to 'the totality' of the proposed trade mark, yet often enough the real risk of confusion may lie in some common feature in the two labels which is liable to linger in the minds of persons requiring a particular kind of goods. Sometimes indeed it will be a word appearing in the labels which either looks or sounds the same.[32] We also accept Mr Brown's submission that the first syllable of a mark or, as here, the first word of a composite mark, ie LUCKY, is usually the most important11 Stichting Lodestar v Austin Nichols [2007] NZCA 61.12 At 134.for comparison: see London Lubricants (1920) Limited's Application where Sargent LJ observed that:13The tendency of persons using the English language to slur the termination of words also has the effect necessarily that the beginning of words is accentuated in comparison and, in my judgment, the first syllable of a word is, as a rule, far more important for the purpose of distinction.[33] In this case, we accept Mr Brown's submission that, applying the tests of look and sound and imperfect recollection, consumers are likely to remember the essential feature LUCKY in the respective marks. The fact of consumer abbreviation of the mark to "luckies" reinforces this conclusion.Significance of the SFE Act[34] Mr Marriott, for N V Sumatra, submits further that, when determining the likelihood of deception or confusion, the High Court erred in failing to recognise that the effect of the SFE Act upon the circumstances of sale of cigarettes and tobacco products would be to reduce the likelihood of deception or confusion to an insignificant level. Mr Marriott points to the fact that the likelihood of confusion is reduced by the requirement to select and ask for a product by brand name, and further by the fact that any confusion can be corrected by the necessary involvement of a shop assistant in the process, or if it is the shop assistant that was confused, corrected by the customer themselves. The High Court appears (in Mr Marriott's submission wrongly) to have found that these factors actually increased the likelihood of deception or confusion. [35] We agree with Clifford J's assessment that the SFE Act increases rather than reduces the likelihood of deception or confusion in this case. This is because of the other actor who must be involved in a purchase, and who is increasingly likely to be a non-smoker unfamiliar with cigarette brands. The Heineken case answers the very point made by Mr Marriott. The fact that any deception might be corrected at the point of sale is irrelevant. Turner J stated: 1413 London Lubricants (1920) Limited's Application (1925) 42 RPC 264 at 279.14 At 141.It was strongly argued that with a quality beer the consumer, if supplied with the wrong article, would protest and require it to be changed. The evidence of John Herbert Philpott, a club steward, was relied on by the appellant in this regard. But even if this argument is accepted, and there remains little likelihood of ultimate deception, at least on any substantial scale, yet it cannot be said that confusion is not likely to result. The argument that any mistake is likely to be remedied, though it may go to deception, cannot be applied to confusion.[36] As Mr Brown submits, it is important to emphasise the scope of what is meant by confusion or deception. These terms are not confined to product-for- product confusion, ie where one product is bought for another. As Richardson J stated in Pioneer Hi-Bred Corn Company v Hy-Line Chicks Pty Ltd: 15"Deceived" implies the creation of an incorrect belief or mental impression and causing "confusion" may go no further than perplexing or mixing up the minds of the purchasing public (New Zealand Breweries Limited v Heineken[1964] 1 NZLR 115, 141). Where the deception or confusion alleged is as to the source of the goods, deceived is equivalent to being misled into thinking that the goods bearing the applicant's mark come from some other source and confused to being caused to wonder whether that might not be the case.[37] We also accept Mr Brown's submission that there are at least two separate situations where confusion or deception could arise. The first is where the consumer asks for "a pack of Luckies" meaning LUCKY STRIKE and using the abbreviation which forms part of two of the registered logo marks. In such a situation it will not be clear what brand is being requested and the stockist may mistakenly supply LUCKY DRAW or LUCKY DREAM cigarettes. [38] The second situation is where the product is stored some distance from the place of ordering because of the restrictions in the SFE Act. The consumer asks for "a pack of LUCKY STRIKE". The sales assistant may be rushed or distracted, or the request is misunderstood because of background noise. He or she may bring LUCKY DRAW or LUCKY DREAM instead so that the consumer receives the wrong product. It may even be that the item is placed in a shopping trolley and the consumer does not immediately notice. [39] Mr Brown offered a third situation, of post-sale confusion. This arises where smokers encounter LUCKY DRAW or LUCKY DREAM outside the sale situation.15 Pioneer Hi-Bred Corn Company v Hy-Line Chicks Pty Ltd [1978] 2 NZLR 50 (CA) at 61- 62.For example, smoker A may be smoking LUCKY DRAW or LUCKY DREAM cigarettes with other smokers in a social location. Smoker A may mention LUCKY DRAW or LUCKY DREAM or even offer smoker B a LUCKY DRAW or LUCKY DREAM cigarette as a gesture. Smoker B may already be smoking LUCKY STRIKE, have previously smoked that brand, or know of it. Smoker B may then be motivated to buy LUCKY DRAW or LUCKY DREAM because he or she thinks that it is connected with LUCKY STRIKE. For commentary on the relevance of post- sale confusion see Levi-Strauss v Kimbyr Investments Ind. 16 We find it unnecessary to consider this situation, which is more arguable.Co-existence in other countries and other classes[40] Mr Marriott also relies on the fact that N V Sumatra's marks have been registered and co-exist with one or more of BAT's marks in other countries and that there are Intellectual Property Office of New Zealand (IPONZ) precedents for co- existence of "LUCKY" marks in other classes. [41] We do not find it of assistance that the marks co-exist in other countries. It is to be noted that (apart from Australia where the registration is being opposed) people in the countries where the marks co-exist do not generally speak English as a first language. We have in any event no evidence of the use of the respective marks in those countries and any confusion that may or may not exist. [42] In relation to N V Sumatra's attempt to put forward the co-existence of other LUCKY brands in other classes, as Mr Brown points out, the IPONZ Practice Manual itself states that each case must be considered on its own merits, citing Robin Jacob QC (now Jacob LJ) in Country Classic Trade Marks: 17I agree that consistency is desirable and that is a sensible practice (particularly in the Registry) to consider a range of similar marks. But in the end it is the mark in question which must be the subject of the decision and I do not think it is incumbent on the Registry when faced with a pile of allegedly similar marks to go into the details of each, drawing fine distinctions here and there. For one thing the hearing officer will not know the detailed circumstances and arguments leading to those registrations.16 Levi-Strauss v Kimbyr Investments Ind [1994] 1 NZLR 332 (HC) at 371.17 Country Classic Trade Marks [1993] RPC 524 at 531.Inevitably there are borderline decisions which might have gone the other way and all one can get is a general impression.Luxury items[43] Next it is argued by Mr Marriott that Clifford J erred in finding that the purchasing decision for the types of products in question would not be a particularly considered one. He says that the evidence of N V Sumatra was that the LUCKY STRIKE cigarettes are a premium brand product. In Mr Marriott's submission, the Court is able to take judicial notice of the fact that cigarettes are no longer sold for insignificant sums of money, and have not been for some considerable time due largely to increased taxes on such products. Furthermore, judicial notice should be taken of the brand loyalty of smokers. In Mr Marriott's submission these factors make the purchasing decision more, rather than less, considered, as does the habitual nature of the purchase. The more familiar a purchaser with the product they are purchasing, the less likely it is that confusion or deception will arise. [44] While we accept that smokers are brand conscious and that this is likely to continue because of the SME Act, we do not accept the characterisation of cigarettes as luxury items. Clifford J was, in our view, merely referring to the fact that, for smokers, cigarettes are an everyday item (or at least one frequently purchased) rather than an occasional, high-priced luxury item. This is a relevant factor.Reasonable possibility or reasonable likelihood[45] Mr Marriott's next argument is that the High Court incorrectly refused registration on the basis of the lower threshold of a reasonable possibility that use of N V Sumatra's trade marks could cause deception or confusion, when s 17(1) of the TMA requires that there be a likelihood of deception or confusion among a substantial number of persons within the relevant target market. [46] Mr Brown, on the other hand, submits that a finding that there is a possibility of deception or confusion means that an applicant has not satisfied the onus thatthere is not a reasonable likelihood of confusion or deception: see for example the way the test is couched in Lever Bros v Newton and Sons.18 Williams J said that: 19The true principle is stated by Lord Watson in Eno v Dunn [(1890) LR 15 App Cas 252 at 257 (HL)], that the applicant must justify the registration of his trade-mark by showing affirmatively that it is not calculated to deceive, and that if there is any doubt whether it is so calculated or not his application ought to be disallowed. [47] Denniston J said:20The [applicants] have not, in my opinion, discharged the onus upon them to, in the words of Lord Herschell [in Eno v Dunn at 261], "to satisfy the Court that there would be no reasonable danger of the public being deceived."[48] Finally, Cooper J held that: 21The burden of showing that the trade-mark which the [applicants] have applied to register is not calculated to deceive the public rests on the [applicants]. They must show this affirmatively, and, to use the words of Lord Watson in Eno v Dunn, it is a necessary consequence that, in dubio, the application to register the trade-mark ought to be disallowed.[49] Clifford J's comments are also, in Mr Brown's submission, entirely consistent with Haslam J's concluding observations in this Court in Heineken: 22The appellant has not only not satisfied me that deception or confusion is unlikely, but I think such a result would be a possible consequence if the Steinecker trade mark were accepted for registration.[50] We do not need to resolve this question for the purposes of this appeal as, whatever the standard, we do not consider it was met by N V Sumatra. Though on one view, the word likelihood must be given meaning, the difference is likely again to be one of semantics once the onus is taken into account. The High Court of Australia has said in Southern Cross Refrigerating Co v Toowomba Foundry Pty Ltdthat: 2318 Lever Bros v Newton and Sons (1906) 26 NZLR 856 (CA).19 At 868.20 At 871.21 At 871.22 At 145 (emphasis added).23 Southern Cross Refrigerating Co v Toowomba Foundry Pty Ltd (1954) 91 CLR 952 at 608 (emphasis added).It is, of course, for the person applying for registration to establish that there is no likelihood of confusion[51] The possibility of deception or confusion among a substantial number of persons may well usually mean that an applicant has not discharged the onus of showing that there is "no likelihood of confusion".Comparison with other cases[52] Mr Marriott submits that the High Court failed to recognise that this case is very similar to Stichting Lodestar and also to the High Court of Australia case ofCooper Engineering Co Pty Ltd v Sigmund Pumps Ltd.24[53] Stichting Lodestar was concerned with a comparison between the WILD TURKEY and WILD GEESE trade marks for spirits. In Stichting Lodestar this Court concluded that the look and sound of the two marks were very different. It said: 25We agree with the Assistant Commissioner that the concept of the word "wild" with a large game bird is not the controlling consideration in this case. We give much greater weight to the stark difference between the words "Turkey" and "Geese", both in spelling and phonetically. The latter is plural, which also differentiates it from the former. The use of these words with the word "Wild" does not detract from those differences. The "concept" is not a strong one – and the class of "hunted birds" is both broad and ill-defined.[54] The Court went on to say that the concepts were in any event different because WILD TURKEY was evocative of the United States whereas WILD GEESE had Irish associations. In Mr Marriott's submission the degree of conceptual similarity was significantly greater than in the present case. [55] Cooper Engineering concerned the marks RAIN KING and RAINMASTER, both of which applied to water spraying installations for agricultural purposes. Mr Marriott submits that the same point regarding the balancing of differences in look and sound against similarities in idea that influenced this Court in Stichting24 Cooper Engineering Co Pty Ltd v Sigmund Pumps Ltd (1952) 86 CLR 536.25 At [31].Lodestar was made by the High Court of Australia in Cooper Engineering. The High Court said: 26In the present case, the prefix of the two words is the same word "Rain", but the suffix "master" differs from "King" in appearance and in sound. This makes the two marks as a whole quite distinct and the marks must be judged as a whole. "Rainmaster" does not look like "Rain King" and it does not sound like it. There is not a single common letter in "master" and in "King". The two words are so unalike to the eye and to the ear that counsel for the appellant was forced to rely upon the likelihood of deception arising from the words conveying the same idea of the superiority or supremacy of the article as a mechanism for making a spray similar to falling rain or artificial rain as it was called during the argument. But it is obvious that trade marks, especially word marks, could be quite unalike and yet convey the same idea of the superiority of some particular suitability of an article for the work it was intended to do. To refuse an application for registration on this ground would be to give the proprietor of a registered trade mark a complete monopoly of all words conveying the same idea as his trade mark. The fact that two marks convey the same idea is not sufficient in itself to create a sufficient resemblance between them, although this fact could be taken into account in deciding whether two marks which really looked alike or sounded alike were likely to deceive.[56] In reply, Mr Brown first submits that caution needs to be applied when making comparisons with other cases. As Turner J observed in Heineken: 27What degree of resemblance is likely to deceive or confuse in any particular case is incapable of definition a priori, and the observation of Judges in other quite different cases are usually of little assistance: Johnston v Orr Ewing(1882) 7 App Cas 219, 220 per Lord Watson.[57] In Mr Brown's submission the extent of reliance on other cases will depend on the circumstances. The fancier the trade mark, the less likely it is that similar trade marks will be allowed. By contrast the more functional a trade mark, the more likely it is that registration of similar marks will be tolerated. Mr Brown submits that, in the case of registered trade marks, there is a continuum between highly distinctive marks at one end and, at the other end, those marks which have descriptive tendencies, which have only achieved registration through providing IPONZ with evidence of acquired distinctiveness through use. This continuum in respect of unregistered trade marks is also referred to by Hill J in Equity Access Pty Limited v Westpac: 2826 At 538.27 At 138 – 139.28 Equity Access Pty Limited v Westpac (1989) 16 IPR 431 (FCA) at 448.The reality is that there is a continuum with at the extremes purely descriptive names at the one end, completely invented names and in between, names that contain ordinary English words that are in some way or other at least partly descriptive.[58] Mr Brown contends that, contrary to N V Sumatra's submissions, Stichting Lodestar is not a very similar case. In particular, there was no evidence in that case that notional fair use of the mark WILD TURKEY included an abbreviation to WILD or WILDS, as is the case with LUCKY STRIKE. Further, the purchase of a bottle of spirits in a self-serve situation at a liquor outlet is likely to be a considered purchase differing from a habitual purchase of cigarettes in a supermarket or convenience store where the packet of cigarettes cannot be self-selected. [59] In addition, Mr Brown submits that the ideas of the marks in Stichting Lodestar were very different, WILD TURKEY being evocative of the region the United States bourbon originated, while WILD GEESE had distinct Irish connotations and was therefore appropriately linked with a brand of Irish whiskey. By contrast, the marks LUCKY STRIKE, LUCKY DRAW and LUCKY DREAM all have the same underlying idea. [60] Mr Brown also submits that this case is quite unlike Cooper Engineering. There the marks in issue were RAIN KING and RAINMASTER in respect of water spraying installations for agricultural purposes. Plainly the word "rain" has a description association with spraying installations unlike the marks in this case. Mr Brown concedes that the Australian High Court was concerned about conferring a monopoly to the trade mark owner in a descriptive word, rain. It is also notable that in that case, the prefix "rain" had been used in other trade marks for goods of the same description sold in the Australian market (eg for RAINWELL, RAINMAKER, RAIN QUEEN and RAINBOX). [61] Mr Brown submits that this case in fact equates to Lever Bros. There the opponent to registration was the owner of four registered trade marks applied to soap namely SUNLIGHT SOAP, SUNLIGHT, SUNBEAM and SUNSHINE. Relying on these registrations, the opponent opposed an application to register the words RISING SUN in connection with soap. Registration was allowed by the Commissioner and by Stout CJ but reversed by this Court. This Court held that,under the relevant statutory provision (which was substantially similar to s 17(1) of the TMA) the principle to be applied where there are existing registrations on the register, was that an applicant must show affirmatively that the mark applied for was not likely to deceive. Denniston J stated, in relation to an identical claim of monopoly as is made here: 29It is said that a decision in the appellants' favour would lead to a monopoly of the word "sun" as a trade mark. If such monopoly of the word, at least as a leading idea, is confined to soap, in respect of which alone the mark is claimed, I see no objection to it. The word has, as I have said, no natural relation to or connection with soap. To adopt the words of Lord Watson inEno v Dunn the argument appears to me to "underrate the resources of the English language".[62] We accept Mr Brown's submission that, with marks that are highly distinctive, the Court will show little tolerance of deception or confusion as to the origin of the goods at issue. Where marks contain elements that are descriptive of goods some level of confusion may be inevitable. The courts will be concerned not to create a monopoly on functional elements. [63] In this case, we accept Mr Brown's submission that the LUCKY STRIKE word mark and logo marks are at the fancy end of the continuum of marks, with no descriptive connotations in respect of tobacco and cigarettes. We also accept Mr Brown's submission on Cooper Engineering. The marks at issue in that case were, by contrast, clearly at the functional end of the spectrum. It is thus not on all fours with this case. [64] We also accept Mr Brown's submission that Stichting Lodestar is not a similar case. In that case, WILD GEESE and WILD TURKEY were applied to different products (very different for the drinkers of the particular spirits involved), the marks had different connotations (linked very much to the origins of the spirits involved) and, apart from the first word, a different sound. It is of significance that the second word of the respective marks in Stichting Lodestar had different syllable lengths and one was plural and one singular.29 At 871.Conclusion[65] None of the challenges to Clifford J's reasoning has been made out. We consider Clifford J was correct to hold that, on a global assessment of all relevant factors, N V Sumatra had not discharged its onus under s 17 of the TMA. The appeal must therefore be dismissed.The cross-appeal: s 16The law[66] Section 16 reads:It shall not be lawful to register as a trade mark or part of a trade mark any scandalous matter or any matter the use of which would be likely to deceive or cause confusion or would be contrary to law or morality otherwise be disentitled to protection in a Court of justice.[67] It has long been accepted that the purpose of s 16 is not to protect competitors but rather to protect the public interest by refusing to accord monopoly rights to a mark, the use of which is likely to deceive or confuse those in the market for goods (or) services. The practical application of s 16 was explained in detail by Richardson J in this Court in Pioneer, 30 in ten propositions, which Richardson J regarded as "clearly settled". [68] It is common ground that the onus is upon the applicant for registration to show that its proposed marks do not fall foul of s 16. However, it was also accepted that there is an initial evidential onus upon an opponent to establish the basis for, or a source of, deception or confusion: Pioneer. 31 The difference between the parties was as to the nature of the obligation with regard to that evidential onus.30 At 61 – 62.31 At 56.Clifford J's decision[69] Clifford J applied, as the test under s 16 of the TMA, that an opponent must establish a "sufficiently substantial reputation in the New Zealand market" to lead to the possibility that the goods covered by the proposed trade mark would be identified with the opponent. [70] On the basis of the evidence provided at first instance and the additional evidence allowed on appeal, Clifford J held that BAT had established that the LUCKY STRIKE brands were known in New Zealand. However, he concluded that BAT had failed to show that its mark had the necessary substantial reputation in New Zealand.BAT's submissions[71] Mr Brown submits that Clifford J was wrong to find that BAT had not met the initial onus of proof of establishing a sufficient awareness or knowledge of its LUCKY STRIKE and LUCKIES trade marks amongst the relevant section of the public (namely smokers) in New Zealand. [72] The evidential onus on an opponent, in Mr Brown's submission, is not to show "substantial reputation" but, as Richardson J stated in Pioneer, "awareness", "cognisance" or "knowledge" of the mark or name relied on by the opponent. Richardson J said:32The reason then for considering reputation and considering it before turning to assess the likelihood of deception or confusion is: In general, consideration of reputation is more helpful as the initial inquiry, for it informs the mind of the circumstances in the trade, against the background of which the two marks must be regarded as notionally in use. (Gaines Animal Foods Ltd's Application (1951) 68 RPC 178, 180). It must be remembered, too, that the object of s 16 is not to protect competitors and potential competitors of the applicant so, the likelihood of damage to the trade of the opponent is not a relevant consideration under32 At 63.s 16 (Hack's Application, pp 106-107). The object is to protect the public interest by refusing to accord monopoly rights to a mark, the use of which is likely to deceive or confuse those in the market for the goods. The concern is with the protection of the relevant New Zealand buying public, not with the protection of the applicant's competitors. Therefore, consideration in narrow terms of questions as to whether the opponent has as established business goodwill in New Zealand, which is important in a passing off action (Alain Bernardin et Compagnie v Pavilion Properties Ltd [1967] RPC 581), is not helpful in an inquiry under s 16. Indeed, by diverting attention from the real issue, it may mislead or at any rate unnecessarily complicate the issues under s 16. And a test framed in terms of the "user" of the opponent's mark in New Zealand has undesirable overtones of proprietorship. For myself I prefer to use a more neutral term such as "awareness" or "cognisance" or "knowledge" and on that basis to ask: having regard to the awareness of the opponent's mark in the New Zealand market for goods covered by the registration proposed, would the use of the applicant's mark be likely to deceive or cause confusion to persons in that market.[73] Mr Brown submits that to require proof of a "substantial reputation" would elevate the evidential onus on an opponent to something akin to proving passing off, a proposition already rejected by this Court and the House of Lords: Pioneer. 33 If the evidential onus were pitched too high, an opponent's case might fail at the threshold even though there was awareness of the opponent's mark or name. As a result the applicant for the trade mark would never be called on to meet the overall onus. This would effectively shift the overall onus of proof that rests on the trade mark applicant to the opponent. [74] Mr Brown further submits that, on the facts, the s 16 ground of objection should be sustained. In his submission, there was evidence of actual and continuous use of LUCKY STRIKE in New Zealand for a substantial number of years and of advertising overseas localised to New Zealand. Although there are differences between ss 16 and 17, BAT submits that the same conclusion that Clifford J reached in respect of the likelihood of confusion or deception under s 17 should logically apply on these facts to the s 16 ground of opposition. This is particularly so, given BAT's continuous use of its brands on cigarettes over decades and the fact that the goods for which N V Sumatra seeks registration in class 34 comprise cigarettes and related products.33 At 62 (emphasis added).N V Sumatra's submissions[75] Mr Marriott submits that the test applied by Clifford J was appropriate and that it does not matter what label is attached to the test, as long as the correct analysis is carried out. It was irrelevant that Clifford J used the words "sufficiently substantial reputation in the New Zealand market" as it was apparent that the Judge was simply asking whether the level of reputation was such that there would be deception or confusion.34 There was nothing wrong with using the word "reputation", and it was clear that the word was not used in the passing off sense. Mr Marriott submits that BAT's arguments failed not because the Judge applied the wrong standard, but rather because BAT failed to provide any evidence as to the size or characteristics of the tobacco market as a whole. The Judge was thus unable determine whether the figures BAT provided as to its own sales translated into reputation or awareness of its marks. [76] Mr Marriott agrees, however, that the exercise of comparing the LUCKY DREAM and LUCKY DRAW marks with the LUCKY STRIKE mark is in this case the same as that carried out in relation to s 17(1).Our assessment[77] We accept Mr Brown's submission that Clifford J set the evidential onus too high. While Richardson J did say Pioneer that an opponent must first establish a "sufficiently substantial reputation", 35 it is clear from the passage of Pioneer quoted at [72] above that this is a relatively low threshold. All that an opponent needs to show is "awareness", cognisance" or "knowledge" of the mark. This means that the opponent will first have to identify the relevant market, then point to evidence showing that a substantial number of persons in that market have awareness, cognisance or knowledge of its mark: Pioneer. 36 What is a substantial number of34 See at [41].35 At 62.36 At 74.persons depends on the nature and size of the market and is relative both to the number of persons involved in and their impact on that market: Pioneer. 37[78] Clifford J required BAT to prove a substantial reputation amongst a substantial number of persons (as against mere awareness or cognisance of the marks among a substantial number of persons). This is not the Pioneer test and is wrong in principle, essentially for the reasons given by Mr Brown (outlined above at [73] above). The focus of s 16 is protection of consumers from being deceived and confused, and it is apparent that deception and confusion might arise even where the reputation of the opponent's mark falls short of being "substantial". [79] Our reasoning on this issue is consistent with the decision of Baragwanath J in Red Bull Gmbh v Meier.38 Insofar as recent High Court cases have approved the test of "quite substantial awareness or knowledge of the relevant mark" in Riviera Leisurewear Pty Ltd v J Hepworth & Son Plc, 39 we accept Mr Brown's submission that this is inconsistent with the New Zealand test as set out in Pioneer.[80] In this case, the evidence provided by BAT clearly met the threshold of showing "awareness", "cognisance" or "knowledge" of its marks. Indeed, it is in fact highly arguable that all BAT needed to show in order to discharge its initial evidential onus was that its marks were registered in New Zealand. Registration, after all, is notice to all of the world. In Pioneer, the question as to reputation or awareness of the opponent's mark had to be addressed because the opponent's mark was not registered in New Zealand. Here, by contrast, BAT's marks have been registered and used in New Zealand for some time. Frankel and McLay comment as follows: 40When analysing whether an application is likely to deceive or confuse [under ss 16 and 17], many factors will be the same whether the likelihood of deception or confusion arises in relation to a registered or unregistered trade mark. The major difference will be that the opponent whose objection is based on an unregistered trade mark will have to prove a reputation in that37 At 74 – 75.38 Red Bull Gmbh v Meier [2002] FSR 42 (HC).39 Riviera Leisurewear Pty Ltd v J Hepworth & Son Plc (1987) 9 IPR 305 at [31].40 Susy Frankel and Geoff McLay Intellectual Property in New Zealand (Butterworths, Wellington 2002) at [8.4.2].trade mark, whereas the opponent relying on a registered mark will not have to prove reputation.[81] Even if it had been necessary for BAT to show substantial reputation in New Zealand, it would in our view have discharged that evidential onus, given the length of time the word mark LUCKY STRIKE has been registered in New Zealand, the evidence of use before registration, the Grand Prix advertising (seen in New Zealand) and the sales figures since 1969. As Mr Brown points out, sales figures do not give the whole story. Others will be aware of LUCKY STRIKE cigarettes by seeing people smoking them and carrying packets bearing BAT's marks. Further, given how well known the brand is offshore and its longevity, it is safe to assume a level of spillover reputation in New Zealand: see the comments of the High Court of Australia in Radio Corporation Ltd v Disney.41[82] In summary, therefore, BAT met its initial evidential onus. It was then for N V Sumatra to prove, in terms of s 16, the absence of a likelihood to deceive or cause confusion. The parties are effectively agreed that, in this case, the s 17 analysis on whether there is a likelihood of deception or confusion would apply to s 16. This means that the cross-appeal must be allowed.Result and costs[83] The appeal is dismissed. [84] The cross-appeal is allowed. [85] The appellant must pay the respondent costs for a standard appeal on a Band A basis plus usual disbursements. We certify for second counsel.Solicitors: James & Wells, Auckland for Appellant Buddle Findlay, Wellington for Respondent41 Radio Corporation Ltd v Disney (1937) 57 CLR 448.