Date: 19991116
Docket: T-2432-97
OTTAWA, ONTARIO, THIS 16TH DAY OF NOVEMBER, 1999.
PRESENT: THE HONOURABLE MR. JUSTICE EVANS
IN THE MATTER of an Appeal pursuant to Section 56 of the Trade-
marks Act R.S.C. 1985, c. T-13 from a Decision of the Registrar
dated September 10, 1997, rejecting the Opposition by Garbo
Creations Inc. to Canadian Trade-mark Application No. 708,840 for
the mark GRETA GARBO & Design filed by Harriet Brown &
Company Inc.
BETWEEN:
GARBO GROUP INC.
(Appellant)
- and -
HARRIET BROWN & COMPANY INC. and
THE REGISTRAR OF TRADE-MARKS
(Respondents)
JUDGMENT
The appeal is dismissed. The parties have 14 days from the date of this decision to make submissions in writing on the award of costs.
"John M. Evans"
J.F.C.C.
Date: 19991116
Docket: T-2432-97
IN THE MATTER of an Appeal pursuant to Section 56 of the Trade-
marks Act R.S.C. 1985, c. T-13 from a Decision of the Registrar
dated September 10, 1997, rejecting the Opposition by Garbo
Creations Inc. to Canadian Trade-mark Application No. 708,840 for
the mark GRETA GARBO & Design filed by Harriet Brown &
Company Inc.
BETWEEN:
GARBO GROUP INC.
(Appellant)
- and -
HARRIET BROWN & COMPANY INC. and
THE REGISTRAR OF TRADE-MARKS
(Respondents)
REASONS FOR JUDGMENT
EVANS J.
[1] Since 1984 Garbo Group Inc., or its predecessor, Garbo Creations Inc., has been selling in retail stores across Canada a range of goods that are described compendiously as "women"s fashion accessories". Between 1984 and 1991 the goods expanded to include the following: jewellery (precious, semi-precious and costume), hair ornaments, button covers, jacket clips, handbags and belts.
[2] Garbo Group is the owner of the registered trade marks GARBO DESIGN, T.G.I.F. BY GARBO and GARBO which it uses in connection with various of its goods. The marks were registered in 1988, 1993 and 1995 respectively. I shall refer to them collectively as the GARBO marks.
[3] In 1992 Harriet Brown & Company Inc. applied to register the trade-mark GRETA GARBO which it proposed to use in Canada in association with cosmetics and toiletries, and eyeglasses and accessories, including sunglasses. Garbo Group opposed this registration on the ground that there was a reasonable likelihood of confusion between the proposed GRETA GARBO mark and the GARBO marks and trade names.
[4] Mr. G.W. Partington, Chair of the Trade-marks Opposition Board, rejected the opposition in a decision rendered in September 1997. He found that confusion was not reasonably likely, largely because of the differences between the goods covered by the proposed and existing trade-marks, and the different channels of trade through which they would normally pass.
[5] Garbo Group has appealed from this decision pursuant to subsection 56(1) of the Trade-marks Act, R.S.C. 1985, c. T-13, alleging that the Board erred in its conclusion that the proposed mark was not reasonably likely to be confused with the appellant"s registered marks. The appellant filed affidavits in support of the appeal that were not before the Board. The respondent did the same.
[6] This case turns on the answer to the following question. When average consumers in the targeted population with imperfect recollection see the name "Greta Garbo" on the applicant"s goods are they reasonably likely to think that they come from the same source as those already being sold by the appellant under the GARBO trade marks?
[7] It is important to bear in mind in the context of this question that the related purposes of trade-mark law are the protection of consumers from being misled about the source of goods, and of the interest of traders in identifying their goods or services to the extent compatible with fair competition.
[8] Although I assume that the decision was made by Mr. Partington in the exercise of powers delegated to him by the Registrar under subsection 63(3), in these reasons I shall refer to the decision under appeal as that of the Registrar.
[9] Gary Grundman is the president of Garbo Group Inc. In the affidavit that he swore in November 1997 for the purpose of this appeal Mr. Grundman stated that in the previous four years the company"s total sales of goods covered by the GARBO marks exceeded $10 million. The best year was 1994 when sales were just under $4 million; thereafter they dropped sharply, but in 1997 recovered to something in excess of $2 million.
[10] He stated also that these goods were sold to 3,000 customers and were found in 600 stores, including such major national department stores at that time as The Bay, Eaton"s, Sears and Zellers, as well as smaller stores specializing in jewellery or women"s fashion and apparel. The company has spent in excess of $1 million on advertising, including advertisements in fashion magazines with a national circulation.
[11] Mr. Grundman has further stated that the word GARBO is a coined word from the letters that start the given names of the principals of the company, Gary and Bonnie Grundman. Whether the word was also intended to suggest to consumers the late Greta Garbo, legendary movie actress of the 1930s and the early 1940s, and thereafter famous and mysterious recluse, is not of immediate relevance here. However, as counsel suggested in argument, the Grundmans" choice of GARBO as the trade-mark for their fashion accessories, rather than some other combination of the initial letters in their names, such as BONGA, may not have been entirely arbitrary.
[12] Harriet Brown & Company is a private company. Its principals are the daughters of a niece of Greta Garbo. Under her aunt"s will the niece was given the right to the name and "persona" of Greta Garbo, which she subsequently assigned to her daughters.
[13] Harriet Brown proposes to use as its mark the words "Greta Garbo" written in script. Garbo Group uses as one of its trade-marks the word GARBO, which is also written in script, but quite different from that proposed to be used by Harriet Brown.
[14] For the purposes of this appeal the following provisions of the Trade-marks Act are particularly relevant.
| 6.(5) In determining whether trade-marks or trade-names are confusing, the court or the Registrar, as the case may be, shall have regard to all the surrounding circumstances including
(a) the inherent distinctiveness of the trade-marks or trade-names and the extent to which they have become known;
(b) the length of time the trade-marks or trade-names have been in use;
(c) the nature of the wares, services or business;
(d) the nature of the trade; and
(e) the degree of resemblance between the trade-marks or trade-names in appearance or sound or in the ideas suggested by them.
|
|
| 6.(5) En décidant si des marques de commerce ou des noms commerciaux créent de la confusion, le tribunal ou le registraire, selon le cas, tient compte de toutes les circonstances de l'espèce, y compris_:
a) le caractère distinctif inhérent des marques de commerce ou noms commerciaux, et la mesure dans laquelle ils sont devenus connus;
b) la période pendant laquelle les marques de commerce ou noms commerciaux ont été en usage;
c) le genre de marchandises, services ou entreprises;
d) la nature du commerce;
e) le degré de ressemblance entre les marques de commerce ou les noms commerciaux dans la présentation ou le son, ou dans les idées qu'ils suggèrent.
|
|
| 56. (1) An appeal lies to the Federal Court from any decision of the Registrar under this Act within two months from the date on which notice of the decision was dispatched by the Registrar or within such further time as the Court may allow, either before or after the expiration of the two months. |
| 56. (1) Appel de toute décision rendue par le registraire, sous le régime de la présente loi, peut être interjeté à la Cour fédérale dans les deux mois qui suivent la date où le registraire a expédié l'avis de la décision ou dans tel délai supplémentaire accordé par le tribunal, soit avant, soit après l'expiration des deux mois. |
(5) On an appeal under subsection (1), evidence in addition to that adduced before the Registrar may be adduced and the Federal Court may exercise any discretion vested in the Registrar.
(5) Lors de l'appel, il peut être apporté une preuve en plus de celle qui a été fournie devant le registraire, et le tribunal peut exercer toute discrétion don"t le registraire est investi.
D. ISSUES AND ANALYSIS
1. Preliminary Matters
(a) standard of review
[15] It is generally said that the standard of review applicable to the Registrar"s decision is that of correctness, although in view of the expertise of the Registrar, and of those to whom the Registrar has delegated decision-making powers under subsection 63(3), their decisions should not lightly be set aside on findings of fact that are central to their specialized jurisdiction, such as whether confusion is likely to be caused by the proposed mark. However, when significant evidence is adduced on the appeal that was not before the Registrar, less weight should be given to the Registrar"s findings: Molson Companies Ltd. v. John Labatt Ltd. (1984), 1 C.P.R. (3d) 494 (F.C.T.D.); United States Polo Association v. Ralph Lauren Corporation (F.C.T.D.; T-1881-93; T-193-95, T-189-96; March 8, 1999).
[16] In my opinion the standard of review applied to the Registrar"s findings of confusion needs to be reformulated to take account of the pragmatic and functional analysis developed in contemporary administrative law jurisprudence for selecting the standard of review appropriate for the issue in dispute. In undertaking this exercise I am following the path already indicated by Lutfy J. in Young Drivers of Canada v. Chan (F.C.T.D.; T-636-97; August, 30, 1999). However, the end result of the analysis may differ relatively little from the Court"s established practice.
[17] The following factors are relevant to a determination of Parliament"s intention on the standard of review to be applied to a finding by the Registrar on the reasonable likelihood of confusion.
[18] First, the language of the constitutive legislation must be examined. Here, it is relevant to note that subsection 56(1) of the Trade-marks Act creates a right of appeal from the Registrar to the Court, and imposes no limitation on its scope. Indeed, subsection 56(5) expressly permits the parties to adduce additional evidence on an appeal that was not before the administrative decision-maker, and authorises the Court to exercise any discretion vested in the Registrar. However, since the likelihood of confusion is a question of fact or mixed fact and law, not discretion, this latter provision is irrelevant to this appeal.
[19] The admissibility of additional evidence on appeal under subsection 56(5) indicates a broad appellate jurisdiction, and a standard of review towards the correctness end of the spectrum. However, it is not as broad as the right of appeal from the human rights tribunal considered in [1982] 1 S.C.R. 202">Ontario Human Rights Commission v. Borough of Etobicoke, [1982] 1 S.C.R. 202, 211; Dickason v. University of Alberta, [1992] 2 S.C.R. 1103, 1126; Zurich Insurance Co. v. Ontario (Human Rights Commission), [1992] 2 S.C.R. 321.
[20] In these cases the Court applied a standard of correctness to the tribunal"s findings of fact and to its assessment of the legal implications of those facts, because the right of appeal was expressly stated to be on questions of fact or law, the appellate court was directed to substitute its opinion for that of the tribunal, and the tribunal was not regarded as having specialized expertise.
[21] A similarly wide right of appeal from a professional discipline committee has also been held to require the appellate court to review the tribunal"s findings of fact on a standard of correctness: see, for example, Richmond v. College of Optometrists of Ontario (1995), 25 O.R. (3d) 448 (Ont. Div.Ct.).
[22] However, it has also been said that professional discipline committees are intended to use their specialized knowledge in assessing evidence of, for example, professional competence, and that consequently, despite the broad right of appeal, an appellate court should give considerable weight to the conclusion of these committees on questions of fact within the area of their expertise: Reddall v. College of Nurses of Ontario (1983), 149 D.L.R. (3d) 60 (Ont. C.A.).
[23] The provision in subsection 56(5) allowing the parties to introduce additional evidence on the appeal may suggest a correctness standard on those findings of fact to which the evidence relates. However, it does not necessarily follow that the same standard should apply to the Registrar"s findings on other facts.
[24] Indeed, even when additional evidence is admitted on appeal, it still may be appropriate to ask whether the Registrar"s decision was wrong in the light of that evidence, rather than how the appellate court would have decided the question if it had been before the court de novo . Of course, the more significant the additional evidence the greater the scope for the court to exercise its own independent judgment on the finding of fact in question.
[25] In [1977] 1 S.C.R. 517">Lamb v. Canadian Reserve Oil & Gas Ltd., [1977] 1 S.C.R. 517, the Court considered the standard of review of findings of fact to be applied by a District Court to which there was a statutory right of appeal in the "form of new hearing de novo" from a specialized administrative tribunal.
[26] Noting that the statutory designation of the appeal as de novo empowered the District Court to admit evidence that was not before the Board, Martland J. nonetheless held (at pages 527-28) that the fact that the first-level decision-maker was a specialized tribunal was relevant to determining the appellate court"s role under the statutory scheme:
| The appeal is from a tribunal which, in dealing with compensation claims throughout the province, acquires an expertise in these matters. The appeal is to a District Court judge, at the judicial centre nearest to the land in question, who is not a specialist in these matters. He hears evidence, and, it may be, fresh evidence. He can assess the credibility of the witnesses, but in determining land values, the adverse effect on other land resulting from the right of entry and compensation for nuisance, he should have some regard for the opinion of the Board. |
I would note that, while the appeal from the Registrar to the Court has often been described as de novo, this is not a term that is found in the Trade-marks Act itself in this context.
[27] Second, a pragmatic and functional analysis requires consideration of the nature of the question in dispute, and the relative expertise of the administrative decision-maker and the reviewing court for deciding it. In determining the likelihood of confusion, the Registrar is directed by subsection 6(5) to have regard to a non-exhaustive list of factors.
[28] Inferring from the primary facts whether confusion is likely is a mixed question of fact and law because it is a finding about whether a statutory criterion has been satisfied and hence calls for some appreciation of the underlying purposes of the statute: Canada (Director of Investigation & Research) v. Southam Inc., [1997] 1 S.C.R. 748.
[29] It also involves making particular findings of fact, and making an informed assessment of the facts, especially when, as will often be the case, some facts point to one conclusion and others to the other. The presence of a significant factual aspect to the issue in dispute, and hence the limited precedential value of individual decisions, suggests a deferential standard of review.
[30] Administrative decision-makers under the Trade-marks Act have regular exposure to the likelihood of confusion between actual and proposed trade marks in a wide range of commercial contexts. Experience of this kind is calculated to make for consistent and
informed decisions. The Registrar and the assistants to whom decision-making power has been delegated also make their decisions on the basis of a hearing of a broadly adjudicative nature. On the other hand, it is also true that courts are not unfamiliar with the closely related issues that arise in passing-off cases and the related causes of action codified and modified in section 7 of the Trade-marks Act.
[31] It is relevant to observe here that the legal powers and functions of the Registrar in determining the opposition to the registration of a trade mark are of an essentially adjudicative nature. They do not involve the discharge of the kind of broad regulatory responsibilities that the Supreme Court in Pezim v. British Columbia (Superintendent of Brokers), [1994] 2 S.C.R. 557 identified as indicative of a deferential standard of review. Or, to put the point somewhat differently, the issues involved in the proceedings before the Opposition Board are bipolar, not polycentric, in nature: Pushpanathan v. Canada (Minister of Citizenship and Immigration), [1998] 1 S.C.R. 982.
[32] Third, the reasons for vesting decision-making power in the Registrar rather than in a court include, presumably, the desirability of having decisions of this kind made by those who encounter the issues on a regular basis and thereby develop an expertise in the area, and of minimising the expenses and delays often associated with litigation. These considerations suggest that on appeal the Court is not intended to review findings of fact within the expertise of the Registrar on a standard of correctness, at least in the absence of additional evidence.
[33] Fourth, the nature of the rights at stake in trade mark opposition proceedings must also be considered as part of the "pragmatic or functional" mix. They are primarily economic in nature: the protection of consumers from being misled, and of producers and distributors of goods from unfair competition. Important as these interests are, they do not rank in our legal system with statutorily protected human rights, or even with the right of a person to pursue a profession or other vocation. Thus, the nature of the rights involved indicates a more deferential standard of review.
[34] To conclude, it is my opinion after weighing these factors that, despite the inclusion in the Trade-marks Act of an untrammelled right of appeal and the right to adduce additional evidence, a considerable degree of deference is called for on the part of the appellate Court when reviewing the Registrar"s finding of confusion, provided at least that no significant new evidence has been adduced on a factual issue and it is not alleged that an error of law has been committed.
[35] Given, in particular, the expertise of the Registrar on the question of confusion, the reasons for vesting decision-making in the Registrar, and the nature of the rights at stake, "unreasonableness simpliciter", which was said in Southam , supra, to be synonymous with "clearly wrong", is in my opinion the most appropriate of the three standards of review currently available.
[36] I am fortified in reaching this conclusion by the similar views expressed by Lutfy J. in Young Drivers, supra, even though on the facts of that case he did not find it necessary to reach a definitive conclusion on the issue. He said:
| If the issue of confusion can properly be characterised as a matter of mixed law and fact, the standard of review may be further down the spectrum from correctness. The special knowledge and expertise of the Registrar may require greater deference, where no new evidence is adduced on appeal. The determination on appeal may be closer to whether the Registrar"s decision is "clearly wrong" or "unreasonable". |
[37] As for the impact on the standard of review of the filing of additional evidence on an issue at the appeal, much will depend on the extent to which the additional evidence has a probative significance that extends beyond the material that was before the Registrar. If it adds nothing of significance, but is merely repetitive of existing evidence without enhancing its cogency, its presence should not affect the standard of review applied by the Court on the appeal.
[38] If, on the other hand, the additional evidence goes beyond what was in substance already before the Registrar, then the Court should ask whether, in the light of that material, the Registrar reached the wrong decision on the issue to which that evidence relates and, perhaps, on the ultimate decision as well. The more substantial the additional evidence, the closer the appellate Court may come to making the finding of fact for itself.
[39] However, when the additional evidence is so significant and extensive that the case is fundamentally different from the one before the Registrar, the Court may conclude that, rather than simply reversing the decision, the more appropriate disposition would be to remit it to the Registrar. Otherwise, the legislative scheme conferring first level decision-making power on the Registrar, a specialist tribunal, may be undermined.
[40] Of course, for the Court to remit the matter to the Registrar may delay the ultimate disposition of a trade mark application that has been opposed. Further, it may be inferred from the right of the parties to adduce additional evidence on the appeal that Parliament thereby intended to enable the Court to make the final decision.
[41] However, given the specialized nature of the Registrar"s jurisdiction, it is also reasonable to impute to Parliament an intention that the parties not be encouraged to treat the administrative hearing in the opposition proceeding as merely a prelude to an appeal to this Court. Hence, in my opinion it is consistent with the statutory scheme for the Court to remit in appropriate cases, rather than itself making a de novo finding on the question of confusion on the basis of very different evidence from that adduced before the statutorily designated specialized tribunal.
(b) burden of proof
[42] It was not disputed that an applicant for the registration of a trade-mark has the burden of proving that there will be no reasonable likelihood of confusion with an existing mark, and that the burden remains on the applicant both before the Registrar and on an appeal to the Court: McDonald"s Corp. v. Coffee Hut Stores Ltd. (1994), 55 C.P.R. (3d) 463 (F.C.T.D.). In other words, any doubt is to be resolved in favour of the opponent, in this case, Garbo Group.
(c) test to be applied
[33] The relevant principles were aptly stated by Rouleau J. in Bally Schuhfabriken AG/Bally"s Shoe Factories Ltd. v. Big Blue Jeans Ltd. (1992), 41 C.P.R. (3d) 205, 209 (F.C.T.D.), where he said:
| The issue of confusion is a question of fact to be determined on the merits of each case. The test is one of the impression having regard to the average consumer with imperfect recollection. [emphasis added] |
[43] The second point to be made here is that, as section 6 of the Act provides, the relevant question is not whether there is a reasonable likelihood of confusion in the abstract between the proposed mark and any existing mark. Rather, it is whether the applicant"s proposed mark is reasonably likely to induce consumers into thinking that the goods to which the mark will be attached emanate from the same source as, or are otherwise associated with, the goods covered by the opponent"s mark.
[44] In this case the applicant contended that the average consumer was likely to think that the goods to be sold under its GRETA GARBO mark were in some way associated with the late Greta Garbo, not with the opponent"s goods sold under the GARBO mark.
[45] Counsel went on to argue that the opponent was estopped from asserting that customers would think that the goods sold under the GARBO mark were associated with the actress, and hence that the applicant"s goods might be thought to be associated with the opponent"s.
[46] This is because, when the opponent first applied to register its GARBO mark in 1987 (three years before the death of Greta Garbo), it was challenged on the ground that, contrary to paragraph 9(1)(k) of the Act, it falsely suggested an association between the goods and the actress. Garbo Group vigorously denied this and the Trade Mark Office found that GARBO did not suggest an association with Greta Garbo.
[47] Counsel for Harriet Brown contended, therefore, that Garbo Group could not now be heard to say that goods sold under the GRETA GARBO mark might be thought to be associated with goods sold under the GARBO mark. The Registrar rejected this argument, on the ground that the test for determining whether a proposed mark falsely suggests an association with a living person is not the same as the test for determining whether a proposed mark will cause confusion with an existing mark. I agree. No estoppel therefore arises and, indeed, the appellant did not argue that the respondent"s mark was confusing with its own because consumers associated the GARBO mark with Greta Garbo.
| 2. The Likelihood of Confusion |
[48] Counsel for Garbo Group submitted that the only point on which the Registrar found in favour of the applicant was the dissimilarity between the goods of the applicant and those to which the GARBO mark attached, and hence the different channels of trade through which they would normally pass. Thus, counsel argued, if the Registrar erred on this point, I must allow the appeal.
[49] Seductive as this submission is, I cannot accept it. Whether there is a reasonable likelihood of confusion depends on a consideration of the facts in the light of all the specific factors set out in subsection 6(5), not all of which carry the same weight, including "all the surrounding circumstances". The Court may still conclude that the Registrar"s decision was not wrong, even though, in the light of additional evidence on one issue, the Registrar"s finding on that issue was erroneous.
[50] The specific factors that must be considered are: the inherent distinctiveness of the trade-marks and the extent to which they have become known; the length of time that the trade-marks have been in use; the nature of the wares; the nature of the trade; and the degree of resemblance between the trade-marks in appearance, sound or in the ideas suggested by them. I examine below the decision of the Registrar and the evidence submitted by the parties on each of these points, particularly the evidence that was not before the Registrar.
(a) Distinctiveness of the marks
(i) Registrar"s decision
[51] The Registrar found GRETA GARBO to be a weak mark because it is the name of a person, who is now dead. Similarly, the Registrar found that the GARBO marks possessed only a "limited degree" of inherent distinctiveness given "the surname significance associated with the word Garbo", even though "Garbo is a relatively rare surname and may therefore not necessarily be perceived by the average consumer as possessing a surname significance."
[52] The Registrar also found that the applicant"s mark had not acquired distinctiveness either, because it had not become known at all in Canada in connection with the wares for which it was claimed. In this regard, he stated:
| I do not consider that the fame associated with the former actress Greta Garbo translates into a certain measure of reputation with the trade-mark GRETA GARBO Design which the applicant intends to use in Canada in association with certain wares. |
[53] He thus found that, for the purpose of determining the likelihood of confusion, the criterion of distinctiveness weighed in the opponent"s favour, "although only to a limited extent."
(ii) additional evidence
[54] In the affidavit that he swore for this appeal Gary Grundman stated that over the past twelve years his company had spent more than $1 million in advertising its products. The amounts varied between $40,000 and $150,000 per year. However, there was no direct evidence that average consumers recognized the mark GARBO, or that they associated it with the appellant"s products.
[55] The respondent relied on affidavit evidence to demonstrate that a number of businesses operate across Canada with the word "Garbo" in their name, including hair salons, and women"s clothing and accessories stores. The respondent also relied on affidavit evidence of a search of Canadian telephone directories that showed thirteen individuals who had this surname.
[56] The respondent also adduced evidence of the continued fame of Greta Garbo, particularly with its target consumers, in order to support the distinctiveness of the name and its claim that consumers would associate the products with the actress and not with the appellant"s goods.
(iii) analysis
[57] The Registrar"s conclusion on the lack of inherent distinctiveness of the respondent"s proposed mark was not challenged on appeal. As for the additional evidence of Greta Garbo"s continuing fame in Canada, the issue is whether the respondent"s mark has acquired distinctiveness in connection with the wares for which it is claimed. The Registrar found on the evidence before him that it had not, and the additional evidence does not demonstrate that his conclusion was wrong.
[58] Similarly, the Registrar found that the appellant"s marks, GARBO and GARBO DESIGN, have little inherent distinctiveness since the word "Garbo" has "surname significance". Even if, as the appellant"s evidence suggested, the fame of Greta Garbo is waning, her residual fame and the thirteen entries in the telephone directory with the name "Garbo" are enough to give surname significance to the marks GARBO and GARBO DESIGN: compare Cartier Men"s Shops Ltd. v. Cartier Inc. (1981), 58 C.P.R. (2d) 68 (F.C.T.D.).
[59] Further, the use of "Garbo" in the names of hair salons, and clothing and accessories stores also undermines the distinctiveness of the name by suggesting a more general association with an attractive personal appearance. On the other hand, the Registrar also held that the mark T.G.I.F. BY GARBO possessed "a fair degree" of inherent distinctiveness when considered in its entirety.
[60] The appellant introduced as evidence of acquired distinctiveness a statement of the expenses that it had incurred for advertising and promotion over the period between 1984 and 1998. However, there was no direct evidence that consumers recognized the mark GARBO and associated it with the wares sold by the appellant. Indeed, what evidence of this kind that there was indicated that the marks had little acquired distinctiveness.
[61] Thus, in an affidavit submitted on behalf of the respondent, Joan Harting Barham, the editor of Toronto Life Fashion Magazine, stated that she had heard of the Garbo Group, but she could not recall when or where. In her evidence she said that she associated the brand name with, if anything, wooden bangles.
[62] There is thus no reason for concluding that the Registrar"s finding on the issue of distinctiveness was wrong, let alone unreasonable. In short, therefore, the respondent"s marks have little or no inherent or acquired distinctiveness. Since the appellant"s marks GARBO and GARBO DESIGN have some degree of inherent and acquired distinctiveness, and the mark T.G.I.F. BY GARBO has rather more, this factor favours the appellant, but only to a limited extent.
(b) Duration of use of the mark
[63] The Registrar found that this factor favoured the appellant, as the GARBO marks had been used in association with the appellant"s goods since 1984, while the mark GRETA GARBO had not yet been used in Canada. No additional evidence was put forward on this issue.
[64] It was held in Cochrane-Dunlop Hardware Ltd. v. Capital Diversified Industries Ltd. (1976), 30 C.P.R. (2d) 176, 185 (Ont. C.A.) that a presumption that goodwill attached to a mark applied when it had been promoted by its owners and had been on the register for twelve years. The GARBO marks have been in use for a similar period of time and have been promoted by their owner through non-trivial, but not huge, expenditures on advertising and other forms of promotion over that period. The Registrar was thus clearly correct to conclude that the second factor favoured the appellant.
(c) Nature of the wares
[65] The Registrar found that "the applicant"s cosmetics, toiletries, eyeglasses and accessories differ from the opponent"s jewellery" and that the goods registered under the mark T.G.I.F. BY GARBO differed from the respondent"s goods. No new evidence was led on this point on the appeal.
[66] There is no basis on which to interfere with the Registrar"s conclusion on this issue, as far as it goes.
(d) Nature of the trade
(i) Registrar"s decision
[67] On this issue, the Registrar simply concluded that he "would expect that the normal trade channels associated with jewellery on the one hand and those associated with cosmetics, toiletries, eyeglasses and accessories on the other would be quite distinct."
(ii) additional evidence
[68] While no evidence appears to have been before the Registrar on this issue, the extent to which the goods passed through the same channels of trade was the subject of considerable additional evidence on the appeal.
[69] Thus, the appellant submitted an affidavit from Gordon Hayward, a manager at the Hudson"s Bay Co., who stated that in Bay department stores, jewellery and cosmetics are often placed in close physical proximity to each other. This is both because they are high profit items and are therefore displayed prominently, and because they appeal to the same target market.
[70] An appendix to Mr. Hayward"s affidavit contained a floor plan of the Bay store in Victoria which showed accessories, jewellery and cosmetics displayed close together. Photographs of Bay stores in Toronto and Laval showing Garbo jewellery displayed near cosmetics counters were included in the affidavit of Gary Grundman. However, as Kevin O"Neil pointed out in an affidavit filed on behalf of the respondent, while cosmetics and jewellery are sold on the same floor in The Bay, so are many others. Further, eyeglasses are sold on a different floor.
[71] The appellant also relied on evidence from Ms. Barham, who stated in cross-examination on her affidavit that many of the goods sold by the parties to this appeal could be characterized as women"s fashion accessories, with the exception of make-up. During the cross-examination of Ms. Barham the appellant also produced as an exhibit an advertisement from Toronto Life Fashion Magazine , which showed products such as skin cream and jewellery being advertised together.
[72] On the other hand, the respondent also led evidence on the issue of the channels of trade. Thus, Mr. O"Neil deposed that he had visited a number of stores in the Bloor-Yorkville area of Toronto where cosmetics, toiletries or eyeglasses were sold, but not jewellery, and vice versa.
[73] Further, counsel for the respondent noted that, while Ms. Barham suggested that a number of the items in both parties" trade mark applications could be considered part of the "fashion accessory" industry, she later stated that the term "fashion accessory" is extremely broad. Indeed, she stated that it is broad enough to include even automobiles.
(iii) analysis
[74] Hughes on Trade Marks, looseleaf (Toronto; Butterworths; 1984) says (at page 629) the following about the inquiry to be made when determining whether the nature of the trade in different products is likely to cause confusion:
| The risk of confusion is greater where the wares or services, though dissimilar, are distributed in the same types of stores or are of the same general category of goods; for example, if both items are in the general category of household products and are sold in similar places, then confusion is more likely. However, where one mark refers to household products and the other to automotive products and they are distributed in different types of shops, there is less likelihood that consumers will mistake one mark for the other. |
[75] The evidence in this case indicates that there is an overlap in the retail outlets for the parties" products. Both are sold in major department store chains, often in close proximity. On the other hand, the evidence also indicates that the goods of only one of the parties are likely to be sold in smaller boutiques and specialty shops.
[76] A finding of a likelihood of confusion may be made on the basis of the similarity of the channels of trade, even though the goods in question are not sold in the same stores. Thus, in Cartier Inc. v. Cartier Optical Ltd. (1988), 20 C.P.R. (3d) 68, 74 (F.C.T.D.), Dubé J. said:
| However, so as to establish the likelihood of confusion, it is not necessary to prove that the wares are sold in the same places, provided they are of the same general class, could eventually be sold in the same places, and the parties are entitled to do so. |
See also Eminence SA v. Registrar of Trade Marks (1977), 39 C.P.R. (2d) 40 (F.C.T.D.) and Everex Systems Inc. v. Everdata Computer Inc. (1992), 56 F.T.R. 132 (F.C.T.D.).
[77] There is no doubt in this case that the goods in question could eventually be sold in the same places, although some stores may only carry the goods of one party. Thus, the question remains whether the products can be classified as belonging to the same "general category of goods."
[78] In the Eminence case, supra, Dubé J. held that hair products and perfume were in the same general category of goods for the purpose of this test. Similarly, as I have already noted, Hughes discusses general categories of goods such as household products and automotive products in the context of the nature of the trade.
[79] In this case, the question is whether goods such as jewellery, belts and handbags can be said to be in the same general category of goods as fragrances, cosmetics, lotions, eyeglasses, sunglasses and eyeglass accessories.
[80] The evidence of Ms. Barham is especially helpful on this point. Under questioning from the appellant"s counsel, she stated that belts, handbags, sunglasses, eyeglasses, makeup, fragrances, watches, hosiery, and jewellery could all be considered to be fashion accessories. She later stated that fashion accessories are part of the fashion industry: respondent"s Application Record , page 249, Question 110. However, she subsequently modified this statement by saying that, of the products on this list, she would not regard make-up as a fashion accessory: see page 289 of the Application Record.
[81] Ms. Barham is an expert in the fashion industry. She is the editor of a national fashion magazine, and is a member of an organization called The Fashion Group International, a global, non-profit, professional organization with over 6,000 members.
In paragraph 5 of her affidavit, Ms. Barham stated that in the course of her career she has acquired "expertise in all aspects of the fashion industry, including jewellery, toiletries and eyeglasses". The products of this industry, whether it is characterized as the fashion industry or the fashion accessory industry, would seem to me to fall into the same "general category of goods."
[82] The respondent noted that Ms. Barham stated on re-examination that the term "fashion accessory" is very broad and may include things as diverse as automobiles. Accordingly, counsel submitted, there can be no presumption that the goods of such a diverse industry normally flow through the same channels of trade. In my opinion, the diversity of the goods included within the fashion accessory industry does not necessarily detract from a finding that some, at least, of the products in the parties" applications are within the same general category of goods.
[83] A category such as household goods may also be very broad and may contain very different products. However, the purpose of the inquiry is to determine whether the products are in the same general category of goods and so could be sold in the same, or similar, stores in the future.
[84] In light of the evidence that was not before the Registrar, I find that the goods sold by the appellant and those proposed to be sold by the respondent are not too diverse to prevent them from being in the same general category of goods. Further, while not identical, the channels of trade through which they normally pass overlap.
[85] Thus, contrary to the Registrar"s conclusion, the "nature of the trade" factor supports a conclusion that there might be a reasonable likelihood of confusion. However, since the marks relate to different goods within a broad category, I do not regard the support as being particularly strong.
(e) Resemblance between the trade marks or trade names
[86] The Registrar found that the marks GARBO and GARBO DESIGN, and GRETA GARBO, were similar in appearance and sound. On the other hand, he concluded that the idea conveyed to consumers by the respondent"s mark was the actress Greta Garbo, while the appellant"s GARBO and GARBO DESIGN marks conveyed to some consumers the idea of the surname Garbo. In contrast, the Registrar found that the marks GRETA GARBO and T.G.I.F. BY GARBO were dissimilar in sound, appearance and the ideas that they conveyed.
[87] The respondent advanced four arguments with respect to the Registrar"s finding that the GARBO marks and GRETA GARBO are similar in sound and appearance, but not in the ideas that they convey.
[88] First, the respondent argued that the association established between the mark GRETA GARBO and the famous actress is so clear as to eliminate any possibility of confusion. That is, on seeing this mark consumers would associate it with the actress or her estate, and not with the appellant"s goods.
[89] Admittedly, there was little direct evidence on this point, save that of Ms. Barham which was not before the Registrar. She deposed that, if she saw the name "Greta Garbo" on goods, she would assume that they came from the estate of the famous actress and from no other source. It is true that Ms. Barham is not an average consumer with imperfect recollection, but an expert in the fashion industry. However, I do not think that this detracts from the weight of her evidence on this issue because it is not related to her expertise.
[90] Ms. Barham"s conclusion on this issue was no doubt reinforced by her view that the GARBO marks had acquired little consumer recognition. She testified that, although an expert on the fashion industry, she had little recollection of goods sold under the GARBO marks. Hence, I attach considerable weight to Ms. Barham"s evidence on the question of whether average consumers are likely to associate the mark GRETA GARBO with the appellant"s goods.
[91] Second, the respondent pointed out that, for purposes of comparison, the first portion of a trade-mark should be treated as being of particular significance: Molson Companies Ltd. v. John Labatt Ltd. (1990), 28 C.P.R. (3d) 457, 461 (F.C.T.D.). Hence, the fact that "Greta" is the first part of the respondent"s trade-mark goes a long way towards establishing that the marks are not similar in appearance or sound.
[92] Third, apart from the words themselves, counsel also submitted that the marks are sufficiently different in visual appearance so as to make confusion unlikely, even though the names are both written in script. However, GARBO is written in a somewhat exuberant, stylized script, while GRETA GARBO appears in neat and even script that is rather understated in style.
[93] Fourth, the respondent relied on statements made against interest by the appellant prior to the registration of the mark GARBO, to the effect that its proposed mark did not suggest an association with the actress Greta Garbo in breach of paragraph 9(1)(k) of the Act.
[94] While this may not strictly amount to an estoppel, it does constitute some evidence that consumers are not reasonably likely to be confused when they see the marks because they have already associated the name Greta Garbo with the appellant"s goods. To the extent that "Garbo" has an association beyond that of a surname, it may merely suggest to consumers the glamour associated with movies.
[95] To summarize, in the light of the additional evidence I cannot conclude that the Registrar was wrong to find that the marks were similar in appearance and sound, but different in the ideas that they suggested, although I would not regard the similarities as being particularly striking.
(f) Conclusion
[96] Despite the additional evidence adduced by the appellant establishing that there was a similarity of the trade and that it passed through overlapping channels, I am not satisfied that the Registrar was wrong in his conclusion that there was no reasonable likelihood of confusion between the appellant"s GARBO marks and trade names, and the proposed mark of the respondent.
[97] This is partly because the additional evidence in the affidavit of Ms. Barham considerably strengthened the Registrar"s finding on the idea suggested by the respondent"s mark as a result of the enduring fame of Greta Garbo, and on the relative obscurity in the fashion industry of the appellant"s goods, trade names and marks.
[98] While the Registrar found in favour of the appellant on most of the criteria listed in subsection 6(5), namely, the distinctiveness of the marks, the extent to which they were known and the duration of their use, they were not such as clearly to indicate that confusion was likely. Rather, much of the Registrar"s analysis was a comparison of how the parties fared on the relevant criteria. The additional evidence on the overlap of the channels of trade was insufficient to satisfy me that the Registrar"s decision was wrong, particularly in light of the evidence of Ms. Barham.
[99] The factors listed in subsection 6(5) are not to be approached mechanically, but with judgment and having regard to "all the surrounding circumstances". Taken as a whole the evidence does not suggest that Harriet Brown"s use of the GRETA GARBO mark is likely to subject the Garbo Group to unfair competition by inducing customers to think that the applicant"s goods came from the same source as those sold under any of the GARBO marks.
[100] For these reasons the appeal is dismissed. The parties have 14 days from the date of this decision to make submissions in writing on the award of costs.
OTTAWA, ONTARIO "John M. Evans"
NOVEMBER 16, 1999 J.F.C.C.
Email this Content