IMPERIAL TOBACCO CANADA LIMITED v. MARLBORO CANADA LIMITED, 2018 FC 503

Date: 2018 0615 Dockets: T-619-13 T-620-13 T-2608-14 T-2609-14 T-2610-14 T-2611-14 T-2612-14 T-2613-14 T-2614-14 Citation: 2018 FC 503 Ottawa, Ontario, June 15 , 2018 PRESENT: The Honourable Mr. Justice Annis BETWEEN: IMPERIAL TOBACCO CANADA LIMITED and MARLBORO CANADA LIMITED Applicants and PHILIP MORRIS BRAND SARL Respondent AMENDED JUDGMENT AND REASONS TABLE OF CONTENTS I. Nature of the Matter 4 II. Factual Background and History between the Parties 5 A. The Parties 5 B. History of the acquisition and use by ITL of the trade-mark MARLBORO in Canada 5 C.

History of the redesigning and relaunch of internationally successful Philip Morris’ Marlboro brand of cigarettes 6 D. PM introduces the MATADOR and MAVERICK products sold in Canada 7 E. The “Dark Market” created by cigarette market regulations 9 F. The launch of the “no-name” packaging in 2006 and ensuing litigation 10 G. Non-cigarette wares 11 H. Federal Courts decisions relating to confusion of PM’s no-name packages and ITL’s MARLBORO mark 12

(1) Philip Morris 2010 12

(2) Philip Morris 2012 13 I. The Applications 17 J. The Oppositions 17 III. The Decisions under Review 18 IV. Issues 19 V. New Evidence on Appeal 19

A. The Applicant’s 1st New Evidence 20

(1) ITL’s interpretive evidence 20

(2) ITL’s survey evidence 20 B. Respondent’s Reply Evidence 22

(1) The Redesigned Rooftop litigation evidence — (pending litigation) 22

(2) PM’s reply survey evidence 23 C. Applicant’s 2nd New Evidence 23

(1) The Redesigned Rooftop litigation evidence — (pending litigation) 23

(2) International rooftop design use evidence 24

(3) Cigar wares packaging evidence 24 VI. Standard of Review 24 VII. Analysis 25 A. Legal Principles on the Admission of New Evidence and Survey Evidence 25

(1) Admitting new evidence 25

(2) Admitting survey evidence 26 B. Is the Mills Survey evidence admissible in any of the Decisions? 27

(1) DECISION 1: 28

(2) DECISIONS 2 and 3: 32

(3) DECISION 4: 34 (

a) Should the conclusions of the Mills Survey also apply to TMA 1,460,676 (Roof Design) and TMA 1,478,470 (Roof Design w/ Empty Roof Line)? 34 • TM Application 1,460,676 (Roof Design) 35 • TM Application 1,478,470 (Roof Design with Empty Roof Line) 36 (

b) Did the Mills Survey questionnaire ask the wrong question? 39 • Requirements that the questions provide a concrete context without being speculative or suggesting an answer 39 • Survey questions are not suggestive or speculative by presenting the design marks in a no-name getup as opposed to the marks as they are sought to be registered 43 (

c) The Admissibility of the Applicant’s remaining new evidence 47

(4) DECISIONS 5 and 6: 48

(5) DECISION 7, 8 and 9: 49 C. Is there likelihood of confusion between each of the applied-for design marks in the Decision 4 and the registered trade-mark MARLBORO? 50

(1) Overall consideration of subsection 6(5) factors and all surrounding circumstances 51

(2) Degree of resemblance factor favours the Applicant 53 D. Are the remaining Decisions reasonable? 58 VIII. Conclusion 59 I. Nature of the Matter

[ 1 ] These are nine appeals brought by the Applicant, Imperial Tobacco Canada Limited [ITL] against the Respondent, Philip Morris [PM] pursuant to

section 56 of the Trade-Marks Act, RSC 1985, c T-13 [Act] from the decisions of a Hearing Officer of the Trade- Marks Opposition Board [the Board or Registrar], dated November 21, 2012 [Decisions 1 and 2, as defined below] and October 6, 2014 [Decisions 3-9, as defined below] [together the Decisions] concerning registration proceedings by PM for 17 trade-marks [the TM Applications]. The Decisions rejected all of ITL’s oppositions to the registration of PM’s marks on the ground that they were not confusing with its registered trade-mark MARLBORO (no. TMDA 55,988) pursuant to section 12(1) (

d) and 6(5) of the Act. [ 2 ] This case draws on a series of previous jurisprudence between these parties, and more specifically with respect to the issue of the reliance on survey evidence, which informed the outcome of Decisions of the Board, and additional survey evidence sought to be introduced in these appeals with the view to affecting the outcome of these matters. In particular, two Federal Courts decisions, the first in Philip Morris Products S.A. v. Marlboro Canada Limited , 2010 FC 1099 [ Philip Morris 2010 ] and the second, the Federal Court of Appeal case overturning it in Marlboro Canada Limited v.

Philip Morris Products S.A. , 2012 FCA 201 [ Philip Morris 2012 ] are significant in this respect. [ 3 ] For the reasons that follow, new evidence is admitted in the appeal of Decision 4 in the matter T-2609-14 resulting in the appeal being allowed and the TM Applications associated with the Decision being refused. No new evidence is admitted in the remaining Decisions, resulting in their appeals being dismissed. II. Factual Background and History between the Parties A. The Parties [ 4 ] ITL is the largest tobacco company in Canada.

It is a majority shareholder of the Applicant Marlboro Canada Limited and an exclusive licensee of Marlboro Canada with respect to the registered trade-mark MARLBORO (no. TMA 55,988). ITL is wholly owned by British-American Tobacco Plc [BAT]. [ 5 ] The Respondent Philip Morris Brands Sàrl [PM Brands] is a subsidiary of PM and is the registered owner of a number of trade- marks and trade-mark applications (including what are described as the " “Rooftop Design Marks” " at issue) in Canada.

PM Brands licenses these trade-marks to a related subsidiary of PM in Canada, Rothmans Benson & Hedges Inc. [RBH], which competes with ITL in the sale of tobacco products. For ease of reference, whenever the Respondent is referred to in any capacity, the reasons will refer simply to the acronym of " “PM” " . [ 6 ] PM and BAT are the two largest publicly-owned tobacco companies in the world and compete for market shares in over 160 countries, including Canada. B.

History of the acquisition and use by ITL of the trade-mark MARLBORO in Canada [ 7 ] The history of the parties dates back to 1924, when the predecessor-in-title of PM assigned its rights to the word mark " “MARLBORO” " to a predecessor-in-title of ITL (Tuckett Tobacco Company Limited). The circumstances surrounding the assignment and related corporate histories of both parties are discussed in Philip Morris Inc. v. Imperial Tobacco Ltd., 1987 CarswellNat 701 (FCA) . The word mark was subsequently registered in Canada in 1932 under TMA 55,988 and has since been continuously used in the country.

A print of ITL’s current packaging is provided below: C. History of the redesigning and relaunch of internationally successful Philip Morris’ Marlboro brand of cigarettes [ 8 ] Following the assignment, PM continued to sell unfiltered cigarettes in the U.S. and other countries in the original Marlboro packaging. In 1955, PM redesigned and re-launched the Marlboro brand of cigarettes internationally with a cowboy-themed advertisement campaign promoting a " “full-bodied” " American blend tobacco and a filter.

The cigarettes were sold in a packaging incorporating for the first time the now well-known red " “roof” " chevron design element (the " “Rooftop Design Mark” " or " “International Package” " ) featured at the top of the package, which is the design mark at the heart of these appeals. PM’s Marlboro became the top-selling cigarette in the world by 1972 and this remains the case some 40 years later. It is sold in over 160 countries in the

world with the notable exception of Canada. PM’s International Package is reproduced below: D. PM introduces the MATADOR and MAVERICK products sold in Canada [ 9 ] In 1958, PM started selling in Canada, and is still selling to date, a Virginia blend (milder) cigarette incorporating the same design elements as its Rooftop Design Mark, with the word mark " “MATADOR” " in the place of " “MARLBORO” " on the packaging sold internationally. A similar product was sold using the word mark " “MAVERICK” " in 1970 and subsequently discontinued in 1978.

Neither MATADOR nor MAVERICK ever enjoyed more than a negligible market share. [ 10 ] Pursuant to the Tobacco Products Labelling Regulations (Cigarettes and Little Cigars), SOR/2011-177, tobacco manufacturers are now facing heightened constraints with respect to the display of their design marks, which is illustrated in the right hand version of the MATADOR package, as depicted below: [ 11 ] Between 1980 and 2005, PM registered five different design marks, in respect of cigarettes, claiming prior use in association with MATADOR — drawings and specimens of the MATADOR package had been submitted as proof of usage.

None of these registrations had been opposed by ITL at the relevant times, as the parties do not contest that consumers generally do not make any association with ITL’s MARLBORO mark, when MATADOR or MAVERICK appears on the PM packaging using the Rooftop Design Mark, which was confirmed in Philip Morris 2010 , paras 31, 37, 173-180. [ 12 ] Then in 2006, PM proceeded by registering TMA 670,898 [the Silver Design Mark], which did not claim a similar use as in previous registrations. It featured a Rooftop Design Mark with the " “PM” " initials.

ITL filed a time extension to oppose the registration but elected not to pursue the opposition. This was of no consequence on the matter as the trial judge in Philip Morris 2010 agreed that ITL could not have envisaged that the design mark would be used without a brand name. [ 13 ] These six registrations (illustrated below for reference) will be referred to collectively as the " "Registered Rooftop Design Marks" " . They were the subject of contention in the Federal Courts litigation proceedings. The design on the far right below is that of the Silver Design Mark referred to above.

It was eventually featured as one of three cigarette package designs launched by PM in 2006 without any brand name. Reg. No. TMA252,082 Reg. No. TMA252,083 Reg. No. TMA254,670 Reg. No. TMA 274,442 Reg. No. TMA465,532 Reg. No. TMA670,898 E. The " “Dark Market” " created by cigarette market regulations [ 14 ] Canada’s cigarette retail market has been increasingly heavily regulated over the last years.

Advertising is now almost forbidden and, since 1997, unlike almost all other consumer products, handling of cigarette products prior to purchasing is prohibited by the Tobacco Act , SC 1997, c 13. [ 15 ] Canadian consumers have thus been required to explicitly identify the desired cigarette product at the time of purchase, either by pointing at same or verbally. Starting in 2004, pointing was no longer an option as cigarette products must be hidden from consumers’ view (the so-called " “Dark Market” " ).

The Dark Market was found in the Federal Courts decisions to be the significant contributing factor that led to confusion when PM introduced no-name cigarettes in 2006. The disputes between the parties in this and previous litigation relate to concerns of PM applying its Rooftop Design Mark in some fashion to unbranded wares so as to raise concerns of

confusion with ITL’s MARLBORO mark. F. The launch of the " “no-name” " packaging in 2006 and ensuing litigation [ 16 ] In 2006, shortly after registering the Silver Design Mark (as illustrated above), in the context of the Dark Market which was gradually being implemented throughout Canada, PM Brands’ predecessor-in-title, through its Canadian licensee RBH, began selling, distributing and advertising in Canada American blend cigarettes in the packaging depicted below which was almost identical to PM’s International Package, except for the fact that it did not feature any brand name.

As can be seen below, the package incorporated the slogan " “COME TO WHERE THE FLAVOR IS” " on the front panel, which catchphrase had been used extensively in the context of PM’s worldwide advertising in connection with the International Package. This was also the first time in worldwide cigarette history that a cigarette product was sold without a name. [ 17 ] Other differences with the International Package were (

i) the presence of the mandatory 50% health warning, which basically resulted in the shrinking of the design of the International Package by the removal of the MARLBORO mark, with the effect of making the Rooftop Design Mark more prominent in the overall packaging; (ii) the previously advertised phrase " “COME TO WHERE THE FLAVOR IS” " (never before on packaging) replaced the " “20 CLASS A CIGARETTES” " situated above the bottom line on the International Package; (iii) the insertion of number " “20” " in the middle of the bottom line, but otherwise maintaining it; and (iv) the addition of the mention of " “WORLD FAMOUS IMPORTED BLEND/MÉLANGE IMPORTÉ DE RENOMMÉE MONDIALE” " , which appears only on Canadian packaging, on the side, and implicitly refers to the switch to the full-bodied American blend tobacco sold internationally. [ 18 ] All these design and word features in combination were referred to in the Federal Courts litigation as the " “no-name” " package.

This characterization is opposed to that with the getup limited to any single element of the design on a package, such as the Rooftop Design Mark. In the Dark Market the no-name package was found by the Federal Court of Appeal in Philip Morris 2012 to be confusing with ITL’s MARLBORO mark and a permanent injunction issued barring its use. G. Non-cigarette wares [ 19 ] PM’s Registered Design Marks that are being opposed as confusing also apply to non-cigarette wares which include other smoking products such as cigars and cigarillos, and smokers’ articles which include ashtrays and lighters.

As part of its marketing campaign in support of its 2006 no-name cigarettes, PM provided ashtrays and lighters to retailers bearing only its Rooftop Design Mark as demonstrated in the pictures below. H. Federal Courts decisions relating to confusion of PM’s no-name packages and ITL’s MARLBORO mark

(1) Philip Morris 2010 [ 20 ] Concurrently with the launch of its no-name packaged cigarettes, PM commenced its action seeking a declaration that its newly released no-name packages did not infringe the rights of ITL in its MARLBORO registration ( Philip Morris 2010 ). [ 21 ] The trial decision is of limited significance in this matter inasmuch as the Court of Appeal concluded that due to errors in lower court’s reasoning, it should revisit and decide the matter anew. Nevertheless, three points stand out. The trial decision concluded that despite the shortcomings of the survey [the Chakrapani Survey], the evidence demonstrated that there was " “a significant degree of

confusion [as to] how to refer to the no-name product, especially among consumers”" and that a "“large number of respondents seem toassociate the Plaintiffs’ [ITL’s] product to the international PM Marlboro”" (Philip Morris 2010, para 282). [22] The second relevant aspect of the trial decision was the Court’s conclusion to deny the aforesaid confusion between the marksdisclosed by the Chakrapani Survey evidence largely because the phrase "“ideas suggested”" in paragraph 6(5)(

e) of the Act "“should berestricted to those ideas inherent to the nature of the trade-marks in question (for example, the design of a Penguin giving the idea of apenguin)”" (Philip Morris 2010, para 290). This finding was the basis upon which the trial decision was set aside. [23] Finally, the decision in Philip Morris 2010 impacted on this matter by its conclusion that PM’s Registered Rooftop Design Marksexcluded a finding of infringement based on what was described as the Remo defence. This refers to the decision in Remo ImportsLimited v.

Jager Cars Limited, 2007 FCA 258 at paragraph 111 to 113 which adopted the Ontario Court of Appeal decision in MolsonCanada v. Oland Breweries Ltd,. (2002), (ON CA), 19 C.P.R. (4th) 201. It was applied by the trial Judge toconclude that PM’s registrations afforded them an absolute defence to infringement.

The decision is of importance in this respect onlybecause in overturning the trial decision, the Court of Appeal distinguished between the confusion arising from the combination ofelements in the no-name packaging getup that was found to be infringing and any confusion alleged to be caused by a single element ofthe packaging getup. In particular, the Court concluded that one could not extract any single element from the packaging getup toconclude that it would be confusing in its own right.

(2) Philip Morris 2012 [24] The Federal Court of Appeal in Philip Morris 2012 overturned the declaration of the trial decision that the no-name package did notinfringe ITL’s MARLBORO word mark. Instead, it found confusion in the marks and issued a permanent injunction against PM’s no-name package. [25] In carrying out her own analysis of the confusion issue, Madam Justice Gauthier concluded that paragraphs (

a) to (

d) of subsection6(5) favoured ITL, as had been the trial judge’s finding. However, the Court proceeded to consider the registered word markMARLBORO with each version (red, gold and silver) of PM’s no-name getup as proposed by ITL. Justice Gauthier concluded that thecombination of elements on the packaging fell within the definition of a trade-mark in

section 2 of the Act (Philip Morris 2012, para 67). [26] Thereafter, turning to paragraph 6(5)(e), the Appeal Court rejected the trial Judge’s limited

interpretation of "“ideas suggested”".Instead at paragraphs 76 to 78 of its reasons, the Court concluded that the confusion evidenced from the Chakrapani Survey could not beignored, based upon "“a purposive and contextual

interpretation of paragraph 6(5)(e)”". Moreover, Justice Gauthier added that the"“resemblance in unusual ideas suggested by any one of the marks once established would have to be considered as part of thesurrounding circumstances (opening words of subsection 6(5))”". This conclusion thereby paved the way for the Chakrapani Surveybeing applied to establish confusion between the marks.

This in turn was the foundation for the Court’s broad permanent injunctionenjoining PM from using the no-name package in association with cigarettes or other tobacco products (Philip Morris 2012, para 126). [27] Another significant aspect of the Philip Morris 2012 decision is the Court’s admonishment at paragraph 76 that "“when oneinvokes a resemblance based on something out of the ordinary, [i.e. not limited to those ideas inherent to the nature of the trade-marks inquestion, per the trial judge] evidence will be required to satisfy the Court that the particular association or suggestion does indeed existas a matter of fact before it is considered in the analysis under paragraph 6(5)(e)”." [Emphasis added.] [28] The Court interprets this passage as emphasizing the need for persuasive evidence to establish confusion when there is no inherentassociation between the two trade-marks which bear no resemblance to each other.

In other words, standing back and examining the twoFederal Courts decisions, each largely turned on the inadmissibility or admissibility of the Chakrapani Survey evidence as beingrelevant. The Court finds that the admissibility of the new survey evidence of Mr. Mills [the Mills Survey] is similarly the singledeterminative issue in this matter. [29] The Court of Appeal also was required to explain why the Remo defence would have no application to prevent ITL from succeedingon its application.

First, it pointed out that the confusion in the marks only arose long after the registrations when legislation created theDark Market. ITL could not be criticized for not challenging proceedings when no basis to do so existed at the time of registration.

It isfor this reason that the court found the Remo defence to not apply in these appeals, even though it concluded that PM’s Rooftop DesignMarks are confusing with the MARLBORO mark and despite them being identical to some of PM’s Registered Rooftop Design Marks. [30] Second, the Court distinguished between the combination of elements on the no-name packaging, and the individual markscomprised in the packaging.

It concluded that it was required to "“determine whether it is the unregistered [no-name] combination aloneor the individual marks, used essentially as registered, that are confusing”" (Philip Morris 2012, para 103). At paragraph 108 of thedecision, the Court pointed out that the surveys [implicitly referring mostly to the Chakrapani Survey] was not specifically designed totest whether consumers associated the name MARLBORO to cigarette packages bearing only the individual Rooftop Design Mark asregistered.

Instead the Court found that "“[w]hat was presented to the participants in the survey by both parties’ experts was the no-namepackage as a whole”. "[Emphasis added.] [31] Based on these findings, the Court concluded that there was no probative evidence that PM’s Registered Rooftop Design Markswere confusing with the word mark MARLBORO (Philip Morris 2012, para 111).

Essentially, the Chakrapani Survey failed to assess theindividual Registered Rooftop Design Marks for confusion with the MARLBORO word mark. [32] The Court’s analysis not only formed a basis for it to reject ITL’s attack on PM’s Registered Rooftop Design Marks, moresignificantly for these proceedings, it also struck at the heart of ITL’s case in all nine opposition proceedings before the Registrarseeking to challenge PM’s TM Applications. ITL based its case in every opposition before the Board upon the same Chakrapani Surveyevidence.

It was found to be irrelevant by the Board because it contained no evidence demonstrating that any of PM’s individual TMApplications were confusing with the MARLBORO word mark in accordance with the decision in Philip Morris 2012. [33] In a situation when there is no visual or other physical resemblance between PM’s design marks and the MARLBORO word mark,

the absence of probative survey evidence demonstrating confusion was fatal to all of ITL’s opposition proceedings before the Board. I. The Applications [ 34 ] Between 2006 and 2010, PM applied to register 17 design marks in Canada (also referred to as the TM Applications, as described above), the particulars of which are found in

Annexe A to these reasons. These design marks all incorporate, what might be described as different variants of the Rooftoop Design Mark. All 17 TM Applications were approved and subsequently advertised in the Trade-Marks Journal. J. The Oppositions [ 35 ] ITL opposed all seventeen TM Applications filed by PM, including TMA 1,298,547 and 1,299,494 (Decision 1 below), which had been filed in 2006 — a few months prior to the launch of the no-name package. The grounds of opposition raised before the Registrar were related to confusion-based claims in respect of

a) registrability pursuant to paragraph 12(1) (

d) of the Act as the TM Applications are confusing with ITL’s MARLBORO word mark,

b) non-entitlement to registration pursuant to paragraphs 16(1) (

a) and 16(3) (

a) of the Act as the word mark MARLBORO has been used by ITL since before the priority dates of filing of PM’s TM Applications, and

c) non- distinctiveness of the trade-marks at issue pursuant to

section 2 of the Act. [ 36 ] ITL relied on the Chakrapani Survey as its primary evidence in its submissions before the Board. In nine decisions (Decisions 1 to 9 as defined below), the Board rejected all oppositions, citing most prominently the absence of evidence demonstrating confusion between marks bearing no resemblance to each other on the basis that the Chakrapani Survey was not relevant. [ 37 ] Between 2013 and 2014, ITL filed notices of appeal to this Court of all nine Decisions.

In June 2016, in response to a request by the parties, Prothonotary Morneau issued a Consolidation Order, allowing the parties to serve additional evidence [the Consolidation Order]. ITL’s additional evidence and related PM’s response are discussed below. In the context of the present appeal, ITL focuses its submissions primarily on the confusion between PM’s Rooftop Design Mark and its registered word mark pursuant to paragraph 12(1) (

d) of the Act and in particular the factors of confusion described in subsection 6(5) of the Act . III. The Decisions under Review [ 38 ] In the Decisions, the Board first established that ITL bears the initial evidential burden to adduce sufficient admissible evidence to support the facts alleged in support of its grounds of opposition, and therefore that PM has the legal onus of establishing, on a balance of probabilities, that its applications comply with the Act , and that ITL’s grounds of opposition at issue should not prevent registration of the design marks at issue.

The Board then proceeded to review the history of the use of the Rooftop Design Mark, its launch in the Canadian market, and the legal disputes that ensued. [ 39 ] In addressing the non-registrability of the trade-marks at issue pursuant to paragraph 12(1) (

d) of the Act , the Board outlined the test for confusion, summarized subsection 6(2) of the Act and stated that the Board is directed by subsection 6(5) of the Act to " “have regard to all the surrounding circumstances” " . The Board determined that the overall consideration of subsections 6(5) (a), (b), (c), and (

d) statutory confusion factors, relating to the inherent distinctiveness of the marks and the extent to which they have become known, the length of use, the natures of wares, and the nature of the trade, favoured ITL. Nevertheless, the Board dismissed the oppositions on the basis of a lack of visual or phonetic resemblance between the marks, pursuant to paragraph 6(5) (

e) relating to the degree of resemblance factor. [ 40 ] The Board noted that there is no evidence of the extent to which these marks have become known in Canada and could not conclude to an intangible association with the graphical component. The Board agreed with PM in that there was no degree of resemblance between the parties’ marks in the ideas they suggested. In this regard, the Board found that the Chakrapani Survey evidence was of no assistance to ITL in any of the proceedings because, as was pointed out in Philip Morris 2012 , Dr.

Chakrapani’s studies were not designed to test consumer reactions to the instant Marks (i.e. the Rooftop Design Mark in isolation), but to a cigarette package made of a particular combination of various elements, which included the Rooftop Design Mark. IV. Issues [ 41 ] The following issues arise in this application: 1 . What is the applicable standard of review of the Board’s decisions in view of the new survey evidence? 2 . In light of the new survey evidence, are the trade-marks at issue confusing with ITL’s registered word mark " “MARLBORO” " ? 3 . Are the Board’s decisions reasonable? V.

New Evidence on Appeal [ 42 ] ITL has filed new evidence pursuant to subsection 56(5) of the Act . It may generally be categorized into three groupings: first, consisting of evidence to assist the Court in interpreting certain descriptions of wares (which could have been introduced in any event in submissions); second, the Mills survey evidence in the form of a market research study by Mr. Don Mills intended to address the limited scope of the Chakrapani Survey evidence described in Philip Morris 2012 ; and third, other evidence described below.

In reply, PM has tendered evidence critiquing the methodology and conclusions of the Mills Survey.

A. The Applicant’s 1st New Evidence

(1) ITL’s interpretive evidence [ 43 ] ITL has filed the Affidavit of Ms. Andrea Plouffe [the Plouffe Affidavit], a legal clerk employed by ITL’s solicitors, dated October 15, 2013. The Plouffe Affidavit provides excerpts from the following legal resources relied upon by ITL as interpretive aids for the purpose of the appeals: (

i) dictionary definition of the words " “tabac” " , " “tobacco” " , and " “including” " ; (ii) copies of the Tobacco Act and Excise Act , printed from the Department of Justice Website; (iii) excerpts from Canadian Intellectual Property Office’s [CIPO] Trademarks Examination Manual; (iv) certified copies of various trade-mark registrations; and (

v) a printout of the Supreme Court of Canada Decision dismissing the application for leave to appeal from the judgment in Philip Morris 2012 . ITL claims that these resources establish that the phrase " “[t]obacco, raw or manufactured, including.…” " in the description of wares in Decision 1 extends to include " “cigarettes” " .

(2) ITL’s survey evidence [ 44 ] The survey evidence filed by ITL is a 15-page Affidavit of Mr. Mills sworn on November 4, 2013 [the Mills Affidavit]. He is a survey expert and co-founder, Chairman and Chief Executive Officer of Corporate Research Associates Incorporated [CRA], a national market research company. Mr. Mills designed, implemented and conducted a national survey of 1,257 daily adult smokers from a sampling of 19 Canadian communities.

The methodology as set out in the survey report entitled " “Market Association of Rooftop Design Marks — A Study among Adult Smokers” " (AR, p 21,570) provides for testing a print of a three-dimensional tobacco product, featuring one of the three representative design marks, and a " “Control Design” " , placed in front, side and top views, in black and white. The rationale behind the utilization of three representative design marks was discussed in the Smart & Biggar letter of instructions dated August 28, 2013 (AR, p 21,140). As depicted below, the mandatory health warnings were included in each sample. Mr.

Mills was cross- examined on his Affidavit in December 2016. [ 45 ] The Mills Affidavit attests that the survey establishes a link between the three representative design marks of PM in the format presented to survey participants, taken in isolation, and ITL’s MARLBORO word mark.

The relevant passage in the Mills Affidavit disposing of the question of " “association” " — on a first / immediate impression basis — relied upon by ITL in their Memorandum of Fact and Law, at paragraph 69, reads as follows: Marlboro is statistically significantly more likely to be associated with each of the Rooftop Design marks tested than with the Control Design (Rooftop Full=22%, Rooftop Thin=12% or Rooftop Window=10%, Control Design=2%). (AR, p. 21,119, para 25) [ 46 ] ITL has also filed the Affidavit of Ms. Joyce Rees, dated November 5, 2013 [the Rees Affidavit]. Ms.

Rees is co-founder and President of Advitek, a data collection service company for the market research industry, specializing in quantitative market research. In her affidavit, Ms. Rees attests that CRA mandated Advitek to collect data underlying the findings in the Mills Survey report. The Rees Affidavit explains that interviews were conducted using the questionnaire developed by CRA, which was coded in Advitek’s own proprietary survey software system, the Tapestry Computer Assisted Personal Interviewing system. Ms.

Rees attests that she prepared instructions to be given to interviewers and personally attended all training sessions of said interviewers in preparation for the field surveys. B. Respondent’s Reply Evidence

(1) The Redesigned Rooftop litigation evidence — (pending litigation) [ 47 ] PM has filed the Affidavit of Ms. Adriana Morillo, legal assistant with PM’s affiliate, sworn July 18, 2014. The Affidavit attaches printouts of the current MATADOR packaging as well as PM’s redesigned (Rooftop) packaging launched on July 23, 2012 as depicted below [the Redesigned Rooftop], which is the subject of the pending trade-mark infringement action brought by ITL (Court File No. T- 1280-14).

(2) PM’s reply survey evidence [48] In response to the Mills Survey, PM has also filed the Affidavit of Dr. Ruth Corbin [the Corbin Affidavit], Chair and formerManaging Partner of CorbinPartners Inc., a marketing science company, sown July 17, 2014. In her Affidavit, Dr. Corbin opines on thereliability of the Mills Survey, in particular the survey methodology and conclusions. Dr. Corbin explains that the survey is not relevantas the line of questioning did not assess a likelihood of confusion and addressed the brands rather than the ideas suggested (CorbinAffidavit, paras. 42-3). Dr.

Corbin was cross-examined on her Affidavit in December 2016. ITL submits that the criticisms in the CorbinAffidavit are of little consequence as the affiant failed to conduct a study of her own. C. Applicant’s 2nd New Evidence

(1) The Redesigned Rooftop litigation evidence — (pending litigation) [49] ITL has filed additional evidence following the Consolidation Order. The Affidavit of Van Khai Luong [the Luong Affidavit],paralegal employed by Counsel for ITL, dated July 29, 2016. The Luong affidavit provides various documents filed in the context of theRedesigned Rooftop litigation evidence: (

i) the parties most recent pleadings’ in the pending redesigned packaging, (ii) photographs ofthe red, gold and silver variants of the "“redesigned”" packaging at issue in the said proceedings, (iii) PM’s promotional materialsaddressed to retailers in the context of the launch of the 2012 Redesigned Rooftop package; and (iv) photographs of ITL’s currentMARLBORO packaging.

ITL submits that these documents demonstrate that PM’s efforts in recent years have been shifting towardshaving less emphasis on the word mark, and focusing on the primacy of the Rooftop Design Mark in the packaging — in Canada andinternationally. [50] PM disagrees that this evidence is relevant, noting that this argument is premature.

(2) International rooftop design use evidence [51] The Luong Affidavit also contains pictures of the red and gold variants of PM’s International Package available for sale inGermany and Italy, and various printouts of documents relating to PM’s international products. ITL submits that this is another exampleof PM marketing the dominant association of the Rooftop Design Mark with its cigarette and other associated products in its worldwideadvertising.

(3) Cigar wares packaging evidence [52] The Luong Affidavit further contains photographs of cigarillos and cigars sold in similar packaging to that of cigarettes. ITL claimsthat this additional evidence establishes the similarity between the shape of the packaging used for cigarettes, cigarillos, and small cigars,in support of the relevance of the survey evidence to those wares. VI.

Standard of Review [53] It is common ground that an appeal from a decision of the Registrar of Trade-marks in opposition proceedings is reviewable underthe reasonable standard, unless new evidence is put forward on appeal that would materially affect the Registrar’s findings. In such acase, the Court must reassess the decision of the Registrar, as the date of the Court’s decision, on the basis of the extended record anddraw its own conclusions on appeal: Shell Canada Ltd. v. PT Sari Incofood Corp., 2008 FCA 279 at para 22; Molson Breweries v.

JohnLabatt Ltd., (FCA), [2000] 3 FC 145 at para 11. [54] PM submits that there is no basis for reconsidering ITL’s oppositions as the new evidence put forward by ITL "“suffers from theexact same flaws as the “old” survey evidence before the Opposition Board: it fails to test consumer reactions to each of the applied-fortrade-marks, in isolation, and in association with the full range of goods listed in the applications”". PM claims that the decisions shouldbe reviewed on the reasonableness standard, based on the record before the Board.

ITL, on the other hand, stresses the probative value ofthe new survey evidence urging the Court to reconsider the validity of the oppositions afresh. [55] The Court will first consider whether the new evidence would materially alter the Board’s conclusions concerning the likelihood ofconfusion in the nine Decisions. This analysis leads the Court to conclude that the new evidence would not affect the conclusions onconfusion in any of the Decisions, with the exception of the TM Applications considered under Decision 4.

The new evidence isadmitted in regard to that Decision, and along with it the supplementary evidence relating to the International Packages andadvertisements. Thereafter, the Court will review Decision 4 de novo, based on the old and newly admitted evidence, to conclude that theapplications therein are confusing with the MARLBORO mark. The appeals in the remaining Decisions are rejected based on a standard

of review of reasonableness. VII. Analysis A. Legal Principles on the Admission of New Evidence and Survey Evidence

(1) Admitting new evidence [ 56 ] In assessing whether new evidence filed on appeal is material pursuant to subsection 56 (5) of the Act , the Court must assess to what extent the new evidence has a probative significance which extends beyond the material that was before the Registrar: Levi Strauss & Co. v. Vivant Holdings Ltd. , 2005 FC 707 at para 27 . [ 57 ] The applicable principles to determine whether new evidence would materially affect the Registrar’s findings is described by Mr. Justice de Montigny in Suzuki Motor Corp. v.

Hayabusa Fightwear Inc. , 2014 FC 784 at para 26 as follows: When considering the impact of the new evidence filed, quality is more significant than quantity… Additional evidence filed to fill the gaps or remedy deficiencies identified by the Registrar will generally be considered material . On the other hand, if the new evidence merely replicates what was already before the Registrar without substantially adding to the nature of the information already filed, it will not be considered sufficient to warrant a review on the standard of correctness […] [Emphasis added.]

(2) Admitting survey evidence [ 58 ] Consumer surveys can be admitted as evidence, if relevant and properly designed: Masterpiece Inc. v. Alavida Lifestyles Inc. , 2011 SCC 27 at para 94 [ Masterpiece ]. [94] The use of consumer surveys in trade-mark cases has been recognized as valid evidence to inform the confusion analysis. As Binnie J. noted in Mattel , often the difficulty with survey evidence is whether it meets the first of the Mohan requirements: relevance.

At para. 45, he further divided the question of relevance into two sub-issues: As to the usefulness of the results, assuming they are elicited by a relevant question, courts have more recently been receptive to such evidence, provided the survey is both reliable (in the sense that if the survey were repeated it would likely produce the same results) and valid (in the sense that the right questions have been put to the right pool of respondents in the right way, in the right circumstances to provide the information sought). [Emphasis in original.] [ 59 ] The considerations relevant to the admissibility of the Mills Survey in this matter, apart from it being new evidence, may be summarized from paragraphs 46 to 48 of Mattel U.S.A.

Inc. v. 3894207 Canada Inc. , 2006 SCC 22 [ Mattel ] as follows, with the Court’s emphasis: carried out in an impartial and independent manner; required to demonstrate a likelihood, and not merely a possibility of confusion so as to: o be relevant, in the sense of having sufficient probative value to support a finding of fact having connexion to the issue to be considered (that of confusion as to source); o be reliable, i.e. if the survey were repeated, likely producing the same results; o be valid, in the sense that: ▪ the right questions have been asked, i.e. directed to the issue of confusion, not to measure the public recognition of a word; ▪ the trade-mark used in the survey is precisely the trade-mark applied for [this issue is discussed further below]; ▪ put to the right pool of respondents, i.e. not exclude, but include appropriate respondents; and ▪ in the right way in the right circumstances to provide the information sought, i.e. not merely showing a design logo without any context and then subsequently immediately removing it from their site and asking questions, or not providing relevant information, or including questions suggestive of the answers.

B. Is the Mills Survey evidence admissible in any of the Decisions? [ 60 ] Inasmuch as the Applicant claims that the Mills Survey is admissible in all of the Decisions, which vary substantially in form and content, the Court will consider the issue in the case of each Decision.

(1) DECISION 1:

1,298,547 (Roof Design) 1,299,494 (Roof Design) [ 61 ] The marks in these two applications are, in the first case very similar and, for the second, apparently identical to the mark applied to the packaging used in the Mills Survey. However, there is no specific mention of " “cigarettes” " in the description of wares they claim to apply to. The description for both wares is identical in each application, as follows with the Court’s emphasis:

(1) Tobacco, raw or manufactured, including cigars, cigarillos, tobacco for roll your own cigarettes, pipe tobacco, chewing tobacco, snuff tobacco, tobacco substitutes (not for medical purposes); smokers' articles, namely, cigarette paper and tubes, cigarette filters, tobacco tins, cigarette cases and ashtrays not of precious metals, their alloys or coated therewith; pipes, pocket apparatus for rolling cigarettes, lighters; matches. [ 62 ] The Registrar concluded that the wares did not include " “cigarettes” " .

The Applicant argues that the Registrar erroneously indicated that the two applications do not cover cigarettes as a ware falling under " “Tobacco … manufactured” " . It might be noted that the Applicant does not make a similar argument that " “cigarettes” " should be included as tobacco products in TM Application 1,478,472 of Decision 6 where " “namely” " is used to describe the specific wares covered, i.e.

" “[t]obacco, raw or manufactured; tobacco products, namely cigars, cigarillos, …” " [emphasis added.] [ 63 ] Based on CIPO’s Trade-mark Examination Manual and the definition of the word " “including” " in dictionaries, the Applicant submits that the broad encompassing description of goods in " “[t]obacco, raw or manufactured” " when matched with the term " “including” " is not limited by the specific wares that follow.

Additionally, it argues that it is commonplace that a " “cigarette” " is not only a manufactured tobacco product, but the most common [the Court’s emphasis]. [ 64 ] The Respondent replies that pursuant to paragraph 30 (

a) of the Act , a statement of goods and services must be both " “specific” " and defined in " “ordinary commercial terms” " . Therefore, the Applicant’s preferred

interpretation of the term " “including” " in the context of these specifications of goods runs contrary to the Act and applicable jurisprudence [none referred to]. It notes that the purpose of the technical requirements for a trade-mark application, found in

section 30 of the Act, is to ensure that the scope of the claimed monopoly is precisely defined in the application. It would thus, be antithetical to the purpose of

section 30 of the Act to " “read in” " other goods not explicitly listed in the specification as the Applicant suggests. [ 65 ] The Court agrees with the Respondent that it is not appropriate to read in the term " “cigarettes” " in the description of wares. It recognizes that the term " “including” " in relation to a list of wares can encompass items not specifically set out in the list. Nevertheless, in addition to the Respondent’s argument, the context affects the

interpretation of the description of the wares. [ 66 ] Cigars, cigarillos and tobacco for roll your own cigarettes are smoking wares specified as included in the description of manufactured tobacco. These wares are similar to cigarettes. Indeed, the Applicant provided new evidence on the appeal consisting of photographs of cigarillos and cigars sold in Canada in packaging similar in shape and format to the parties’ cigarette packaging. [ 67 ] In addition, the Applicant somewhat euphemistically describes cigarettes as the " “most common” " tobacco product.

In fact, they constitute the overwhelming most purchased tobacco product dwarfing all other tobacco products in sales and value.

It is beyond the ken of reasonability therefore, that the Respondent would include cigars, cigarillos and tobacco for roll your own cigarettes in its list of tobacco products, without the intention of purposely not including " “cigarettes” " by the term’s absence from a list of similar wares of relative insignificance to that of cigarettes. [ 68 ] In arriving at this conclusion, the Court also rejects the Applicant’s argument at paragraph 124 of its memorandum that " “the participants were shown and asked about an unidentified packaging of ‘tobacco product’, not a pack of cigarettes”.

" In making this argument, the Applicant attempts to rely upon the submission that " “while the shape of the packaging did resemble that of the parties’ respective cigarette packages, the additional evidence of Ms. Luong shows that similar shapes are also used for cigarillos and small cigars, which are manufactured tobacco products, and other than cigarettes” " . The Applicant advances this argument despite the fact that Mr.

Mills acknowledges that the packaging and placement of the design elements was based upon the actual Philip Morris rooftop packs of cigarettes (Respondent’s Memo, para 100). [ 69 ] There is no indication that the cigarillo or cigar packaging provided in the Applicant’s new evidence is a product of either the Applicant or the Respondent. None of the packages had any semblance of the rooftop design apparent on them.

The Applicant did not provide evidence that cigars or cigarillos were products made by the parties, or that there could be some basis in confusion as to the source of these products emanating from their production and sale by either party. [ 70 ] There was also no evidence before the Court that any of the participants smoked cigarillos or small cigars. Without such evidence, the Court concludes that the vast majority of persons questioned on the nature of the so-called " “tobacco products” " would only be smokers of cigarettes. The Court could not imagine that this information was not available to Mr.

Mills. [ 71 ] The Court also concludes that any inherent distinctiveness of the Respondent’s marks was based upon the marketing and sale of its extremely successful Marlboro cigarette product alone.

It is the Applicant which describes the Respondent’s " “iconic” " international Rooftop Design associated with the MARLBORO mark as being created almost entirely on its success in marketing the packaging design and sale of cigarettes. [ 72 ] Indeed, if the Court did not conclude that nearly all of the survey participants would have interpreted " “tobacco products” " to mean cigarettes when presented with images of the packaged product, its main concern would be that the question was too vague with respect to wares. If this were not the case, I might agree with the Respondent that Mr.

Mills ought to have posed the question using the trade- mark alone without any packaging getup so as to apply to all the tobacco product wares. This would include tobacco products for roll

your own cigarettes, pipe tobacco, chewing tobacco, snuff tobacco and tobacco substitutes. [ 73 ] Too generic a question in respect of wares without giving context to what is meant by tobacco products is just as irrelevant as too narrow a question without any context of wares.

It is the packaging that gives context to the question as applying to cigarette wares. [ 74 ] The Court finds the Applicant’s argument with respect to Decision 1 including non-cigarette wares to be without merit. [ 75 ] On this basis, the Mills Survey evidence, although new, is inadmissible with respect to the two marks in this Decision because the Mills Survey did not seek responses on any wares other than that of cigarettes.

These comments similarly apply with respect to any non- cigarette wares in the remaining Decisions (TM Applications 1,478,480, 1,478,483, 1,478,470, 1,478,474, 1,478,481, 1,478,472, 1,478,482, 1,424,232, 1,424,347, 1,420,345) to which the Mills Survey has no application.

(2) DECISIONS 2 and 3: 1,335,783 (Wide Open Flavour & Design) 1,460,679 (Three Line & Roof Design) 1,460,681 (Two Line & Roof Design) [ 76 ] The Applicant has acknowledged that the Mills Survey did not specifically test TM Application 1,335,783 in Decision 2 which includes the words " “WIDE OPEN FLAVOUR” " as part of the survey. The survey evidence is therefore not admissible as new evidence with respect to TM Application 1,335,783 in Decision 2. [ 77 ] The design marks in TM Applications 1,460,679 and 1,460,681 in Decision 3 apply to cigarettes.

The Applicant argues that the Mills Survey applies to any proposed mark that contains the Rooftop Design Mark, such as these two marks as shown accompanied by three and two sidebar lines. The Court disagrees with this submission. As stated above in reference to the Mattel decision, the trade-mark used in the survey must be " “precisely” " the trade-mark applied for. These two design marks are not precisely, or even substantially similar to the marks used in the survey.

The mere fact that each of the design marks at issue comprises a Rooftop Design Mark (or some variation thereof) does not mean that the other design elements in those trade-marks can simply be ignored. [ 78 ] In addition, Mr. Mills admitted in cross-examination that he failed to actually test these trade-marks, as follows: Q. The study you did does not allow conclusions to be drawn about those marks you didn’t test, though? A. Well, I don’t know if I can answer that question. I know that we didn’t test them. Q. So if you didn’t test them you can’t draw conclusions about them? A.

Personally I could not, no. [ 79 ] Accordingly, the Mills Survey evidence is not admissible with respect to the marks in TM Applications 1,460,679 and 1,460,681 in Decision 3 .

(3) DECISION 4: 1,460,676 (Roof Design) 1,478,480 (Roof Device (Bottom Panel)) 1,478,483 (Roof Design (w/ Lozenge)) 1,478,470 (Roof Design (w/ Empty Roof Line)) (

a) Should the conclusions of the Mills Survey also apply to TMA 1,460,676 (Roof Design) and TMA 1,478,470 (Roof Design w/ Empty Roof Line)? [ 80 ] PM acknowledges that the Mills Survey " “purports” " to test consumer perceptions with respect to TM Applications 1,478,480 (Roof Device Bottom Panel) and 1,478,483 (Roof Design with Lozenge), the two middle designs in Decision 4 above (both applying to cigarettes). PM also accepts that solid black (Roof Design) mark in TM Application 1,299,494 in Decision 1 was similarly the subject of the Mills Survey.

Although not applying to cigarettes, this mark is relevant to these discussions as a benchmark to compare its similarity with that of TM Application 1,460,676 (Roof Design), being the left-hand mark above.

The Court is satisfied that those three marks identified by PM were indeed the subject matter of the Mills Survey. [ 81 ] There is no acknowledgment by PM that the survey findings of TM Application 1,299,494 (Roof Design) in Decision 1, in relation to non-cigarette wares, would also apply to TM Application 1,460,676 (Roof Design) in Decision 4, despite their similarity apart from one being black and the other gray coloured.

Similarly, there is no acknowledgment by PM that 1,478,470 (Roof Design with Empty Roof Line), the right-hand mark above, is also similar to TM Application 1,299,494 (Roof Design) in Decision 1, despite their similarity apart from the empty roof line and the lack of color. [ 82 ] Because of the close resemblance between these latter two marks and the Roof Design in the survey which was identical to the

Roof Design in Decision 1, and has been acknowledged as so, the Court is required to consider the Applicant’s argument that the results for the Mills Survey should apply as representative of the other two Roof design marks in Decision 4.

Another way to frame this issue is whether the meaning given to a " “precise” " resemblance as the term is used in Mattel , should be broad enough to capture the latter two Roof Design applications in Decision 4? • TM Application 1,460,676 (Roof Design) [ 83 ] The Court understands the fact that TM Application 1,460,676 is shaded in gray (described as “ Polygons with dark surfaces or parts of surfaces” ), while the Mills surveyed mark is black.

This should not be a factor of distinction with respect to the visual similarity of the two marks. [ 84 ] The Court understands that on the basis of the color norms applying to TM Applications, any color can be applied to either mark. The Mills Survey presented the design marks in black. The Court finds that the question in the survey focused on shape of the marks rather than colour, particularly as it appears that red is the predominant color applied by the Respondent and its affiliates to the Rooftop Design Mark.

The survey therefore appears to be intended to apply to all similar marks of whatever colour. [ 85 ] The Court concludes that the Mills Survey testing for the package bearing the Rooftop Design Marks is precisely the same as the design in the TM Application 1,460,676 in Decision 4. • TM Application 1,478,470 (Roof Design with Empty Roof Line) [ 86 ] In this debate regarding the precision of marks, the Roof Design with Empty Roof Line mark differs only from the Mills Survey Roof Design mark by the additional line added to the bottom peak of the design.

It stretches the ordinary meaning of " “precise” " used in Mattel , i.e. defined in the dictionary to mean " “minutely exact” " , to conclude that the mark in TM Application 1,478,470 is precisely the same as the Mills Survey Roof Design mark. [ 87 ] In examining this issue, the Court notes that the Mattel decision referred to the term " “precisely” " in a fact scenario concerning confusing word marks, BARBIE and BARBIE’S, not design marks as in this case. [ 88 ] Moreover, the issue in Mattel was not about confusion over the word marks, but the questions posed that formed the basis of the new survey evidence introduced on the judicial review of the Registrar’s decision.

In other words, the comments in Mattel on the trade- mark used in the survey being required to be " “precisely” " that applied-for did not consider the degree of precision required for confusing design marks. [ 89 ] The Supreme Court in Mattel cited the Registrar’s decision in Canada Post Corp. v. Mail Boxes Etc. USA Inc. , 77 C.P.R. (3d) 93 [ Canada Post ] for the proposition that the trade-mark used in the survey must be precisely the trade-mark applied for . However, Canada Post involved a confusing word mark.

The extra added word was found to be a dominant element that would have skewed answers to the survey as described and commented on as follows at paras 13 and 18: [13] The opponent has also submitted the Humphreys affidavit which evidences a survey conducted by Ms. Humphreys' company respecting consumer reaction to a trade-mark similar to the applied for mark but which includes the word CANADA after the words MAIL BOXES ETC. […] [18] However, I do not consider that the survey results are equally applicable to the trademark which is the subject of the present application.

I agree with the applicant's submission that the fact that the trade-mark which was the subject of the survey contained the word CANADA as a dominant element would have skewed the respondents' responses toward Canada Post or the Canadian postal system. Thus, I must give significantly diminished weight to the survey results in the present case. [Emphasis added.] [ 90 ] It is well understood that the ratio decidendi of a case is based upon its facts.

While obiter dictum may have persuasive value, and recognizing that comments coming from the Supreme Court of Canada must be credited with the most persuasive value possible, the Court is nevertheless of the view that a factual context may give rise to reasonable distinctions or meanings not considered, even by the Supreme Court.

On this point, it appears reasonable to consider whether there is significant distinction in the precision required to compare word marks as opposed to design marks, such that " “precisely” " may be given a wider berth in the comparison of designs. [ 91 ] Words carry meaning, in most cases very precise, with ambiguity arising from the different

definitions of the words. In other words the confusion arises not entirely from the visual impact, but because humans think in words making the connexion in meaning almost direct. The word " “Canada” " in the Canada Post decision, introduced an entirely new concept and meaning into the word mark in question. On the other hand, design marks (not containing words or an abbreviation for a word or phrase) rely entirely upon a visual impact that may convey confusion, in this case by vague impressions conjured up by an association with an absent word mark.

There is no one to one relationship on the impact to meaning by adding a word as opposed to a single line to a mark. This is evident simply by comparing the marks considered relating to Decision 3 which involves marks with two lines versus three lines, as opposed to that involving lines and words.

The point is that whether the design mark has an extra line on the bottom rooftop, should have little effect when comparing two Roof Design marks. [ 92 ] Additionally, the Court is of the view that when deciding whether distinctions between design marks are sufficiently " “precise” " , reference should be made to the test of confusion of marks.

It measures confusion in " “the mind of a casual consumer somewhat in a hurry who sees the mark at a time when he or she has no more than an imperfect recollection of the [prior] trade-marks and does not pause to give the matter any detailed consideration or scrutiny, nor to examine closely the similarities and differences between the

marks”" (Masterpiece at para 40 citing Veuve Clicquot Ponsardin c.

Boutiques Cliquot Ltée, 2006 SCC 23 [Veuve Clicquot] at para 20).With so little difference in the marks, it is hard to see how the casual observer could make such a distinction in a design mark based uponthe addition of a fairly non-descript additional line on a rooftop peak. [93] In conclusion, the Court finds the variance in the addition of a single line to the Roof Design mark to be a "“distinction without adifference”" when the comparison is made between the mark used in the Mills Survey and TM Application 1,478,470 (Roof Design withEmpty Roof Line) on the basis of a single repeated line: Promafil Canada Ltee v.

Munsingwear Inc., (FCA), [1992]F.C.J. No. 611(FCA) at para 15. In the Court’s view, the latter mark is sufficiently "“precise”" as to give rise to the same confusion asthe Roof Design mark used in the Mills Survey to be a basis for a finding of confusion with the Applicant’s MARLBORO word mark.Accordingly, TM Application 1,478,470 (Roof Design with Empty Roof Line) in Decision 4 constitutes a fourth design mark to whichthe conclusions of the Mills Survey should apply. (

b) Did the Mills Survey questionnaire ask the wrong question? • Requirements that the questions provide a concrete context without being speculative or suggesting an answer [94] PM submits that the Mills Survey tests consumer reactions to a cigarette packaging configuration not at issue in these proceedings,rather than testing reactions to the trade-marks that are at issue.

It argues that by asking survey participants the wrong questions, anyconclusions about consumer perceptions that could be drawn from the results of the Mills Survey are irrelevant to the issues on appeal. [95] In considering whether the survey evidence is new, PM points out that the principal problem with the Chakrapani Surveyconsidered by the Registrar was that the survey did not actually test consumer reactions to the design marks at issue.

Rather it testedreactions to a particular variant of the no-name package at issue in the Federal Courts decisions which was itself a composite of variousother word and design elements arranged in a particular three-dimensional configuration.

By repeating the same error, PM argues thatITL has not provided new relevant evidence for the Court’s review. [96] As described above, the tested trade-marks in the Mills Survey were displayed to survey participants, although describedgenerically as "“tobacco products”", on what the Court concludes in the eyes of the participants were three-dimensional cigarettepackages on a two-dimensional surface.

PM argues that the participants should have been simply provided with the trade-marks in a two-dimensional fashion, as actually applied for, as the marks alone on a sheet of paper without any other visual context. [97] Moreover, from admissions during questioning of Mr.

Mills on his affidavit, PM argues that the express purpose of the three-dimensional packaging was to mimic the no-name package, which was previously found to be infringing in Philip Morris 2012.Accordingly, the Mills Survey was prepared with the view to suggesting a cigarette packaging configuration already found to beinfringing. [98] The Court finds PM’s submissions raise issues that set the directions from the Mattel decision somewhat in opposition to eachother.

On the one hand, survey questions should not suggest answers, while on the other the survey should provide sufficient context forconsumers to be able to answer the questions intelligibly. [99] In this latter regard, the Court views the Federal Court decision in Mattel to be instructive. I refer to its requirement that surveyquestions be sought in manner that provides context to the participants, i.e. in this matter providing a context that the marks relate tocigarettes, while not either engendering speculation or suggesting responses. [100] Mr.

Justice Rouleau in Mattel U.S.A Inc. v. 3894207 Canada Inc., 2004 FC 361 (affirmed on appeal to the SCC), declined to admitsurvey evidence because the survey excluded knowledgeable participants, which he found was an error in principle. However, inparagraphs 31 to 33 of the reasons, Rouleau J. described the conflicting parameters of the necessity to provide "“concrete context”",while not suggesting the answers or leading participants into areas of speculation, as follows: [31] The applicant alleges that Mr.

Thexton's choice to exclude those who were familiar with the respondent's restaurants aspotential participants for the survey was based on the fact that the test for confusion is one of “first impression”. Thisreasoning is baffling. In fact, the application of the ‘first impression” test does not require that the consumer be unaware ofthe context and the exact nature of the wares, services or business in association with which the trade-mark at issue is used.This would amount to disregarding the existence of paragraph 6(5)(

c) of the Act. [32] Quite the contrary, it is precisely the participants who were familiar with the services of the respondent's restaurant andbusiness who would have been in the best position to enlighten us about the likelihood of confusion between the two trade-marks. In Walt Disney Productions v. Fantasyland Hotel Inc. [1994] A.J.

No. 484, this is precisely what led the AlbertaQueen's Bench to disregard a survey carried out, in part, on people who had never visited the defendant's establishment (ahotel), when trying to determine if there was a risk of confusion between the name of this hotel and the plaintiff's trade-mark. At paragraph 271 of his judgement, Rooke J. stated: Furthermore, if they do not know of the Hotel they fall outside the requirement of being "prospective customers" within thetest set out in the Warnink "Advocaat" case (supra).

Rather, it is an ordinary person, making an uninformed guess on noinformation about the Hotel operated by the Defendant.

I note that the Safeway case held (headnote and at 1317-8) thatevidence of confusion by persons only casually acquainted with the business is entitled to little weight in determining thelikelihood of confusion, and that the target universe must consist of people most likely to purchase the defendant's products -in this case, I would take it to mean people who know of the defendant's business. [33] It is therefore fundamental to understand that such surveys, to have probative value, cannot take place in a vacuum, aswas the case here.

It is not sufficient to ask abstract questions without revealing the concrete context underlying the issues.Furthermore, many questions included in the survey are suggestive. As Rooke J. pointed out in Walt Disney Productions,supra, the question “should not direct the person answering the question into a field of speculation upon which that person

would never have embarked had the question not been put” (paragraph 271). [Emphasis added.] [ 101 ] Justice Rouleau’s directions might be summarized as follows: 1 . the target participants should be persons most likely to have knowledge of the wares or services and how they are employed; 2 . the survey questions cannot be abstract or take place in a vacuum, but must provide concrete context that relate to the underlying issues; 3 . the survey questions should not direct persons into a field of speculation; and 4 . the survey questions must nevertheless not be suggestive of an answer. [ 102 ] The most salient issue raised by PM is whether showing the marks on a three-dimensional no-name getup of cigarette packages infringes these rules.

Although there may be concerns in favour of PM’s submissions because a packaging getup has been adopted, in the Court’s view the Mills Survey evidence does not ask the wrong question. • Survey questions are not suggestive or speculative by presenting the design marks in a no-name getup as opposed to the marks as they are sought to be registered [ 103 ] Before describing why the Court rejects PM’s argument that the wrong question has been asked, it will review a number of other factors in support of admitting the Mills Survey.

First, it is not seriously contested that the Mills Survey evidence is intended to " “fill the gap” " by responding to the deficiencies identified by the Registrar in the Chakrapani Survey evidence . The Registrar adopted the conclusions of the Federal Court of Appeal that distinguished between providing evidence of confusion associated with a combination of elements on a cigarette package and that arising from a single mark on the same package.

This w as described by Justice Gauthier at paragraph 108 of Philip Morris 2012 , as follows: [108] In any event, the surveys filed in evidence were not specifically designed to test whether consumers associated the name Marlboro to cigarette packages bearing only each individual ROOFTOP design mark essentially as registered.

What was presented to the participants in the survey by both parties’ experts was the no-name package as a whole. [Emphasis added.] [ 104 ] Premised on this conclusion, the Registrar specifically refused to adopt the Applicant’s argument that the Chakrapani Survey demonstrated that confusion would arise for the Rooftop Design Mark in isolation, as described at paragraph 78 of Decision 4 as follows: [78] … Nonetheless, the fact remains that it was the particular combination of elements used on the no-name packaging alone that was found confusing by Madam Justice Gauthier.

Similar to the present cases, there was no evidence pointing to the fact that each of the individual ROOFTOP design marks was, by itself, confusing with the trade-mark MARLBORO. [Emphasis added.] [ 105 ] Mr. Mills posed a survey question that was intended to remedy the limitations of the Chakrapani Survey introduced in the Philip Morris 2012 .

By extracting the Rooftop Design Mark from the remaining elements of the no-name packaging " “as a whole” " , it was intended to present the " “individual applied-for marks in isolation” " to participants to determine whether standing alone they gave rise to confusion with the MARLBORO word mark. Mr. Mills concluded that presenting only the individual rooftop design element on a package of no-name cigarettes avoided any argument that it was a combination of trade-mark elements as meant by the Court of Appeal in Philip Morris 2012 . [ 106 ] Secondly, the Court rejects the conclusions of Dr.

Corbin that the results of the survey do not support a finding of likelihood of confusion because " “asking people what brand they ‘associate’ with the design is not determinative of confusion”. " The Court agrees with Mr. Mills’ statement that the findings of the survey support a finding that a significant percentage of daily adult smokers identify each of the three TM Application design marks tested with the MARLBORO word mark. In the Court’s view, this indeed substantiates a finding of confusion as to source of the wares.

The Court notes in addition that this issue was not pursued in the Respondent’s memorandum, or to any extent in its oral argument. [ 107 ] Third, the Court also agrees that the questions provided a concrete context by the TM Applications being pictured on a no-name cigarette package. There is no issue as to speculation as to their use on other wares. However, as indicated and to the opposite effect, this format undermines the Applicant’s argument that the survey evidence is relevant to the marks for non-cigarette wares.

It would be speculative to apply the survey conclusions beyond the concrete depiction of the use of the cigarette package design presented. [ 108 ] In addition, Mr.

Mills deposed that the packaging getup used in the survey was modeled on the Respondents’ actual 2012 redesigned packaging (but importantly that includes the " “ROOFTOP” " word mark) of its rooftop packs of cigarette launched after the Philip Morris 2012 decision, as described at paragraph 19 of his sworn statement: …The design elements were placed on the three-dimensional tobacco product packages in the exact same location as on the actual Philip Morris ROOFTOP packs of cigarettes. [ 109 ] On another point, it is not contested that the findings are statistically sufficiently significant to demonstrate confusion with respect to the three surveyed marks as falling within acceptable ranges.

Satisfactory statistical survey evidence was recently discussed by Justice Boswell in Diageo Canada Inc. v. Heaven Hill Distilleries, Inc., 2017 FC 571 :

[94]…Indeed, the rate of confusion Courts have previously found to be sufficient to establish a likelihood of confusion ranges from: 4.8% to 8.2% in Walt Disney Productions v Triple Five Corp , 1994 ABCA 120 , [1994] AJ No 196 at para 44 , 53 CPR (3d) 129 (Alta CA); 11% in New Balance Athletic Shoes Inc v Matthews, [1992] TMOB No 358, 45 CPR (3d) 140 (TMOB) ; and 13.5% in Kirkbi AG v Ritvik Holdings Inc , 2002 FCT 585 at para 128 , 20 CPR (4th) 224, aff'd without discussion of this issue, 2003 FCA 297 , [2003] FCJ No 1112, aff'd 2005 SCC 65 , [2005] 3 SCR 302. [ 110 ] Despite the foregoing conclusions, PM nevertheless contends, using Justice Gauthier’s words in Philip Morris 2012 , that the survey was a presentation not of the " “cigarette packages bearing only each individual Rooftop Design Mark [TM Applications] essentially as registered, … [but] the no-name package as a whole” " .

In other words, it submits that the Applicant has not overcome the distinction that Justice Gauthier made in Philip Morris 2012 between the mark as registered and its combination with other elements of the packaging getup. [ 111 ] As best the Court can frame PM’s submission, it would appear to argue that a significant element relating to confusion is that the package contains no distinguishing word mark. The presence of the word mark is admitted by both parties as being the key element affecting the confusion between PM’s Rooftop Design Mark and the MARLBORO word mark.

Conversely, its absence must therefore be a key contributor to confusion. The confusion does not arise from the Rooftop Design Mark itself, but rather the circumstances of its presentation — it depends on packaging. [ 112 ] However, these are not the same issues as were considered in Philip Morris 2012 . The Court recognizes the conundrum the no- name circumstance presents to the Applicant in this matter. Ultimately, it only seeks to demonstrate that the design mark is confusing when applied on cigarette packaging without an accompanying word mark. It was required to convey this context in its questioning.

True, there was another option. If ITL wished to test the confusion of the mark in a no-name circumstance, it could have specified those circumstances by stating to the participant that the design mark would be placed on a package of cigarettes unaccompanied by any other word mark.

The resolution of the issue seems to come down to whether by using the visual packaging format to express the no-name circumstance, this contributed to the confusion, or as PM argues that it suggested a response? [ 113 ] A picture is capable of conveying more information than words, but I do not find that by using packaging, ITL did anything more than concretizing the proposition that it wanted smokers to consider; namely " “was the Rooftop Design Mark on a no-name package confusing with its MARLBORO mark?” " In other words, coming back to the question asked in Philip Morris 2012 at paragraph 103, I would reframe the issue in the form of a question in these circumstances: whether the packaging getup is conveying a different message to consumers than the TM Applications themselves when the question relies on cigarette packaging without any accompanying word mark.

The answer is " “No” " . The question is simply posed more clearly when using the no-name packaging format. [ 114 ] In

summary, the Court finds that the survey questionnaire based upon the marks being shown on a three-dimensional tobacco product, even though similar to the actual Phillip Morris Redesigned Rooftop packs of cigarettes, not to be suggestive and not asking the wrong question, as this issue is framed by the Respondent. The survey meets the requirements of relevancy and reliability sufficient for it to be admitted as new evidence in this matter. [ 115 ] The Mills Survey evidence is therefore admissible with respect to the four TM Applications in Decision 4, but not in respect of any other Decision. (

c) The Admissibility of the Applicant’s remaining new evidence [ 116 ] The interpretive evidence from dictionaries etc. has already been considered, while that regarding the packaging of cigars and cigarillos was noted and not found probative for its purpose of comprising other tobacco products as being the subject matter of the Mills Survey. [ 117 ] The evidence pertaining to the relaunched Redesigned Rooftop product, including the pleadings, the photographs and promotional material that were attached as exhibits to Ms.

Luong’s affidavit [subparagraphs a, b and c at paragraph 51 of the Applicant’s Memorandum] is not considered relevant except for the purpose of demonstrating that the packaging used by Mr. Mills was based upon its getup, without the word mark. [ 118 ] The remaining category of additional new evidence consists of the Respondent’s international Marlboro packaging and printouts of articles and photographs pertaining to the international Marlboro product , which was attached as an exhibit to Ms. Luong’s affidavit [subparagraphs d, e and f].

In light of the consideration of the Mills Survey evidence, this supplemental evidence is relevant to the Applicant’s arguments on confusion and whether the opposition proceedings are grounded on speculation, all as more particularly discussed below when analyzing the evidence.

(4) DECISIONS 5 and 6:

1,478,474 (Roof Device (w/ Square Pattern)) 1,478,481 (Roof Device (w/ T Pattern) ) 1,478,472 (Roof Device (w/ Triangle Pattern)) 1,478,482 (Roof Device (w/ Lozenge & Triangle Pattern) [ 119 ] TM Application 1,478,472 in Decision 6 has already been identified as not applying to cigarettes. The Mills Survey is not admissible therefore with respect to this design mark. [ 120 ] While TM Applications 1,478,474 and 1,478,481 in Decision 5 and TM Application 1,478,482 in Decision 6 apply to cigarettes, they are not precisely, or even substantially similar to the marks used in the Mills Survey.

No survey questions were asked with respect to these marks. The Court’s remarks relating to the inadmissibility of the survey evidence with respect to Decision 3 apply mutatis mutandis to these Decisions. The survey evidence is also not admissible therefore, with respect to TM Applications 1,478,474 and 1,478,481 in Decision 5, and TM Application 1,478,482 in Decision 6.

(5) DECISION 7, 8 and 9: 1,424,232 (BLACK MENTHOL & Design) 1,424,347 (BLACK MENTHOL & Design) 1,420,961 (Roof Device Finger Print Label) 1,420,345 (Roof Device Finger Print Label) [ 121 ] The Applicant has acknowledged that the Mills Survey did not specifically test TM Applications 1,424,232 or 1,424,347 of Decisions 7 and 8 respectively that include the words " “BLACK MENTHOL” " (Applicant’s Memorandum, para 66).

The survey evidence is therefore not admissible as new evidence with respect to Decisions 7 and 8. [ 122 ] While the remaining two marks in TM Applications 1,420,961 and 1,420,345 in Decision 9 apply to cigarettes, they are not precisely, or even substantially similar to the marks used in the Mills Survey. No survey questions were asked with respect to these Applications. The Court’s remarks relating to the inadmissibility of the survey evidence with respect to the preceding Decisions apply mutatis mutandis to these Decisions concerning cigarettes.

The Mills Survey evidence is not admissible therefore, with respect to TM Applications 1,420,961 and 1,420,345 in Decision 9. C.

Is there likelihood of confusion between each of the applied-for design marks in the Decision 4 and the registered trade-mark MARLBORO? [ 123 ] This Court must now determine anew whether there is a likelihood of confusion between each of the four applied-for design marks in Decision 4 and ITL’s registered trade-mark MARLBORO in light of the entire evidentiary record (including the additional evidence). [ 124 ] ITL raised a number of grounds of opposition before the Registrar including unfairly competing and intentionally creating confusion contrary to paragraph 30(

i) of the Act , issues of priority of filing dates and confusion pursuant to paragraphs 16(1) (

a) and (

c) and paragraphs 16(3) (

a) and (

c) of the Act, in addition to claiming that PM’s marks are non-distinctive. [ 125 ] Ultimately however, the parties focused their submissions on the determinative issue with respect to the non-registrability of the marks pursuant to paragraph 12(1) (

d) of the Act , namely that the marks are confusing with the Applicant’s registered trade-mark MARLBORO. This in turn led to the parties to narrow their submissions primarily to the test for confusion pursuant to subsection 6(5) of the Act.

(1) Overall consideration of subsection 6(5) factors and all surrounding circumstances [ 126 ] In determining whether th

[…]