England and Wales Court of Appeal (Civil Division) Decisions

IN THE COURT OF APPEAL (CIVIL DIVISION)
ON APPEAL FROM THE HIGH COURT OF JUSTICE, BUSINESS AND PROPERTY COURTS OF ENGLAND AND WALES, INTELLECTUAL PROPERTY LIST (ChD), INTELLECTUAL PROPERTY ENTERPRISE COURT
Nicholas Caddick KC sitting as a Deputy High Court Judge
[2022] EWHC 941 (IPEC)

		Royal Courts of Justice Strand, London, WC2A 2LL
		27 April 2023

Lord Justice Arnold:

Introduction

1. This is a trade mark dispute about the use of the words MATCH and MUZMATCH in relation to online introduction and dating services. The Claimants are members of a group of companies which provide such services worldwide, including in the UK, to consumers of all religions and none. The Claimants are the proprietors of two European Union trade mark registrations for MATCH.COM and a UK trade mark registration for "match" in a lower case font together with a small heart device ("the Trade Marks"). The First Defendant ("Muzmatch") provides online introduction (or "matchmaking") services to Muslims in the UK. The Second Defendant ("Mr Younas") is the founder of Muzmatch's business and its chief executive officer. The Claimants contend that Muzmatch has infringed the Trade Marks and committed passing off through the use of the word MUZMATCH, various logos incorporating the word MUZMATCH and certain phrases that included the word "match" which were used by Muzmatch for search engine optimisation ("SEO") purposes; and that Mr Younas is jointly liable for such acts. The Defendants deny infringement and passing off, but have not challenged the validity of the Trade Marks.
2. In addition to the claims outlined above, the Claimants claimed that the Defendants had infringed an EU trade mark registration for the word TINDER and a UK trade mark registration for a slightly stylised version of that word ("the Tinder Marks") through the use of phrases including the word "tinder" for SEO purposes. The Defendants submitted to judgment in respect of that claim. The Claimants nevertheless seek to rely upon those acts as being relevant to the claims in dispute.
3. Nicholas Caddick KC sitting as a Deputy High Court Judge in the Intellectual Property Enterprise Court held for the reasons given in a careful and detailed judgment dated 20 April 2022 [2022] EWHC 941 (IPEC) that the Defendants had infringed the Trade Marks and committed passing off. I granted the Defendants permission to appeal on certain grounds, but not others.

The Trade Marks

4. The Trade Marks are as follows:

i) EU Trade Mark No. 182253 MATCH.COM registered as of 1 April 1996 in respect of "information and consultancy services in the nature and field of on-line dating and introduction services" in Class 42.

ii) EU Trade Mark No. 16246639 MATCH.COM registered as of 13 January 2017 in respect of inter alia "downloadable software in the nature of a mobile application for internet-based dating and introduction" in Class 9, "providing a website featuring technology in the field of social media, namely, a website that enables users to send status updates to subscribers of web feeds, upload and download electronic files to share with others" in Class 42 and "dating services; internet based social networking, introduction and dating services" in Class 45.

iii) UK Trade Mark No. 3097217 shown below registered as of 3 March 2015 in respect of inter alia "providing social introduction and date-arranging services" and "dating agency services; matchmaking services; computer dating services; provision of dating agency services via the Internet" in Class 45.



5. The present claims were initiated by a claim form issued on 30 July 2020, after the United Kingdom had ceased to be a Member State of the European Union, but before that withdrawal became fully effective on 31 December 2020. Since 31 December 2020, the two EU Trade Marks have been replaced, so far as the UK is concerned, by two comparable UK registrations. As explained in more detail below, the infringing acts complained of commenced long before 31 December 2020 and continued afterwards. Neither side suggests that either the change in the registrations which occurred on that date, or the consequential change in the legal basis for the Claimants' infringement claims in so far as based upon the EU registrations, is material to the issues before the judge or on appeal. Accordingly, this can be ignored.

The factual background

6. The factual background to the dispute may be summarised as follows.
7. The Claimants' predecessor in business started providing online introduction and dating services in the USA via a website located at www.match.com in about 1995. These services were (and remain) principally aimed at people who are looking to find another person with whom they can build a long-term relationship, including marriage. The services were launched in the UK in 2001. The European, including UK, business was acquired by the Second Claimant in 2009. The Claimants created a Twitter handle in January 2009, a YouTube channel in December 2009 and a Facebook page in March 2010. In 2010 the Claimants launched an alternative service known as Match Affinity. In July 2011 the Claimants launched an app, that is to say, software downloadable to a mobile phone or device to enable users to access the Claimants' services via mobile telecommunications networks connected to the internet. In July 2017 the Second Claimant acquired Tinder, which remains a separate business.
8. The judge described the evolution of the logos used by the Claimants on their website and app in his judgment at [16]-[23]. The Claimants adopted the device which forms the UK Trade Mark as a logo on the website in 2015. Throughout the Claimants' trading in the UK, the branding of their services generally involved some level of emphasis on the MATCH element of the Trade Marks by way of placement, size, colour, or emboldening of the word MATCH or by it being the sole word used to identify the Claimants as the source of the services.
9. The Claimants advertised and promoted their services by television advertisements, radio advertisements, newspaper advertisements, posters, magazine inserts, PR, social media and online advertising. The Claimants spent substantial sums on these efforts, for example just under £5 million on TV advertising alone in 2010. The Claimants succeeded in creating a high level of awareness for the MATCH brand. For example, a brand monitoring survey commissioned from the well-known market research company TNS found that in 2009 "Match.com" had a "top of mind" awareness score of 28%, a spontaneous awareness of 37% and an aided awareness of 65%, far ahead of any of its competitors. Later years' reports referred to the brand as "Match" interchangeably with "Match.com". By 2014 the figures had increased to 47%, 71% and 87%. There were 342,086 UK visitors to the Claimants' website in May 2011, and the Claimants had around 129,000 subscribers and sales of more than €29m in 2011. By the end of January 2012 the YouTube channel had had over 500,000 views.
10. In around April or May 2011 Mr Younas set up a business under the name Muzmatch with the aim of providing online matchmaking services to the Muslim community in the UK.  Mr Younas was (and remains) a practising Muslim. He wanted to provide a service that would allow Muslims to find a marriage partner in a way that was compatible with Islamic values. His concern was that, for many Muslims, the only way to meet a marriage partner was through family contacts or intermediaries, a process which many Muslims found awkward and outdated. Whilst introduction websites were becoming popular with some Muslims, many other Muslims disapproved of them, and in particular of the more "mainstream Western" websites, because they were perceived to promote casual relationships rather than focussing on finding a marriage partner. As a result, Mr Younas decided to set up Muzmatch to provide Muslims with the opportunity to find marriage partners using the more modern means of a website, but one which specifically sought to be compatible with Islamic values.
11. To this end, Mr Younas created a pay-as-you-go website with features designed to weed out users who were not serious in their search for a Muslim marriage partner. He registered "muzmatch.com" as a domain name in April 2011 and launched the website in May 2011. Mr Younas did this in his spare time away from his main job as a Vice President of the Institutional Equities Division of Morgan Stanley. Initially, he operated the business as a sole trader, but at the end of 2014 he left Morgan Stanley and in January 2015 he set up Muzmatch to take over the business. In about March 2015 Muzmatch launched an app. Thereafter the website was retained to promote Muzmatch's services, but visitors to it were directed to the app. The app has various features designed to ensure that users respect Islamic values. Until around March 2017 the app was free to use, but gradually paid-for features were introduced, including a premium monthly subscription service. Muzmatch's focus continued to be to provide a service targeted at Muslims seeking a marriage partner. Thus its app was described on the Apple App Store as "Muzmatch: Single Muslim dating; Halal, Arab & Muslim marriage" and on the Google App Store as "Muzmatch: Muslim & Arab Singles, Marriage & Dating".
12. Mr Younas and Geoffrey Craig, another senior employee of Morgan Stanley who had no role or interest in the Defendants' business, gave evidence as to the choice of the name "Muzmatch". The judge found that it was probably Mr Craig who suggested the name. Mr Younas liked the name because it carried a clear and simple message, namely that the service helped users to find a Muslim match, and was catchy and memorable. As the judge noted, the Defendants' website and app did not use the word "match" descriptively, although the app did use the phrase "it's a muzmatch" when notifying a user about a potential partner.
13. The judge set out in his judgment at [53] a table showing the various logos used by the Defendants from May 2011 to 2022. For present purposes it is sufficient to note the following. Initially the logo featured the word "muzmatch.com" in a lower case font with the addition of a small heart device over the dot (this was before the Claimants adopted such a device). In April 2011 the Defendants dropped the ".com". In 2016, as result of a complaint from the Claimants, they adopted a logo featuring the word "MUZMATCH" in an upper case font preceded by a butterfly device. In January 2020 they adopted a logo featuring the word "muzmatch" in a lower case font with a different butterfly device above it.
14. Prior to 2018 the Defendants promoted the business by business cards and leaflets handed out at mosques and Muslim social events, word of mouth and organic social media content. By February 2015 the business had 8,122 users in the UK and there had been some 1.5 million views of pages from its website. The business grew more rapidly after the launch of the app. It had 7,824 signups in the UK in March 2016, generating some 25.8 million online swipes. By September 2021, those figures had grown to 666,069 signups and an average of 71 million swipes a month. From 2018 the app was promoted by means of paid-for social media content. These included two advertisements featuring plays on the words from well-known songs by Lionel Ritchie (which reached over 100,000 people in February 2018) and Adele. The same advertisements also appeared on the London Underground and on buses in London and Manchester. In addition there were some television advertisements on the Islam Channel and Sky TV and on Muzmatch's YouTube page.
15. In around January 2012 Mr Younas put in place an SEO strategy which involved making a list of some 5000 words or word combinations (referred to as "keywords") that related in some way to Muslim marriage or Muslim matchmaking. A page, known as a landing page, was then created on the Muzmatch website for each of these keywords. These landing pages would be indexed by search engines, such as Google, so that a user who carried out a search using one of those keywords would be presented with search results which included a link which, if clicked on, took the user to the relevant landing page on Muzmatch's website. The aim was to increase visits to the website.
16. Although the word "match" was not itself a keyword, there were some 132 keywords that contained that word, including, for example, "match-muslim", "muslimmatch", "muslim-match" and "UK-muslim-match". The list also included SingleMuslim, Shaadi, SimplyMarry, Salaam Love and Asian Flames, which were the names of other dating or introduction websites.
17.  In around March 2015 Mr Younas revised the list of keywords to include keywords more related to mobile apps such as "muslim-app" and "muslim-mobile". He also added "muslim-tinder", "tinder" and "halal-tinder" as keywords. The Defendants accept that the use of these keywords infringed the Tinder Marks.
18. The judge found that, as a result of the Defendants' SEO strategy, a reasonably substantial number of users who had carried out searches using keywords that included the word "match" had clicked through from links in the search results to the landing pages on the Muzmatch website.
19. Match learned of Muzmatch's activities in around January 2016. On 26 January 2016 its trade mark attorneys wrote to Muzmatch asserting that those activities infringed the 1996 EU Trade Mark and the UK Trade Mark (amongst other marks) and amounted to passing off, and threatening proceedings if suitable undertakings were not given by 15 February 2016. This led to settlement discussions in which the parties sought to agree terms on which, subject to an agreement on branding style, Muzmatch could continue to use the sign MUZMATCH. Although no agreement was reached, the Muzmatch logo was changed as mentioned in paragraph 13 above. These discussions broke down in late 2016, however. Subsequently, from September 2017, the parties were involved in negotiations as to whether the Claimants could buy Muzmatch. In the course of these negotiations, the Claimants made four separate offers, all of which were rejected by Mr Younas, and the negotiations ended in January 2019. In the meantime, the Claimants successfully opposed an application by Muzmatch to register MUZMATCH as an EU trade mark both before the Opposition Division (January 2018) and Board of Appeal (September 2018) of the European Union Intellectual Property Office. The Claimants also successfully opposed an application by Muzmatch to register MUZMATCH as a UK trade mark (February 2020). Nevertheless, the Claimants did not bring proceedings until July 2020.
20. A final but important point of background is that, because the aim is to connect people who do not know each other, users of introduction and dating services have to provide sensitive personal information about themselves and about what they are looking for in a partner. As a result, it is important that users feel that they can trust the platform(s) that they are using. Some users use more than one platform.

The legislative framework

21. The Claimants' claims for infringement of the EU Trade Marks are made under Article 9(1)(b) and (c) of Council Regulation 207/2009/EC of 26 February 2009 on the Community trade mark trade (codified version) and Article 9(2)(b) and (c) of European Parliament and Council Regulation 2017/1001/EU of 14 June 2017 on the European Union trade mark (codification) ("the EUTM Regulation"). There is no material difference between these provisions, and for convenience I shall refer only to the latter.
22. The Claimants' claim for infringement of the UK Trade Mark is made under section 10(2) and (3) of the Trade Marks Act 1994, which implemented successively Article 5(1)(b) and (2) of First Council Directive 89/104/EEC of 21 December 1988 to approximate the laws of the Member States relating to trade marks, Article 5(1)(b) and (2) of European Parliament and Council Directive 2008/95/EC of 22 October 2008 to approximate the laws of the Member States relating to trade marks (codified version) and Article 10(2)(b) and (c) of European Parliament and Council Directive 2015/2436/EU of 16 December 2015 to approximate the laws of the Member States relating to trade marks (recast). (Directive 89/104 is not of temporal application to this case, but is referred to in much of the case law cited below.) There is no material difference between these provisions, and for convenience I shall refer solely to Article 10(2) of Directive 2015/2436, which is mutatis mutandis in the same terms as Article 9(2) of the EUTM Regulation.
23. Article 10(2) of Directive 2015/2436 provides:

"Without prejudice to the rights of proprietors acquired before the filing date or the priority date of the registered trade mark, the proprietor of that registered trade mark shall be entitled to prevent all third parties not having his consent from using in the course of trade, in relation to goods or services, any sign where:

(a) the sign is identical with the trade mark and is used in relation to goods or services which are identical with those for which the trade mark is registered;

(b) the sign is identical with, or similar to, the trade mark and is used in relation to goods or services which are identical with, or similar to, the goods or services for which the trade mark is registered, if there exists a likelihood of confusion on the part of the public; the likelihood of confusion includes the likelihood of association between the sign and the trade mark;

(c) the sign is identical with, or similar to, the trade mark irrespective of whether it is used in relation to goods or services which are identical with, similar to, or not similar to, those for which the trade mark is registered, where the latter has a reputation in the Member State and where use of that sign without due cause takes unfair advantage of, or is detrimental to, the distinctive character or the repute of the trade mark."

24. Although the EUTM Regulation, the Directives and the 1994 Act provide for certain defences to infringement, there is no defence for "honest concurrent use".

The relevant dates

25. It was common ground before the judge that the relevant dates for the assessment of infringement are as follows: March or April 2011 (when Mr Younas first starting using the sign MUZMATCH) and January 2012 (when the use of "match" in the SEO keywords started) for the 1996 EU Trade Mark; March 2015 for the UK Trade Mark; and January 2017 for the 2017 Trade Mark. The judge primarily focussed on March/April 2011, although he was careful also to consider the position at the later dates. A point which the judge did not remark upon is that the Claimants' claims for financial remedies in respect of infringements prior to 30 July 2014 are barred by limitation. Neither side suggests that this affects the infringement analysis.

Infringement under Article 9(2)(b) of the EUTM Regulation/section 10(2) of the 1994 Act

The law

26. In order to establish infringement under Article 9(2)(b) of the EUTM Regulation, Article 10(2)(b) of Directive 2015/2436 and section 10(2) of the 1994 Act, six conditions must be satisfied by the proprietor of a registered trade mark: (i) there must be use of a sign by a third party within the relevant territory; (ii) the use must be in the course of trade; (iii) it must be without the consent of the proprietor of the trade mark; (iv) it must be of a sign which is at least similar to the trade mark; (v) it must be in relation to goods or services which are at least similar to those for which the trade mark is registered; and (vi) it must give rise to a likelihood of confusion on the part of the public
27. The manner in which the requirement of a likelihood of confusion in Article 9(2)(b) of the EUTM Regulation and Article 10(2)(b) of Directive 2015/2436, and the corresponding provisions concerning relative grounds of objection to registration in the Directive and the Regulation, should be interpreted and applied has been considered by the Court of Justice of the European Union in a large number of decisions. In order to try to ensure consistency of decision making, a standard summary of the principles established by these authorities, expressed in terms referable to the registration context, has been adopted in this jurisdiction. The current version of this summary (see e.g. Sazerac Brands LLC v Liverpool Gin Distillery Ltd [2021] EWCA Civ 1207, [2021] ETMR 5 at [8]) is as follows:

"(a) the likelihood of confusion must be appreciated globally, taking account of all relevant factors;

(b) the matter must be judged through the eyes of the average consumer of the goods or services in question, who is deemed to be reasonably well informed and reasonably circumspect and observant, but who rarely has the chance to make direct comparisons between marks and must instead rely upon the imperfect picture of them he has kept in his mind, and whose attention varies according to the category of goods or services in question;

(c) the average consumer normally perceives a mark as a whole and does not proceed to analyse its various details;

(d) the visual, aural and conceptual similarities of the marks must normally be assessed by reference to the overall impressions created by the marks bearing in mind their distinctive and dominant components, but it is only when all other components of a complex mark are negligible that it is permissible to make the comparison solely on the basis of the dominant elements;

(e) nevertheless, the overall impression conveyed to the public by a composite trade mark may, in certain circumstances, be dominated by one or more of its components;

(f) and beyond the usual case, where the overall impression created by a mark depends heavily on the dominant features of the mark, it is quite possible that in a particular case an element corresponding to an earlier trade mark may retain an independent distinctive role in a composite mark, without necessarily constituting a dominant element of that mark;

(g) a lesser degree of similarity between the goods or services may be offset by a greater degree of similarity between the marks, and vice versa;

(h) there is a greater likelihood of confusion where the earlier mark has a highly distinctive character, either per se or because of the use that has been made of it;

(i) mere association, in the strict sense that the later mark brings the earlier mark to mind, is not sufficient;

(j) the reputation of a mark does not give grounds for presuming a likelihood of confusion simply because of a likelihood of association in the strict sense; and

(k) if the association between the marks creates a risk that the public might believe that the respective goods or services come from the same or economically-linked undertakings, there is a likelihood of confusion."

28. The same principles are applicable when considering infringement, although it is necessary for this purpose to consider the actual use of the sign complained of in the context in which the sign has been used: see Specsavers International Healthcare Ltd v Asda Stores Ltd [2012] EWCA Civ 24, [2012] FSR 19 at [45], [87] (Kitchin LJ, as he then was).
29. It is well established that there are two main kinds of confusion which trade mark law aims to protect a trade mark proprietor against. The first, often described as "direct confusion", is where consumers mistake the sign complained of for the trade mark. The second, often described as "indirect confusion", is where the consumers do not mistake the sign for the trade mark, but believe that goods or services denoted by the sign come from the same undertaking as goods or services denoted by the trade mark or from an undertaking which is economically linked to the undertaking responsible for goods or services denoted by the trade mark. I discussed the distinction between the two in Sazerac v Liverpool Gin at [10]-[14].

The judge's assessment

30. The judge set out the applicable legal principles at [79]-[80]. He cited the earlier version of the standard summary set out in Specsavers v Asda at [52]. That will potentially have been to the Defendants' advantage, because paragraph (k) of that version referred to there being a likelihood of confusion "if the association between the marks causes the public to wrongly believe that the respective goods [or services] come from the same or economically-linked undertakings", whereas the version set out above correctly reflects the jurisprudence of the CJEU that it is sufficient "if there is a risk that the public might believe that the respective goods or services come from the same or economically-linked undertakings": see in particular Case C-39/97 Canon KK v Metro-Goldwyn Mayer Inc [1998] ECR I-5507 at [29]. This is a point repeatedly noted by Kitchin LJ in his later judgments in IPC Media Ltd v Media 10 Ltd [2014] EWCA Civ 1439, [2015] FSR 12 at [39], Maier v Asos plc [2015] EWCA Civ 220, [2016] Bus LR 1063 at [76] and Comic Enterprises Ltd v Twentieth Century Fox Film Corp [2016] EWCA Civ 41, [2016] FSR 30 at [32]. In future, decision makers should be careful to use the correct version of the summary.
31. The judge stated at [72] that the test for infringement under both section 10(2) and section 10(3) was to be "conducted through the eyes of a hypothetical person referred to as the average consumer". It is correct that infringement must be assessed from the perspective of the average consumer of the relevant goods or services, who is deemed to be reasonably well-informed and reasonably observant and circumspect; but the notion of "a hypothetical person", while supported by authority, must be treated with care. The point of assessment from the perspective of the average consumer is that one excludes from consideration those who are either ignorant or have specialist knowledge and those who are either careless or excessively careful, but otherwise one takes into account the characteristics of the relevant class of consumers. Such consumers are not an undifferentiated mass, but have the spread of relevant characteristics that human beings have. Moreover, it is sufficient to establish a likelihood of confusion that a significant proportion of the relevant class of consumers is likely to be confused even if many would not be: see Interflora Inc v Marks & Spencer plc [2014] EWCA Civ 1403, [2015] Bus LR 492 at [129] (Kitchin LJ giving the judgment of the Court of Appeal). It is clear from what the judge went on to say at [93(l)] ("a relatively substantial number of users ... are likely to have been confused" by the SEO keywords and landing pages complained of) that the judge understood this.
32. The judge found at [73]-[76] that the average consumer of the services in issue would be a member of the general public who is or may be looking for a partner by means of an online dating or introduction service. The average consumer might be of any race, creed, age or background. The level of attention paid by the average consumer was likely to be higher than for normal consumer items.
33. The Defendants admitted that conditions (i)-(v) set out in paragraph 26 above were satisfied, and in particular that Muzmatch's services were identical to those for which the Trade Marks were registered. So far as condition (iv) was concerned, there was a dispute as to the extent of the similarities between the Trade Marks and the signs complained of. The judge's assessment at [86]-[87] was that, overall, there was a medium level of similarity. The MATCH element in the signs meant that there was a clear visual and aural similarity with the Trade Marks. There was also a clear conceptual similarity: the average consumer would perceive that the word MATCH was being used as a brand, but would also see that it was a reference to the nature of the services (to find a match); the addition of "MUZ" or "muslim" or "UK" would be seen as merely restricting that same concept somewhat (to find a Muslim match in the UK). The presence of the .COM element in the EU Trade Marks made little difference. Nor did the fonts or the heart and butterfly devices.
34. The judge held at [93] that there was a likelihood of confusion as a result of Muzmatch's use of signs comprising the word MUZMATCH for reasons that can be summarised as follows. By April/May 2011 the 1996 EU Trade Mark had acquired a very substantial degree of distinctiveness and reputation in the UK as a badge of origin for the Claimants' services. That distinctiveness and reputation also attached to the UK Trade Mark and the 2017 EU Trade Mark from their filing dates. The average consumer would have seen the word MATCH as the dominant element in each of the Trade Marks. The average consumer would have been well aware that the word MATCH in the context of online dating services was used both descriptively and distinctively, and would be able to distinguish between such uses. The use of MATCH in a distinctive sense other than by the Claimants was not widespread in April 2011. Accordingly, the average consumer would have associated distinctive uses of MATCH with the Claimants' services as at all relevant dates. In the case of the MUZMATCH signs, the use was in a distinctive sense.
35. The judge also found, for similar reasons, that there was a likelihood of confusion as a result of Muzmatch's use of SEO keywords comprising the word "match".
36. The judge summarised his conclusion at [94] as follows:

"Taking the evidence as a whole and, in particular, the evidence of Match's dominant presence in the market as at 2011 (and continuing thereafter), I find that the use of the Muzmatch Signs was likely to lead the average consumer in this case to conclude that Muzmatch was connected in some material way with the business that used the [EU Trade] Marks and, after 3 March 2015, with the business that used the [UK Trade] Mark. In particular, I find that the average consumer would have thought that Muzmatch was a sub-brand of that business specifically targeted at Muslim users and particularly at those Muslim users who felt that the services of a mainstream on-line dating service provider were not in accordance with Islamic values."

37. Although the judge did not use the term "indirect confusion", counsel for the Claimants submitted, and I agree, that this is a finding of a likelihood of indirect confusion, rather than a finding of a likelihood of direct confusion.
38. The judge went on to say at [95] that, in reaching this conclusion, he had taken into account the point made by counsel for the Defendants that Muzmatch had been trading since 2011, and yet, despite "hugely extensive searches by the parties for the purposes of disclosure", the Claimants had only been able to find three instances which they alleged (but the Defendants disputed) were of actual confusion.
39. The judge cited, directly or indirectly, a series of authorities as supporting the proposition that the absence of evidence of actual confusion is "rarely decisive" in trade mark infringement and passing off cases, but this depended on the facts: Electrolux Ltd v Electrix Ltd (1953) 71 RPC 23 at 31 (Evershed MR); Harrods Ltd v Harrodian School Ltd [1996] RPC 697 at 716 (Millet LJ) and 722 (Sir Michael Kerr, dissenting); Phones 4U Ltd v Phone4u.co.uk. Internet Ltd [2006] EWCA Civ 244, [2007] RPC 5 at [45] (Jacob LJ); Specsavers International Healthcare Ltd v Asda Stores Ltd [2010] EWHC 2035 (Ch), [2011] FSR 1 at [101] (Mann J); Fine & Country Ltd v Okotoks Ltd [2012] EWHC 2230 (Ch) at [84]-[87] (Hildyard J); and Jack Wills Ltd v House of Fraser (Stores) Ltd [2014] EWHC 110 (Ch), [2014] FSR 39 at [99]. It is not necessary for present purposes to traverse all these authorities again, although it is worth noting that in Harrods Millett LJ thought that the absence of evidence of confusion after a period of 15 months was "not without significance" and in Phones 4U Jacob LJ said at [42] that "evidence of substantial side-by-side trade without significant confusion or deception gives rise to a powerful inference that there is no such confusion or deception". It is sufficient to repeat what I said in Jack Wills and in Samuel Smith Old Brewery (Tadcaster) v Lee [2011] EWHC 1879 (Ch), [2012] FSR 7 at [95]:

"… absence of evidence of actual confusion is not necessarily fatal to a claim under Article 5(1)(b) [of Directive 89/104]. The longer the use complained of has gone on in parallel with use of the trade mark without such evidence emerging, however, the more significant it is."

"In considering the weight to be attached to this factor, it is relevant to consider what opportunity there has been for confusion to occur and what opportunity there has been for any such confusion to have been detected."

40. The judge went on:

"99. In the present case, notwithstanding that Muzmatch has been trading since April 2011 (albeit not on a particularly large scale until after 2016), I find that the absence of evidence of actual confusion is not because there was no relevant confusion but rather reflects the fact that Match's and Muzmatch's services have been provided via their websites and, more recently, their apps. Where services are provided to a customer in that manner, then any other dealings with that customer are likely to be conducted via that same website or app and are likely to relate to the services that that customer had received via that website or app. On this basis, …, it is hard to see how or why any confusion that there might be in the customer's mind would come to light. It would rarely be relevant to the customer's issues - and that this is so even for those customers who use … two or three different on-line dating platforms. Given this, it seems to me that, despite the longer period of trading involved, this case falls squarely within the scope of what Mann J said in Specsavers as set out above.

100.     In arguing that confusion would have come to light, [counsel for the Defendants] pointed to Mr Younas's evidence that Muzmatch had been contacted by customers of Minder (another provider of dating services) in relation to matters concerning Minder's services. The difficulty with this is that I know nothing about Minder and its services, save that Mr Younas has said that it had copied a number of app features and it looked similar to the Muzmatch app. Otherwise, I have no evidence as to how Minder operated and I do not think that this evidence helps me in considering the likelihood of confusion arising from two businesses like Match and Muzmatch that are using similar names."

41. Turning to the three instances of alleged actual confusion, no witnesses had been called to testify to these. The judge did not place any great reliance on them. The most that could be said was that two of the instances provided some, albeit limited, support for the Claimants' case:

"102. The first instance is based on an email received by Match on 9 May 2021 from a person regarding 'an account on muz match' which, that person said, was not his account but which had used his pictures. … on its face, this may not be evidence of relevant confusion given that the person concerned claimed not to be a user of any dating websites. However, the fact that someone had contacted Match with a complaint regarding Muzmatch is evidence of some sort of a likelihood of confusion arising from the similarity in the names, although in the absence of more details, it is difficult to place too much reliance on this particular instance.

103.     The second instance is in a website chat in July 2018. This involved a person in the USA who appears to have been a customer of both Match and Muzmatch and who complained to Match about a $20 charge levied by Muzmatch. … it is not evidence of actual confusion in the UK. However, the fact that a customer made that mistake in the US shows that there is a likelihood of such confusion arising in similar circumstances in the UK. Again, in the absence of more details, it is difficult to place too much reliance on this."

The appeal

42. Since the judge's conclusion that there was a likelihood of (indirect) confusion involved a multi-factorial evaluation, this Court can only intervene if he erred in law or in principle: compare Actavis Group PTC EHF v ICOS Corp [2019] UKSC 15, [2019] Bus LR 1318 at [78]-[81] (Lord Hodge) and see Re Sprintroom Ltd [2019] EWCA Civ 932, [2019] BCC 1031 at [72]-[78] (McCombe, Leggatt and Rose LJJ). The Defendants have permission to appeal against the finding of a likelihood of confusion on three grounds.
43. The first ground is that the judge failed to take into account the characteristics of the average consumer when making his global assessment, and in particular his earlier finding that the average consumer would pay a higher level of attention than for normal consumer items. It is true that the judge did not explicitly state that he was taking the average consumer's level of attention into account when considering likelihood of confusion, but he did assess the matter from the perspective of the average consumer. There is no reason to think that he did not take into account his own finding as to their level of attention less than 20 paragraphs previously.
44. The second ground is that the judge failed to take into account the differences between Muzmatch's services (aimed specifically at the Muslim community with features designed to reflect Islamic values) and the Claimants' services (aimed at the general public), which was important context for the assessment of the likelihood of confusion. I do not accept this for two reasons. First, as noted above, the Defendants admitted that their services were identical to those for which the Trade Marks were registered. Furthermore, the judge found that the average consumer could be of any religion. In those circumstances there is no separate contextual factor which can be relied upon by the Defendants as negativing a likelihood of confusion. Secondly, and in any event, given the judge's findings as to the distinctiveness and reputation of the Trade Marks, the key question with respect to likelihood of confusion is how signs comprising MUZMATCH would be perceived by consumers familiar with the Trade Marks. The judge found at [75] that, not only were the Claimants' services not targeted at any particular group of users, but also it was probable (based on evidence from later years) that a reasonable number of the Claimants' users would have identified themselves as Muslims. Thus there was in fact an overlap between the consumers targeted by the Claimants' services and those targeted by the Defendants' services.
45. The third ground cannot be dismissed so easily. This is that the judge erred in failing to take into account the absence of evidence of actual confusion despite the parties having traded side-by-side for over 10½ years by the time of the trial in January 2022. (Although, as counsel for the Claimants pointed out, the Defendants' business appears to have been on a small scale prior to the launch of the app in March 2015, that is still nearly seven years prior to trial.) In well-focussed submissions counsel for the Defendants argued that the judge's reasoning suffered from six identifiable flaws.
46. First, she pointed out that the judge's reasoning at [99] was not that there had been no opportunity for confusion to occur, but rather that confusion was unlikely to have been detected. This was based on the proposition that, where services are provided to a customer by a website or app, any other dealings with that customer are likely to be conducted via that website or app and are likely to relate to the services that the customer had received via that website or app. Counsel for the Defendants submitted that this was erroneous because it ignored the judge's findings that both sides had promoted their respective services by social media. It followed that customers were just as likely to communicate with the parties via social media as via the respective websites or apps. In those circumstances, she submitted, the observation I made in Glaxo Wellcome UK Ltd v Sandoz Ltd [2019] EWHC 2545 (Ch), [2019] RPC 27 at [265] was pertinent: "… the ubiquitous use of social media makes it much easier than it used to be for instances of confusion to be publicised and for well-resourced claimants to find them". In this regard, she relied upon the fact that, between them, the parties had (as the judge recorded in a footnote) searched more than 53.7 million records using 17 keywords for the purposes of disclosure.
47. Although this submission appears at first blush to have some force, the difficulty with it is that its cogency depends on the answers to evidential questions which do not appear to have been investigated at trial. Not only was Glaxo v Sandoz not cited to the judge, but also it is far from clear that relevant searches were carried out by the parties. Counsel for the Defendants relied upon statements made in the Claimants' List of Documents and in a response by the Claimants to a Request for Further Information. The List of Documents stated that the Claimants had searched "retained databases of customer care records, being the Akio database of customer care records for Match UK prior to 22 May 2018, the Easiware database of customer care records for Match from 22 May 2018 onwards, the database of Match US, and the database of customer care records [of] Tinder 2011-2019 (fed via the web desktop, app, Twitter)". The response to the RFI stated that these databases comprised approximately 9.8 million records, 710,000 records, 16.2 million records and 18.1 million records respectively, a total of 44.81 million records. In the case of the Akio and Easiware databases, however, the response to the RFI stated that these included records relating to other UK, EU and Nordic brands, but in both cases the whole database was searched for efficiency. Moreover, although counsel for the Defendants particularly relied upon the reference to Twitter, it is clear from the response to the RFI that this related to the database of customer care records for Tinder. There is no statement that the other databases of customer care records were fed via Twitter (or other social media platforms). Moreover, so far as the evidence shown to this Court goes, it does not appear that the Claimants carried out any direct searches of social media for instances of confusion. Nor does it appear that the Defendants did so. In those circumstances I do not consider that the judge can be criticised for not taking into account the possibility of instances of confusion being detected by means of searches of social media.
48. Secondly, counsel for the Defendants submitted that the judge had been wrong at [100] to dismiss Mr Younas' evidence concerning users of Minder contacting Muzmatch about issues with Minder. The point of that evidence was that it showed that instances of confusion could come to light although Minder's services were also provided via an app. In my view, however, the judge was entitled to consider that this evidence was of little weight without knowing more about Minder and the relevant circumstances. In particular, there does not seem to have been any evidence as to how the users in question contacted Muzmatch.
49. Thirdly, counsel for the Defendants submitted that the judge's reasoning was contradicted by the two instances of alleged confusion he discussed at [102]-[103]. In both cases, a supposedly confused person had contacted the Claimants about Muzmatch. The judge was well aware of this, however, and I do not think that these instances invalidate his assessment that confusion was unlikely to be detected. As counsel for the Claimants pointed out, the judge found that there was a likelihood of indirect confusion rather than direct confusion. Indirect confusion is inherently less likely to come to light than direct confusion.
50. Fourthly, counsel for the Defendants submitted that the judge had failed to take into account Mr Younas' unchallenged evidence, which the judge had recorded at [46(c)], that users of Muzmatch were encouraged to report inappropriate content either through an in-app reporting function or by email and that some 3,915 reports had been filed in September 2021 alone. Again, however, the judge was well aware of this evidence and it does not invalidate his assessment that confusion was unlikely to be detected.
51. Fifthly, counsel for the Defendants submitted that, based on the judge's reference to Specsavers at [100], the judge appeared only to have considered the likelihood of confused users raising complaints with the wrong provider when confusion might manifest itself in other ways. I accept that confusion could manifest itself in other ways, and that Specsavers is distinguishable from the present case for a number of reasons, but I consider that the judge was entitled to take the view that confusion was most likely to be detected as a result of complaints. Confusion which did not give rise to any complaint would be even less likely to be detected.
52. Sixthly, counsel for the Defendants submitted that the judge had failed to give any real weight to the extensive disclosure exercise carried out by the parties. I have already considered this in the specific context of the argument about social media. So far as the more general point is concerned, the judge acknowledged the extensive searches that had been carried out. He was entitled to take the view that the carrying out of extensive searches did not prove that actual confusion was likely to have been detected.
53. Having considered each of the Defendants' arguments, I would add that these are all questions of assessment of the evidence falling squarely within the remit of the judge. The judge was alive to the point that there was very little, if any, evidence of actual confusion despite over 10 years of parallel trade. The weight to be accorded to this factor was a matter for him. He made no error of principle in deciding to give it little weight in the particular circumstances of this case.

Passing off

54. The judge found passing off established for essentially the same reasons as he found infringement under Article 9(2)(b) of the EUTM Regulation and section 10(2) of the 1994 Act. In this Court it was common ground between the parties that the two stood or fell together.

Infringement under Article 9(2)(c) of the EUTM Regulation/section 10(3) of the 1994 Act

The law

55. A proprietor of a registered trade mark alleging infringement under Article 9(2)(c) of the EUTM Regulation, Article 10(2)(c) of Directive 2015/2436 and section 10(3) of the 1994 Act must show that the following requirements are satisfied: (i) the registered trade mark must have a reputation in the relevant territory; (ii) there must be use of a sign by a third party in the relevant territory; (iii) the use must be in the course of trade; (iv) it must be without the consent of the proprietor; (v) it must be of a sign which is identical with or similar to the trade mark; (vi) it must be in relation to goods or services; (vii) it must give rise to a link between the sign and the trade mark in the mind of the average consumer; (viii) it must give rise to one of three types of injury, that is to say, (a) detriment to the distinctive character of the trade mark, (b) detriment to the repute of the trade mark, or (c) unfair advantage being taken of the distinctive character or repute of the trade mark; and (ix) it must be without due cause.
56. It is sufficient for the use of the sign to give rise to a link in the mind of the average consumer that the sign would call the registered trade mark to mind even if the average consumer would not be likely to be confused as a result: see Case C-252/07 Intel Corporation Inc v CPM United Kingdom Ltd [2008] ECR I-8823 at [60]. This must, like the question whether there is a likelihood of confusion, be appreciated globally taking into account all factors relevant to the circumstances of the case: see Intel at [41].

The judge's assessment

57. The issues between the parties were as to requirements (i), (vii), (viii) and (ix). So far as requirement for a reputation is concerned, the judge found at [110]-[111] that the 1996 EU Trade Mark had a reputation by 2011, that the same reputation would have attached itself to the 2017 EU Trade Mark upon its being registered and that the UK Trade Mark also had a reputation as soon as it was registered.
58. The judge found that the requirement for a link was satisfied for the reasons he expressed at [115] as follows:

"Although I am considering the application of s.10(3) on the assumption that I was wrong to conclude that there was a likelihood of confusion, the factors that led me to reach that conclusion would, in my judgment, establish a link in the mind of the average consumer as between the Muzmatch Signs and the [Trade] Marks. In summary, by 2011 the 1996 [EU Trade] Mark had established a very significant reputation in the market (which was also enjoyed by the 2017 [EU Trade] Mark from its filing date), there is medium level of similarity between the [EU Trade] Marks and the Muzmatch Signs, and there is either an identity or a reasonably high level of similarity between the parties' goods/services. As the use made of the Muzmatch Signs was clearly use intended to be distinctive (as a brand) rather than descriptive, I am satisfied that the average consumer would have made a link between those Signs and the [EU Trade] Marks. I am aware that the evidence of Mr Younas and Mr Craig was that the Signs did not create any link in their minds. However, in this respect, I do not think that their views are representative of those of the average consumer."

Although it is not clear from this paragraph, it is evident from what the judge went on to say that this finding extended to the UK Trade Mark.

59. The judge found that the use of the signs complained of took unfair advantage of the reputation of the Trade Marks and that this was without due cause. His reasons for reaching these conclusions do not matter for the reasons explained below.

The appeal

60. The Defendants have permission to appeal against the judge's finding that the use of the signs complained of would give rise to a link in the mind of the average consumer. There was a dispute between the parties as to whether or not the Defendants had also been given permission to appeal on the question of unfair advantage, but counsel for the Defendants did not pursue the point.
61. As to the judge's finding of a link, counsel for the Defendants submitted that the judge had erred for the same reasons that she submitted that he had erred in relation to likelihood of confusion. Since I have rejected those reasons, this ground of appeal falls with them. I would add, however, that, even if the judge had erred in finding that there was a likelihood of confusion, his finding of a link was in my view unimpeachable.

Honest concurrent use

62. I now turn to the most important issue raised by this appeal. The Defendants contend that they have a defence to the Claimants' claims of trade mark infringement and passing off by virtue of "honest concurrent use". The Claimants dispute that honest concurrent use is a free-standing defence, as opposed to a factor to be taken into account when assessing infringement, and contend that the judge took the long concurrent use of MATCH and MUZMATCH fully into account when finding that there was a likelihood of confusion and that the Defendants' use had taken unfair advantage of the reputation of the Trade Marks. In the alternative the Claimants contend that the Defendants' use of the signs complained of was not honest.

The judge's assessment

63. The judge adopted at [138] my suggestion in W3 Ltd v easyGroup Ltd [2018] EWHC 7 (Ch), [2018] FSR 16 at [287] that, in cases brought under section 10(2) of the 1994 Act and Article 9(2)(b) of the EUTM Regulation, honest concurrent use was not a separate defence, but rather a factor that falls to be taken into account as part of the global assessment. Moreover, he considered that the same was true of cases brought under section 10(3)/Article 9(2)(c). The judge nevertheless went on to reject the defence on its merits for reasons he expressed at [141] as follows:

"In cases like Budweiser and the other honest concurrent user cases relied on by [counsel for the Defendants], the use relied on as being honest concurrent use seems not to have been infringing use when it started. Indeed, in each of them, the claimant's mark sued on was registered after (usually well after) the honest concurrent user had started. In contrast, in the present case, Muzmatch's use started as infringing use of Match's prior registered Mark; that use had interfered with the essential function of that Mark, had given rise to a likelihood of confusion and/or had taken unfair advantage of the reputation of that Mark. Moreover, given the 26^th January 2016 letter and Match's (successful) opposition to the registration of the Muzmatch name as an EU and a UK trade mark …, Muzmatch was well aware that Match regarded the Muzmatch Signs as infringing. In the absence of, say, estoppel or statutory acquiescence, it is difficult to see how use that starts as infringing use can be turned into honest use by the fact that it has continued. The very reason why use infringes is because, if left unchecked, its effect would be to educate the public that the mark does not indicate the services of the trade mark proprietor. It must be a very unusual case where (again, in the absence of estoppel or statutory acquiescence) a defendant can argue that continuing to infringe gives rise to a defence to an infringement claim. I do not think that this is such a case and, in my judgment, Muzmatch's use cannot be regarded as honest concurrent use. Whilst I accept that Mr Younas did not intend to create a link to or confusion with the Match Marks (because he wrongly saw use of the word 'match' as descriptive in this context and because he was serving only the Muslim community), I do not see that this can give rise to a defence when, objectively viewed, his actions infringed Match's rights for the reasons set out above."

English trade mark law prior to the 1994 Act

64. Honest concurrent use is a concept which has its origin in English equitable doctrine prior to 1875. As Lord Diplock explained in a magisterial speech (delivered after 13 days of argument which is reported in [1973] RPC 297 at 299-316) with which Lords Reid, Simon of Glaisdale and Kilbrandon agreed (Lord Morris of Borth-y-Gest having fallen ill halfway through the hearing) in General Electric Co v General Electric Co Ltd [1972] 1 WLR 729 at 742-743:

"A right of property of this character [sc. the right of property in a trade mark] calls for an accommodation between the conflicting interests of the owner of the monopoly, of the general public as purchasers of goods to which the trade mark is affixed, and of other traders. This accommodation had been substantially worked out by the Court of Chancery by 1875.

The interest of the general public requires that they should not be deceived by the trade mark. It ought not to tell a lie about the goods. Two main kinds of deception had been the subject of consideration. These were misrepresentation (a) of the character of the goods to which the trade mark was attached, and (b) as to their origin, i.e. that they were the product of some other manufacturer.

But the interest of the public in not being deceived about the origin of goods had and has to be accommodated with the vested right of property of traders in trade marks which they have honestly adopted and which by public use have attracted a valuable goodwill. In the early 19th century trade was still largely local; marks which were identical or which closely resembled one another might have been innocently adopted by traders in different localities. In these their respective products were not sold in competition with one another and accordingly no question of deception of the public could then arise. With the rapid improvement in communications, however, in the first half of the 19th century markets expanded; products of two traders who used similar marks upon their goods could thus come to be on sale to the same potential purchasers with the consequent risk of their being misled as to the origin of the goods. Furthermore, it was accepted that as an adjunct of the goodwill of the business the right to use a trade mark might be acquired by more than one successor if the goodwill of the business were divided, as it might be, for instance, where the business had formerly been carried on in partnership or from more than one manufactory or shop. To meet this kind of situation the doctrine of honest concurrent user was evolved. Under this doctrine a trade mark remained entitled to protection in cases where the use of it had not originally been deceptive but a risk of deception had subsequently arisen as a result of events which did not involve any dishonesty or other wrongful conduct upon the part of the proprietor of the mark. If, however, his own wrongful conduct had played a part in making the use of the mark deceptive, the Court of Chancery would not grant him an injunction against infringement. This was but a particular application of the general equitable doctrine that he who seeks equity must come with clean hands.

In cases of honest concurrent user, neither of the owners of the mark could restrict the other from using it, but as against a usurper who infringed it either owner of the mark could obtain an injunction: Dent v. Turpin (1861) 2 J. & H. 139 and Southorn v. Reynolds (1865) 12 L.T. 75"

65. As Lord Diplock went on to explain, after 1875 the doctrine of honest concurrent use was relied upon by the courts to permit registration of identical or similar trade marks in respect of the same classes of goods pursuant to the general discretion conferred by section 6 of the Trade Marks Act 1875. This basis for concurrent registration received express statutory recognition in section 21 of the Trade Marks Act 1905, and that was repeated in section 12(2) of the Trade Marks Act 1938. Lord Diplock summarised the legal status under the 1938 Act of a registered trade mark the use of which was likely to cause confusion at 751 as follows:

"(1)  The fact that the mark is entered upon the register is prima facie evidence of the validity of the original registration and of the right of the registered proprietor to the exclusive use of the mark, subject however to the rights of concurrent user by any registered proprietor of an identical mark or one nearly resembling it.

(2)  If the mark was likely to cause confusion at the time when it was first registered it may be expunged from the register as 'entry made in the register without sufficient cause' unless the proprietor of the mark at that time would have been entitled to have it entered on the register by reason of his honest concurrent use of the mark as a trade mark before the original registration of the mark.

(3)  If the likelihood of causing confusion did not exist at the time when the mark was first registered, but was the result of events occurring between that date and the date of application to expunge it, the mark may not be expunged from the register as an entry wrongly remaining on the register, unless the likelihood of causing deception resulted from some blameworthy act of the registered proprietor of the mark or of a predecessor in title of his as registered proprietor.

(4)  Where a mark is liable to be expunged under (2) or (3) the court has a discretion whether or not to expunge it and as to any conditions or limitations to be imposed in the event of its being permitted to remain on the register."

66. I would add that, by virtue of section 39 of the 1905 Act and section 4(4) of the 1938 Act, concurrent registration provided a defence to infringement. Section 7 of the 1938 Act also provided a saving for vested rights. Where the defendant had no registration to found a defence under section 4(4) and was not within section 7, the defendant's only remedy was to apply for a stay of the infringement action pending the determination of an application for registration, which if successful would preclude the grant of an injunction but (probably) not an award of damages: see Kerly's Law of Trade Marks and Trade Names (12^th ed) at 15-24 to 15-26.
67. It can be seen from the passage I have quoted from Lord Diplock's speech in paragraph 64 above that the doctrine as he formulated it concerned use of a trade mark which "had not originally been deceptive but a risk of deception had subsequently arisen as a result of events which did not involve any dishonesty or other wrongful conduct upon the part of the proprietor of the mark". Later in his speech, at pages 746, 747, 750, in the passage quoted in paragraph 65 above and at 751, Lord Diplock referred to "blameworthy conduct" and "some blameworthy act". Although Lord Diplock did not say so in terms, it is clear from his speech that whether conduct was "wrongful" or "blameworthy" was to be judged objectively. It did not require an intention to deceive.

English law of passing off

68. Although Lord Diplock was considering the law of trade marks, it is important to note that a parallel doctrine, generally known as "concurrent goodwill", but also sometimes referred to as "honest concurrent use", forms part of the law of passing off, which unlike the law of trade marks is entirely a creature of the common law. In that context Oliver LJ, with whom Watkins and Stephenson LJJ agreed, said in Habib Bank Ltd v Habib Bank AG [1981] 1 WLR 1265 at 1275:

"I think, if I may say so, that Mr. Jeffs's submissions are too ambitious in this sense, that they seek to elevate into a doctrine dignified by a term of art — 'the doctrine of honest concurrent user' — what is, in the sphere of passing off, merely a facet of Lord Diplock's first essential ingredient of misrepresentation in Erven Warnink Besloten Vennootschap v. J. Townend & Sons (Hull) Ltd. [1979] A.C. 731, 742. As Mr. Aldous has pointed out, section 2 of the Trade Marks Act 1938 expressly provides that nothing in the Act is deemed to affect any rights of action for passing off. The fact therefore that two or more people may be entitled to rely upon honest concurrent user of a mark to achieve registration leaves quite unaffected the question of whether they may be entitled to sue one another in a passing off action. What I think Mr. Jeffs is really saying in propounding his doctrine is really this, that where you find that two traders have been concurrently using in the United Kingdom the same or similar names for their goods or businesses, you may well find a factual situation in which neither of them can be said to be guilty of any misrepresentation. Each represents nothing but the truth, that a particular name or mark is associated with his goods or business."

69. Notwithstanding that observation, some subsequent authorities have treated Lord Diplock's statement of principles in General Electric as applicable to passing off cases: see in particular Pete Waterman Ltd v CBS United Kingdom Ltd [1993] EMLR 27 at 44-55 and 50 (Sir Nicolas Browne-Wilkinson V-C). Moreover, in Phones 4U (cited above), Jacob LJ said this about the law of passing off, albeit without citing General Electric:

"21.  In this discussion of 'deception/confusion' it should be remembered that there are cases where what at first sight may look like deception and indeed will involve deception, is nonetheless justified in law. I have in mind cases of honest concurrent use and very descriptive marks. Sometimes such cases are described as 'mere confusion' but they are not really—they are cases of tolerated deception or a tolerated level of deception.

22.  An example of the former is the old case of Dent v Turpin (1861) 2 J&H 139. Father Dent had two clock shops, one in the City, the other in the West End. He bequeathed one to each son—which resulted in two clock businesses each called Dent. Neither could stop the other; each could stop a third party (a villain rather appropriately named Turpin) from using 'Dent' for such a business. A member of the public who only knew of one of the two businesses would assume that the other was part of it—he would be deceived. Yet passing off would not lie for one son against the other because of the positive right of the other business. However it would lie against the third party usurper."

70. A recent example of the application of similar principles to those articulated by Lord Diplock to a claim for passing off, albeit again without citation of General Electric, is Sir Robert McAlpine Ltd v Alfred McAlpine plc [2004] EWHC 630 (Ch), [2004] RPC 36. In that case the claimant and the defendant were both well-known construction companies. The McAlpine family's construction business had been set up in 1869 by Robert (later Sir Robert) McAlpine. By 1935 the business was being run by three of the founder's sons, one of whom was named Alfred. It was agreed in that year that the business should split, essentially on geographical lines: Alfred's company, the defendant, would trade in the North West, the Midlands and North Wales, while the claimant would trade in the remainder of the country. The geographical split of work was maintained until 1983, when it was agreed that it would cease. By that time the name "McAlpine" was very well known in the construction trade, and each company was sensitive to the need to distinguish between the two companies. This was done informally by making sure that each company used the appropriate forename in describing itself and its activities. There was a significant overlap in their work in the construction and civil engineering fields. In 2001 the defendant began a re-branding exercise with the intention of creating a more modern image which involved dropping the word "Alfred" and trading simply as "McAlpine". The claimant succeeded in a claim for passing off. It was common ground between the parties that they had a shared goodwill in the name "McAlpine", although as Wadlow on the Law of Passing Off (6^th ed) points out at 3-395 the better view is that they had independent goodwills in the names "Robert McAlpine" and "Alfred McAlpine" respectively. Unsurprisingly, there was evidence of some level of confusion prior to the defendant's rebranding. Although there was little, if any, evidence that confusion had materially increased after the rebranding, Mann J held that the defendant's new trading style involved a misrepresentation that the defendant was the only company known by that name when it was not and that that misrepresentation was damaging to the claimant's goodwill.
71. Having briefly reviewed Dent v Turpin, General Electric, Habib Bank, Pete Waterman, McAlpine and Phones 4U, Recorder Purvis QC sitting in the Patents County Court concluded in W.S. Foster & Son Ltd v Brooks Brothers UK Ltd [2013] EWPCC 18 at [61]:

"The authorities therefore seem to me to establish that a defence of honest concurrent use in a passing off action requires at least the following conditions to be satisfied:

(i)  the first use of the sign complained of in the United Kingdom by the Defendant or his predecessor in title must have been entirely legitimate (not itself an act of passing off);

(ii)  by the time of the acts alleged to amount to passing off, the Defendant or his predecessor in title must have made sufficient use of the sign complained of to establish a protectable goodwill of his own;

(iii)  the acts alleged to amount to passing off must not be materially different from the way in which the Defendant had previously carried on business when the sign was originally and legitimately used, the test for materiality being that the difference will significantly increase the likelihood of deception."

72. This formulation of the applicable principles appears to exclude the possibility that use of a sign which constituted passing off at its inception may cease to do so even if, through inactivity on the part of the claimant and the passage of time, the relevant class of consumers learns to distinguish between the goods or services denoted by the respective signs, or least most of them do. However, in DaimlerChrysler AG v Alavi [2001] RPC 42 Pumfrey J said at [67]:

"… there must come a time after which the court would not interfere with a continued course of trading which might have involved passing off at its inception but no longer did so: logically, this point would come six years after it could safely be said that there was no deception and independent goodwill had been established in the market by the protagonists."

73. Pumfrey J's reference to six years shows that he was contemplating a claim for passing off being barred by limitation, bearing in mind that passing off is (like trade mark infringement) a tort that is repeated each time the sign is used. Although limitation bars the remedy not the right, it seems to me that Pumfrey J's logic is unassailable so far it goes. But his statement does not answer the question as to what the position is absent a limitation bar.
74. Having cited Pumfrey J's statement at 9-129, Wadlow goes on at 9-131:

"The term 'honest concurrent use' derived from successive Trade Marks Acts²¹⁶ is inappropriate in passing-off.²¹⁷ If there has been concurrent user in fact, then it cannot be conclusive whether or not it was honest. The distinctiveness of marks is frequently destroyed by conduct which would have been actionable, even fraudulent, had the claimant acted in time. A fortiori, a concurrent right to use the mark, or more properly an immunity, can be obtained by use which was less than honest in its inception."

(Footnote 217 cites Habib Bank.)

75. On the other hand, neither the researches of counsel nor my own have found any case in which it has actually been held that conduct which involved passing off at its inception has ceased to do so purely through inactivity of the claimant and effluxion of time (i.e. in the absence of some affirmative defence such as acquiescence), although it is arguable that this is implicit in some of the decided cases. I shall give an example of this below. I shall also give an example of a parallel in the context of trade mark infringement.

The 1994 Act

76. The 1994 Act was passed for the principal purposes of implementing Directive 89/104 and making provision in connection with Council Regulation 40/94/EC of 20 December 1993 on the Community trade mark (the first predecessor to the EUTM Regulation), as well as giving effect to the Madrid Protocol Relating to the International Registration of Trade Marks and the Paris Convention for the Protection of Intellectual Property. Although section 7 provided that, where there had been honest concurrent use of the trade mark for which registration was sought, the registrar should not refuse registration by reason of conflict with an earlier trade mark or other earlier right unless objection on that ground was raised in opposition proceedings by the proprietor of that earlier trade mark or right, that was a "home-grown" provision which did not derive from Directive 89/104. In any event, section 7(5) provides that section 7 does not apply when there is an order in force under section 8 (power to require that relative grounds be raised in opposition proceedings), and the Trade Marks (Relative Grounds) Order 2007, SI 2007/1976, made under section 8 has been in force since 1 October 2007.

EU trade mark law

77. As noted above, neither Directive 2015/2436 nor the EUTM Regulation provide for any defence to trade mark infringement on the ground of honest concurrent use. Furthermore, the CJEU has repeatedly held that Articles 5 to 7 of Directive 89/104 effect a complete harmonisation of the rules relating to the rights conferred by a trade mark, and accordingly define the rights of proprietors of trade marks in the European Union: see Case C-355/96 Silhouette International Schmied GmbH & Co KG v Hartlauer Handlesgeschaft mbH [1998] ECR I-4799 at [25] and [29]; Joined Cases C-414/99 to C-416/99 Zino Davidoff SA v A & G Imports Ltd [2001] ECR I-8691 at [39]; Case C-324/08 Makro Zelfbedieningsgroothandel CV v Diesel SpA [2009] ECR I-10019 at [20]; Case C-127/09 Coty Prestige Lancaster Group GmbH v Simex Trading AG [2010] ECR I-4965 at [27]; Case C-482/09 Budejovický Budvar np v Anheuser-Busch Inc [2011] ECR I-8701 ("Budweiser") at [32]; Case C-661/11 Martin Y Paz Diffusion SA v Depuydt [EU:C:2013:577] at [54]; and Case C-500/14 Ford Motor Co v Wheeltrims srl [EU:C:2015:680] at [44]. Consequently, save for the specific cases governed by Article 8 et seq of Directive 89/104, a national court may not, in a dispute relating to the exercise of the exclusive right conferred by a trade mark, limit that exclusive right in a manner which exceeds the limitations arising from Articles 5 to 7 of the Directive: Martin Y Paz at [55]; and Ford v Wheeltrims at [44]. It is clear, and undisputed, that the same principles apply to the corresponding provisions in Directive 2015/2426 and the EUTM Regulation.
78. The case law of the CJEU establishes that six conditions must be fulfilled in order for there to be infringement under Article 10(2)(a) of Directive 2015/2436 and Article 9(2)(a) of the EUTM Regulation: (i) there must be use of a sign by a third party within the relevant territory; (ii) the use must be in the course of trade; (iii) it must be without the consent of the proprietor of the trade mark; (iv) it must be of a sign which is identical to the trade mark; (v) it must be (a) in relation to goods or services (b) which are identical to those for which the trade mark is registered; and (vi) the use must affect, or be liable to affect, one of the functions of the trade mark. The first five conditions are apparent on the face of the legislation. Condition (vi) has been read into the legislation by the CJEU as a matter of interpretation: see Supreme Petfoods v Henry Bell & Co (Grantham) Ltd [2015] EWHC 256 (Ch), [2015] RPC 22 at [86]-[134] and the cases discussed there.
79. The question of how, if at all, honest concurrent use fits into this legal framework was referred by this Court to the CJEU in Budweiser.
80. Anheuser-Busch ("AB") is a brewer based in Missouri which (together with its predecessors) has sold beer under the name "Budweiser" since about 1875. Budejovický Budvar ("BB") is a brewer based in the town of Ceske Budovice (formerly "Budweis") in the Czech Republic. Since 1895 it (together with its predecessors) has also sold beer under names consisting of or including the word "Budweiser". Although the names are the same, the beers are not. Their get-ups, tastes and prices have always been different. In markets where they have co-existed, consumers have by and large become aware of the difference between the beers, though it is likely that there is some small level of confusion. As world trade expanded, so too the markets of the two companies expanded and overlapped. This has led to more than a hundred years of worldwide trade mark litigation.
81. In the UK BB was the first actively to enter the market in late 1973. AB entered the UK market in 1974, although it had previously supplied reasonably substantial quantities of its beer to US service personnel stationed at US bases in the UK through "PX" stores. In 1979 AB sued BB for passing off. BB counterclaimed. The claim and counterclaim were both dismissed: see Anheuser-Busch Inc v Budejovický Budvar np [1984] FSR 413. Whitford J held and the Court of Appeal (Oliver, O'Connor and Dillon LJJ) agreed, albeit for somewhat different reasons, that both sides were entitled to use the name Budweiser in the UK. In reaching this conclusion the Court of Appeal held that AB's sales to the US service personnel prior to late 1973 were to be disregarded since this did not constitute the carrying on of a business in the UK to which goodwill could attach. It is implicit in the Court of Appeal's reasoning that BB had not established a sufficient goodwill by 1974 to prevent AB from using the name either.
82. Jacob LJ subsequently opined, in his judgment giving the Court of Appeal's reasons for making the reference to the CJEU, that this was a case of honest concurrent use even though that was not how it had been analysed by either Whitford J or the Court of Appeal: see Budejovický Budvar np v Anheuser-Busch Inc [2009] EWCA Civ 1022, [2010] RPC 7 at [11].
83. In 1979 AB applied to register BUDWEISER as a UK trade mark. BB opposed the application. In 1989 BB applied to register BUDWEISER as a UK trade mark. AB opposed the application. The Registrar of Trade Marks, Rimer J and the Court of Appeal (Peter Gibson LJ, Judge LJ and Ferris J) all held that both sides were entitled to register the mark, in the case of BB on the ground of honest concurrent use within section 12(2) of the 1938 Act: see BUDWEISER Trade Marks [2000] RPC 906. Both marks were entered on the register on the same day, 19 May 2000.
84. In 2005 AB applied for a declaration of invalidity of BB's registration pursuant to section 5(1) implementing Article 4(1)(a) of Directive 89/104 (identical marks, identical goods or services) on the ground that its application had pre-dated BB's application, and so its trade mark was an earlier trade mark. Following decisions by the hearing officer and Norris J substantially in favour of AB, the Court of Appeal (Ward LJ, Jacob LJ and Warren J) referred three questions to the CJEU, the third of which essentially asked whether honest concurrent use could be relied upon as an answer to a claim for a declaration of invalidity under Article 4(1)(a).
85. The Court of Justice began its consideration of the third question by noting at [67]-[70] that Article 4(1)(a) was the counterpart in the registration context to Article 5(1)(a) in the infringement context, and thus the Court's interpretation of Article 5(1)(a) was applicable to Article 4(1)(a). It continued:

"71. It follows from the Court's case-law that the exclusive right under Article 5(1)(a) … was conferred in order to enable the trade mark proprietor to protect his specific interests as proprietor, that is, to ensure that the trade mark can fulfil its functions and that, therefore, the exercise of that right must be reserved to cases in which another party's use of the sign affects or is liable to affect the functions of the trade mark. Those functions include not only the essential function of the trade mark, which is to guarantee to consumers the origin of the goods or services, but also its other functions, in particular that of guaranteeing the quality of the goods or services in question and those of communication, investment or advertising (see, inter alia, Case C-487/07 L'Oréal and Others [2009] ECR I-5185, paragraph 58, and Joined Cases C-236/08 to C-238/08 Google France and Google [2010] ECR I-2417, paragraph 77).

72.       It should be added that Article 5(1)(a) … does not require evidence that there exists a likelihood of confusion on the part of the public in order to afford absolute protection in the case of identity of the sign and the trade mark and of the goods or services (LTJ Diffusion, paragraph 49).

73. In the present case, the referring court asks the Court how Article 4(1)(a) of Directive 89/104 should be interpreted in the light of the trade mark's essential function.

74. In that context, it follows from the foregoing that Article 4(1)(a) … must be interpreted as meaning that a later registered trade mark is liable to be declared invalid where it is identical with an earlier trade mark, where the goods for which the trade mark was registered are identical with those for which the earlier trade mark is protected and where the use of the later trade mark has or is liable to have an adverse effect on the essential function of the trade mark which is to guarantee to consumers the origin of the goods.

75. In the present case, it is to be noted that the use by [BB] of the Budweiser trade mark in the United Kingdom neither has nor is liable to have an adverse effect on the essential function of the Budweiser trade mark owned by [AB]."

86. The Court went on to explain at [76]-[81] that the circumstances of the instant case were "exceptional" for five reasons:

i) "[AB] and [BB] have each been marketing their beers in the United Kingdom under the word sign 'Budweiser' or under a trade mark including that sign for almost 30 years prior to the registration of the marks concerned";

ii) "[AB] and [BB] were authorised to register jointly and concurrently their Budweiser trade marks" by the Court of Appeal judgment in 2000;

iii) "both of those companies have from the beginning used their Budweiser trade marks in good faith";

iv) "although the names are identical, United Kingdom consumers are well aware of the difference between the beers of [BB] and those of [AB], since their tastes, prices and get-ups have always been different"; and

v) "it follows from the coexistence of those two trade marks on the United Kingdom market that, even though the trade marks were identical, the beers of [AB] and [BB] were clearly identifiable as being produced by different companies".

87. Accordingly, the Court concluded at [84]:

"… the answer to the third question is that Article 4(1)(a) … must be interpreted as meaning that the proprietor of an earlier trade mark cannot obtain the cancellation of an identical later trade mark designating identical goods where there has been a long period of honest concurrent use of those two trade marks where, in circumstances such as those in the main proceedings, that use neither has nor is liable to have an adverse effect on the essential function of the trade mark which is to guarantee to consumers the origin of the goods or services."

88. When the case returned to the Court of Appeal (Budejovický Budvar np v Anheuser-Busch Inc (No 2) [2012] EWCA Civ 880, [2013] RPC 12), AB argued that, notwithstanding what the Court of Justice had said about the facts of the matter, the earlier cases between the parties had established that there was a more than de minimis level of confusion and thus the essential function of AB's trade mark was being adversely affected by BB's use of the same mark.
89. The Court of Appeal accepted that the level of confusion was more than de minimis. As Sir Robin Jacob, with whom Ward LJ and Warren J agreed, explained by reference to his earlier judgment at [17]:

"I did not hold that the level of confusion was de minimis. I said it was small, not that it was negligible. The level of confusion in the earlier cases was clearly not negligible: otherwise the claims in passing off would have failed for that reason alone, not because there was [a] defence of honest concurrent use or failure to prove goodwill. What I said is entirely consistent with the findings relied upon by [AB]. Common sense prevails here. These two brands have lived side by side for many many years in different get-ups, prices and taste[s] and with large sales: the sales of AB's beer are much greater than BB's but the latter are substantial. You do not have to be a genius to infer from those very facts alone that the public by and large will have got used to that. Or that there will always be some who are confused, albeit that many are not."

90. The Court of Appeal did not accept, however, that the CJEU's ruling was confined to cases where the level of confusion was de minimis. As Sir Robin Jacob explained:

"21.  The Court did not rule that only de minimis levels of confusion are acceptable when there is honest concurrent use. Nor did the Court rule that the inevitable confusion in a same mark/same goods case is enough to take a case out of acceptable concurrent use. Yet that is what [AB]'s submissions involve.

22.  More fundamentally, [AB]'s submissions involve the unstated premise that even where there is long established honest concurrent use the mark of one party must provide a guarantee of origin in that party and not the other. That is quite unrealistic. Here for instance, Budweiser has never denoted AB's beer alone.

23.  So I do not think that there is any impairment of the guarantee of origin – of either side's mark. The guarantee is different given a situation of long established honest concurrent use."

91. Thus honest concurrent use may be relied upon as an answer to a claim for a declaration of invalidity under Article 4(1)(a) where the use of the defendant's trade mark has not had, nor is liable to have, an adverse effect on the origin function of the claimant's trade mark because, even though there is not merely a likelihood of confusion but a small level of actual confusion, the trade mark has come to be understood by most of the relevant class of consumers as denoting different trade sources when used by the two proprietors. There is no dispute that the same applies to a claim for infringement under Article 10(2)(a) of Directive 2015/2436, and hence section 10(2) of the 1994 Act, or Article 9(2)(a) of the EUTM Regulation. Nor is there any dispute that the burden of proving this lies on the defendant: see Supreme Petfoods at [147], [164] and Bentley 1962 Ltd v Bentley Motors Ltd [2020] EWCA Civ 1726, [2021] Bus LR 736 at [35]. As I discussed in those cases, this may be rationalised on the basis that, in double identity cases, there is a rebuttable presumption of a likelihood of confusion, but honest concurrent use can be relied upon by the defendant as rebutting (or, more accurately, sufficiently rebutting) that presumption.
92. Budweiser leaves open at least four questions. First, it does not address the question of how an adverse effect on the other functions of a trade mark may be relevant to honest concurrent use in a double identity case. I considered this question in Supreme Petfoods, but it does not arise here and it is not necessary to say anything about it.
93. Secondly, Budweiser does not address the relevance of honest concurrent use to other kinds of cases, and in particular other kinds of infringement claim. As discussed below, subsequent domestic case law establishes that honest concurrent use can be relied upon to defeat an allegation of infringement under section 10(2) of the 1994 Act implementing Article 10(2)(b) of Directive 2015/2436 or Article 9(2)(b) of the EUTM Regulation, and it is common ground that the same must apply to an allegation under section 10(3)/Article 10(2)(b)/Article 9(2)(b). In the case of a claim for infringement under section 10(2)/Article 10(2)(b)/Article 9(2)(b), the claimant must show that there is a likelihood of confusion. If the claimant succeeds in doing so, it follows that the use of the sign complained of is liable adversely to affect the origin function of the trade mark. Similarly, in the case of a claim for infringement under section 10(3)/Article 10(2)(c)/Article 9(2)(c), the claimant must show that one of the three kinds of injury has occurred. Again, if the claimant succeeds in doing so, it follows that the use of the sign complained of adversely affects one or more of the functions of the trade mark. How then does honest concurrent use fit into the analysis in such cases? Who bears the burden of proof and what is the relevant date for assessment? I shall return to these questions below.
94. Thirdly, Budweiser does not settle the question of the relevant date. That was a case concerning the validity of a registration, and no doubt for that reason it appears that the CJEU was considering the position as at the date on which the marks were registered (2000). Strictly speaking, it is at least arguable that the correct date should be the date on which the later application was filed (1989), but that would not have made any material difference on the facts of that case. But what is the position in the infringement context? The general rule is that infringement is to be assessed as at the date on which the alleged infringing use was commenced: see Case C-145/05 Levi Strauss & Co v Casucci SpA [2006] ECR I-3703. If, however, the defendant uses the sign in a materially different manner or context at a later date, a new global assessment must be made as at that date: see Stichting BDO v BDO Unibank, Inc [2013] EWHC 418 (Ch), [2013] FSR 35 at [94]. Equally, the "honest practices in industrial or commercial matters" proviso to the defences under section 11(2) of the 1994 Act/Article 14(1) of Directive 2015/2436/Article 14(1) of the EUTM Regulation may mean that the defences are available as at some dates but not others: see BDO at [95]. Again, I will return to this point below.
95. Fourthly, Budweiser does not address the criterion for honesty in this context. This has, however, been considered in subsequent domestic case law, as I shall discuss.

Subsequent English cases

96. Since Budweiser there have been a number of English cases in which honest concurrent use has been invoked as a defence to infringement. It is not necessary to mention all of these, but the following should be noted.
97. In IPC v Media 10 the claimant was the publisher of Ideal Home magazine and the proprietor of a UK trade mark registration for IDEAL HOME in respect of, in essence, retail services by mail order, television and the internet. The magazine had been published since 1920. In 2005 the claimant had launched a website under the same name and in 2009 it started retailing goods via that website. The defendant was the owner of the business which ran the biannual Ideal Home Show in London and elsewhere. The exhibition had been held since 1908. The defendant launched its own online shop under the same name. The claimant brought a claim for trade mark infringement and passing off. The claim was dismissed by John Baldwin QC sitting as a Deputy High Court Judge in the IPEC and his decision was upheld by the Court of Appeal. It is important to note that Mr Baldwin held that IDEAL HOME SHOW was not identical to IDEAL HOME, and this does not appear to have been challenged in the Court of Appeal, where the claimant advanced its infringement claim under section 10(2) of the 1994 Act. It appears to have been common ground that passing off did not require separate consideration.
98. Kitchin LJ, with whom Lord Dyson MR and Bean LJ agreed, identified the central issue on the appeal at [40] as being "whether or not the use by [the defendant] of the name Ideal Home Show in relation to internet retail services is liable to have an adverse effect upon the functions of the registered trade mark in the light of the very many years during which the words Ideal Home have been used in connection with the parties' respective businesses". Having summarised the reasoning of the CJEU in Budweiser, he said:

"48. Now it is entirely true to say that the facts of the present case are different from those of Budweiser. Moreover, the circumstances of Budweiser were clearly exceptional. However, I do not understand the reasoning or guidance of the Court of Justice to be limited to only those cases which share all five characteristics of that case. To the contrary, it seems to me that the Court has made it clear that the fundamental question to be asked and answered in any particular case is whether the impugned use does or does not have an effect upon the functions of the trade mark.

49.  Further and importantly in the present context, the Court has not ruled that honest concurrent use cannot avail a trader if the impugned use is liable to cause some confusion. …"

99. Having summarised the reasoning of the Court of Appeal following the CJEU's ruling, Kitchin LJ went on at [51]:

"I respectfully agree that there may well be more than de minimis confusion in a case of honest concurrent use. No doubt many consumers will recognise that the marks are used by different businesses, but others will not. In other words, once honest concurrent use is established, the mark does not solely indicate the goods or services of just one of the users. As Sir Robin Jacob explained, in such a case the guarantee given by the mark is different."

100. Applying these principles to the facts of the case, he held that there was no infringement for reasons which are worth quoting at some length:

"52.  … The words Ideal Home have been used as or as part of the names of a magazine and a show directed to home interest products since 1920. The deputy judge has found that the reputation and goodwill attaching to each of the businesses is very substantial indeed. It follows that neither can complain about the honest use by the other of the words Ideal Home in relation to their respective core businesses, that is to say, in the case of IPC, publishing a magazine concerned with home interest products, and, in the case of Media 10, operating shows of home interest products at which such products are sold. That is so despite the fact that, as the deputy judge also found, consumers may well think that the two businesses are one and the same or are at least connected.

53.  In my judgment the same considerations apply to the provision of home interest retail services online by IPC under the name Ideal Home, and by Media 10 under the name Ideal Home Show. The deputy judge has found that these services are so closely related to the activities the parties have carried on for very many years that the words Ideal Home provide to consumers just the same guarantee that they have always provided, that is to say that the services are those of the entity responsible for running the show or the magazine or, to many consumers, the entity responsible for both of them. In other words, he has found on the facts that the name Ideal Home, when used in relation to such online retail services, does not denote and has never denoted to average consumers the services of a single entity. Rather, it denotes and has always denoted the services of both IPC and Media 10, that is to say the services of the businesses responsible for publishing Ideal Home magazine and running the Ideal Home show. Sometimes the circumstances of the use may suggest that the services are those of one business rather than the other. But that will not always be so. Further and in any event, many consumers believe the businesses are one and the same or are at least connected.

54.  I recognise, as did the deputy judge, that the confusion that occurred whilst the parties limited themselves to running a show and a magazine was more in the nature of an administrative inconvenience. I also accept the finding of the deputy judge that his conclusion opens up the possibility of some confusion between the parties' online retail businesses. But that, it seems to me, is the inevitable consequence of the use by two separate entities of the same or closely similar names in relation to such similar businesses, namely the running of a show of consumer interest products and the publishing of a magazine all about such products. I am satisfied that the provision of online retail services concerning such products was an entirely natural extension of the businesses of each of them. … Of course Media 10 must not take any steps to increase the level of confusion beyond that which is inevitable and so encroach onto IPC's goodwill. But the deputy judge was satisfied that it has not done so and in that regard it is notable that it uses for its services the name Ideal Home Show, just as it does in relation to its core business."

101. He said that it made no difference that the claimant was the senior user in relation to online retail services:

"55. … these activities of IPC have been conducted on a scale and in a manner such that they have not generated any reputation or goodwill separate or in any way different from the enormous reputation the parties enjoy in the name Ideal Home as a result of their core activities. As a result, when used in relation to online retail services the words Ideal Home denote what they have always denoted, namely that these are the services of the entities or entity responsible for the show and the magazine. The likelihood of confusion and the nature of that confusion are just the same as they would have been had the parties begun to offer such retail services at exactly the same time.

56.  I therefore believe that this is one of those rare cases in which the use of the mark complained is indeed honest and that it has not had and will not have an adverse effect upon the essential function of the registered trade mark, and that is so because the guarantee of origin the mark provides is just the same as it has always been. The deputy judge referred to the trade mark as flawed. I prefer to say (as Sir Robin Jacob did) that the guarantee is different from that which it would have been had the mark only ever been used by a single entity."

102. IPC v Media 10 is important for present purposes because it establishes four propositions. First, honest concurrent use may defeat an allegation of infringement under section 10(2) of the 1994 Act, as well under section 10(1). Secondly, this is not because honest concurrent use provides a free-standing defence, but because it shows that the conditions for infringement are not satisfied. Thirdly, it may do so even if there is some actual confusion on the part of consumers. Fourthly, it may do so even if the trade mark proprietor is the senior user in relation to the relevant services (or goods).
103. Although Kitchin LJ said that the guarantee of origin provided by the trade mark (strictly, the two trade marks) in that case was "the same as it has always been", that statement cannot be taken literally given the development of the parties' respective businesses. It is worth considering what the position would have been had the defendant's predecessor brought a claim for passing off against the claimant's predecessor in 1920, by which time the former had been running the show for 12 years. It seems likely that the claim would have succeeded. But no claim for passing off was brought. By 2012 (when the claim for trade mark infringement was commenced) the parties had coexisted for around 92 years. It is plain that the defendant could not have brought a claim for passing off, even though limitation would only have restricted its remedies to those arising out of the claimant's use during the previous six years and even though there was continuing low level confusion. Since 1920 the message conveyed to most consumers by the words "Ideal Home" had changed.
104. In Victoria Plum Ltd v Victorian Plumbing Ltd [2016] EWHC 2911 (Ch), [2017] Bus LR 363 the claimant had traded as Victoria Plumb since 2001 and changed its name to Victoria Plum in 2015, while the defendant had traded as Victorian Plumbing since 2001. Both sold bathroom fittings. The claimant owned a UK trade mark registration for VICTORIA PLUMB. The claimant contended that the defendant had infringed its trade mark and committed passing off by bidding on signs as such "victoria plumb" and "victoria plum" as search engine keywords for digital advertising. The defendant advanced a defence of honest concurrent use. In the end, Henry Carr J held that the defence was inapplicable for a series of reasons which are not relevant for present purposes. He nevertheless considered whether the defendant's use was "honest". For that purpose he discussed the test for honesty in this context.
105. In addition to the guidance given by the CJEU and this Court in Budweiser and IPC v Media 10, it appears to have been common ground between the parties that assistance could be gained from the test applied under section 11(2)(a) of the 1994 Act and the corresponding provisions in the Directives and the Regulations when considering whether the use by a person of their own name was in accordance with "honest practices in industrial or commercial matters". I had considered the case law of the CJEU on that question in Hotel Cipriani Srl v Cipriani (Grosvenor Street) Ltd [2009] EWHC 2032 (Ch), [2009] RPC 9 at [142]-[152], and that passage was not criticised on appeal [2010] EWCA Civ 110, [2010] RPC 16 (see also Samuel Smith at [112]-[118]). Henry Carr J summarised the principles to be gleaned from the discussion in Hotel Cipriani at [75] as follows:

"First, the requirement that the use be in accordance with honest practices constitutes the expression of a duty to act fairly in relation to the legitimate interests of the trade mark proprietor. Secondly, all circumstances should be considered when ascertaining whether or not the use is honest, including whether the defendant can be regarded as unfairly competing with the proprietor of the trade mark. Thirdly, an important factor is whether the use of the sign complained of either gives rise to consumer deception or takes unfair advantage of, or is detrimental to, the distinctive character or repute of the trade mark. If it does, it is unlikely to qualify as being in accordance with honest practices. Fourthly, a likelihood of confusion can be in accordance with honest practices if there is a good reason why such confusion should be tolerated. Finally, whether the defendant ought to have been aware of the existence of substantial confusion or deception is a relevant factor."

106. In relation to the fourth point, Henry Carr J also cited what Kitchin LJ had said in Maier v ASOS at [149]:

"The possibility of a limited degree of confusion does not preclude the application of the defence, however. It all depends upon the reason for that confusion and all the other circumstances of the case."

107. Henry Carr J concluded at [79]:

"In my judgment, the factors which have been considered in the context of honest commercial practices in respect of the own name defence need a degree of adaptation when considering whether 'concurrent use' is honest. In particular: (i) The defendant has a duty to act fairly in relation to the legitimate interests of the trade mark proprietor. (ii) All circumstances must be considered when ascertaining whether or not the use by the defendant is honest, including whether the defendant can be regarded as unfairly competing with the trade mark proprietor. (iii) However, the question is not simply whether use of the sign complained of gives rise to consumer deception, as such deception may have to be tolerated. Similarly, the defendant may well be aware of the existence of such confusion, having lived with it for a considerable period. (iv) The question is whether the defendant has taken steps which exacerbate the level of confusion beyond that which is inevitable and so has encroached upon the claimant's goodwill. (v) Whether the defendant ought to be aware that such steps will exacerbate confusion is a relevant factor."

108. In Walton International Ltd v Verweij Fashion BV [2018] EWHC 1608 (Ch) [2018] RPC 19 both sides had used the trade mark GIORDANO for clothing since the 1980s. The claimants' business was based in Hong Kong, and their core markets were in Asia, Australasia and the Middle East. They owned various UK and EU trade marks, including a UK registration of GIORDANO for clothing dating from 1994. The defendant's business was based in the Netherlands, and its markets were in Europe. The defendant's use in the UK had commenced in 2001, and had steadily grown as a result of organic growth since then. I held at [202]-[216] that the defendant had discharged the onus of establishing that, as a result of its honest concurrent use, its use of the sign GIORDANO neither had had, nor was liable to have, an adverse effect on any of the functions of the 1994 UK trade mark (if, contrary to another finding, it was valid). In doing so I rejected the claimants' argument that there had been no honest concurrent use by the defendant because there had been no real coexistence between the parties' trade marks which had educated the relevant public to distinguish between them.
109. The relevance of this decision for present purposes is two-fold. First, the use of the sign complained of started well after the trade mark was registered. Secondly, although there was no evidence of actual confusion, it was tolerably clear that the main reason for this was that the claimants' UK customers were almost exclusively people who had encountered the GIORDANO brand outside the UK and then continued to purchase clothes from the claimants' online shop. Although it was inherently likely that greater confusion would result if the claimants were to sell their clothes through bricks-and-mortar retail outlets (other than on a small scale through a few "pop-up" shops), that would not be the defendant's fault. That would be "no more than the inevitable consequence of the claimants having allowed the defendant to build up a substantial trade under the GIORDANO trade mark over many years". In other words, most UK consumers had come to understand the trade mark to denote the defendant's goods, not the claimants'. It is implicit in this reasoning that the defendant had infringed the trade mark pursuant to section 10(1) of the 1994 Act when it started using the sign in 2001, but had ceased to do so subsequently.
110. In Bentley 1962 Ltd v Bentley Motors Ltd [2019] EWHC 2925 (Ch), [2020] FSR 15 the defendant was the well-known car manufacturer, which was founded by Walter Owen Bentley in 1919. The claimant was a clothing supplier which was the successor to a line of predecessors starting with a company founded by one Gerald Bentley in 1962. The claimant owned three trade marks consisting of or comprising the word BENTLEY, the earliest of which dated from 1982. The defendant started selling clothing in about 1987. In November 2017 the claimant brought a claim for trade mark infringement against the defendant. The claimant's primary case, which succeeded, was under section 10(1) of the 1994 Act. Among other issues raised by the defendant was a defence of honest concurrent use. The analysis in that case was complicated by the fact that the defendant's use started before the 1994 Act came into force and continued afterwards.
111. His Honour Judge Hacon sitting as a High Court Judge considered whether the defendant had made honest concurrent use of the sign BENTLEY at [147]-[160]. He started by observing at [147] that it was "possible for the status of concurrent use of a trade, i.e. whether it is honest or not, to change over time". He found that there had been honest concurrent use of the sign at least until the 1994 Act came into force, but that from around 2000 the defendant had started to behave in a manner which was inconsistent with honest concurrent use because it amounted to a steady encroachment on the claimant's goodwill. He concluded that by November 2011 (i.e. six years before the issue of the claim form) the defendant had ceased to make honest concurrent use of the sign and it infringed the trade marks. Permission to appeal on this issue was refused, although granted on some others: see the decision of this Court (cited above) at [4].
112. Bentley is relevant for two reasons. First, it was another case in which the defendant's use started after the (earliest) registration, but nevertheless the court found that there had been honest concurrent use. Secondly, it was a case in which concurrent use that was initially honest ceased to be honest at a later date.

The appeal

113. The Defendants contend that the judge made two errors of law in rejecting the defence of honest concurrent use: first, in holding that the defence is only available where the use of the sign complained of was non-infringing use when it started; and secondly, in holding that the defence is only available if the use started before the trade mark was registered. There is an anterior question of law which is not directly addressed by the Defendants' grounds of appeal, however. As noted above, the judge held that honest concurrent use was not a separate defence, but rather a factor to be considered in the infringement analysis. Was he right about that?
114. Although counsel for the Defendants contended in her skeleton argument that the judge was wrong on this point, in oral argument she submitted that it did not matter, and that what mattered was the incidence of the burden of proof and the relevant date. So far as the burden of proof is concerned, she submitted that the burden lay on the claimant to establish a prima facie case of infringement, but if the claimant did so, the burden shifted to the defendant to prove not merely honest concurrent use, but also that there was no adverse effect on any of the functions of the trade mark. As for the relevant date, she submitted that, where honest concurrent use was relied upon, assessment was not confined to the date of first use of the infringing sign, but extended to later dates, particularly the date of the claim form. I agree that the burden of proof and the relevant date are important questions, but I think that one needs to start by determining the legal relevance of honest concurrent use.
115. In my judgment the judge was correct that honest concurrent use is not a separate defence, but a factor in the infringement analysis. As discussed above, neither the EUTM Regulation nor Directive 2015/2436 nor the 1994 Act provide for a defence to infringement on the ground of honest concurrent use. Even if it were open to an English court to read such a defence into the 1994 Act as a matter of domestic law, which may be doubted, it is clear that this is not permissible as a matter of EU law (and the Defendants have not suggested that, with respect to acts committed since 31 December 2020, this is now possible as a result of Brexit despite the heavy imprint of retained EU law in this field). In the case of claims for infringement under section 10(1)/Article 9(2)(a)/Article 10(2)(a), honest concurrent use may be relied upon by the defendant as sufficiently rebutting the presumption that there is a likelihood of confusion. In the case of claims for infringement under section 10(2)/Article 9(2)(b)/Article 10(2)(b), the relevance of honest concurrent use lies in assessing whether there is a sufficient likelihood of confusion in the first place having regard to the difference in the guarantee of origin provided by the trade mark and the sign. Similarly, in the case of claims for infringement under section 10(3)/Article 9(2)(c)/Article 10(2)(c), the relevance of honest concurrent use lies in assessing the existence of one of the specified kinds of injury having regard to that difference.
116. Nevertheless, I think that counsel for the Defendants was correct to accept that, once the claimant has established a prima facie case of infringement, the burden shifts to the defendant to establish that, by virtue of its honest concurrent use, there is nevertheless no adverse effect on any of the functions of the trade mark. This is perhaps most easily seen in a section 10(2) case, where establishing a prime facie case of likelihood of confusion puts the claimant in the same position as the claimant in a section 10(1) case who benefits from a rebuttable presumption of a likelihood of confusion. The way in which this works in practice will become clearer when I come to consider the present case.
117. So far as the date of assessment is concerned, as discussed above the general rule is that the relevant date is the date when use of the sign complained of commenced. It is not the law, however, that that is a once-and-for-all assessment. Counsel for the Claimants accepted during the course of argument that, in theory, a use which was initially infringing could eventually cease to be infringing if the trade mark proprietor took no action, there was substantial parallel trade for a long period, and as a result the trade mark and the sign came to be understood by the relevant class of consumers as denoting different trade origins. In my judgment he was right to do so for the reasons discussed above. I do not think that the judge ruled out this possibility in saying that it would be "a very unusual case". It follows that it is not necessary for the use complained of to have started before the trade mark was registered, and again I do not think that the judge held that it was.
118. In the present case the judge took the conventional course of considering whether there was a likelihood of confusion as at March/April 2011, March 2015 and January 2017 taking into account the available evidence as to the existence or otherwise of actual confusion after those dates. He was correct to do so, because, as is well established, evidence which post-dates the relevant date is capable of throwing light backwards on the position as at the relevant date. Although it would, in principle, be possible to conclude that, while there was a likelihood of confusion as at April/May 2011, there was no such likelihood by the time of the trial, the judge's analysis eliminates that possibility.
119. That is not the end of the Defendants' case, however. As discussed above, honest concurrent use may lead to the conclusion that there has been no infringement, even though there is a small level of actual confusion between the trade mark and the sign, if most of the relevant class of consumers have come to understand that the trade mark and the sign denote different trade origins.
120. Counsel for the Claimants submitted that, even if the Defendants' concurrent use of the signs complained of had been honest, there was no evidence that most of the relevant class of consumers had been educated to understand that MUZMATCH was unrelated to MATCH. It is fair to say, however, that there does not appear to have been any positive evidence of such consumer education in either Budweiser or IPC v Media 10. Rather, the conclusion that most consumers had learnt that "Budweiser" denoted two different beers and that "Ideal Home" denoted two different businesses was a matter of judicial inference from many years' parallel trade on a substantial scale.
121. Although the judge did not in terms ask himself whether most of the relevant class of consumers had been educated to understand that MUZMATCH was unrelated to MATCH, I consider that it is clear from his analysis of the likelihood of confusion that, had he posed himself that question, his answer would have been in the negative. Rather, the essence of his analysis was that it was likely that a substantial proportion of the relevant class of consumers encountering MUZMATCH would assume that it was related to MATCH in the manner he described at [94] at all relevant dates even though almost no such confusion had been detected over the period from March/April 2011 to January 2022. Given that I have concluded that he was entitled to reach that conclusion, it follows that he was also entitled to conclude that the Defendants' concurrent use did not demonstrate otherwise. Unlike cases such as Budweiser and IPC v Media 10, this is not a case where the length and scale of the concurrent use compels the court to infer that most consumers have learnt the difference between the two marks. That conclusion is equally applicable to the claim for passing off. Still less did the Defendants' concurrent use demonstrate that their use of MUZMATCH did not take unfair advantage of the reputation of the Trade Marks, given that that finding was premised upon there not being a likelihood of confusion.
122. In those circumstances it is unnecessary to consider the Claimants' contention that the Defendants' use was not honest. I will therefore confine myself to commenting briefly upon three points. The first is that counsel for the Defendants submitted that Henry Carr J had been correct to conclude in Victoria Plum that the analysis in cases such as Hotel Cipriani and Samuel Smith required a degree of adaptation in this context. I did not understand counsel for the Claimants to dispute this, but I would wish to leave open for further consideration whether it is sufficient, as Henry Carr J might be interpreted as having said, that the defendant has not taken any steps which exacerbate the level of confusion beyond that which is inevitable and thus has not encroached on the claimant's goodwill. Secondly, the Claimants relied upon the fact there is no evidence that Mr Younas carried out a trade mark search in April/May 2011. I agree with the Claimants that this is a relevant factor to take into account, but I adhere to the view that I expressed in Walton at [214] that a mere failure to carry out a search is not sufficient to negate honest concurrent use. Lastly, the Claimants also relied on the Defendants' admitted infringement of the Tinder Marks. In my judgment that is irrelevant to the honesty of their use of MUZMATCH: see Samuel Smith at [119]-[120].

Conclusion

123. For the reasons given above I would dismiss this appeal. I would add that this case is yet another illustration of the enduring truth of Jacob LJ's observation in Phones 4U at [8]:

"Like gardens trade mark cases always get worse with neglect – even if rights are not actually lost, delay is apt to turn what would be over in a few weeks by a quick application into a mini State Trial."

Lord Justice Nugee:

124. I agree.

The Lord Chief Justice:

125. I also agree.