Birlea Furniture Ltd v Platinum Enterprise (UK) Ltd & Anor [2018] EWHC 26 (IPEC) (11 January 2018)

IN THE HIGH COURT OF JUSTICE
BUSINESS AND PROPERTY COURTS OF ENGLAND AND WALES
INTELLECTUAL PROPERTY ENTERPRISE COURT
COMMUNITY TRADE MARK COURT

Her Honour Judge Melissa Clarke:

INTRODUCTION

1. This is a trade mark infringement claim arising out of the sale of beds on Amazon.co.uk (“Amazon”). The Claimant is the registered proprietor of the EU trade mark no.11644416 “BIRLEA” filed on 23 April 2013 and registered on 23 August 2013 for, inter alia, furniture and beds in Class 20 (“the Mark”). It is also an importer and wholesaler of furniture, including beds. It sells to companies who then re-sell to the retail market. One of the Claimant’s current customers is Amazon. For that purpose it is a vendor seller on Amazon.
   
2. The First Defendant is a trading company that imports and sells furniture, including beds. It trades as ‘Bodyease’ but sells beds online, including on Amazon, using the brand ‘bedzonline’. The Second Defendant is an employee, shareholder and director of the First Defendant. It is relevant to note that from April 2010 to June 2013 the Claimant used to supply its products, including a specific style of metal daybeds called ‘Torino’ daybeds, to the First Defendant for re-sale online. The trading relationship between them broke down after that date.
   
3. The Claimant claims that the First Defendant infringed the Mark pursuant to Article 9(1)(a) of the EU Trade Mark Regulation 207/2009 (“2009 EUTMR”) / Article 9(2)(a) of the EU Trade Mark Regulation 2017/1001 (“EUTMR”) by advertising and selling metal daybeds through Amazon via two specific listings which use the Mark. One listing has the unique Amazon identification number (“ASIN number”) B00FS3D4G and relates to Torino daybeds in cream (the “D4G Listing”), and the other has the unique ASIN number B00BFS3D64 and relates to Torino daybeds in black (“D64 Listing”). I shall refer to the D4G Listing and D64 Listing together as “the Listings”. The Claimant claims that the Second Defendant is liable for the same infringing acts as a joint tortfeasor with the First Defendant.
   

THE PARTIES’ POSITIONS AT TRIAL

4. The Claimant’s case is that:
   

i) the Listings were set up by the Claimant, and each use the Mark in a number of places, namely: in the title (e.g. ‘Birlea Torino Single Metal Daybed, Black’); in the identification of the brand in the sub-title ‘by Birlea’; and in the ‘from the manufacturer’ section of the listing page;

ii) the First Defendant has used the Listings to advertise and sell Torino daybeds identical to the beds of the Claimant, and also to sell some units of a different style of bed called ‘Essina’ by fulfilling ‘Torino’ orders with ‘Essina’ beds (for example, it is not disputed that the Claimant’s test purchase of a Torino daybed from the First Defendant using the D64 Listing was fulfilled by the supply of an Essina bed);

iii) the name Birlea has been used on the Listings from before the time that the First Defendant began to use the Listings to advertise and sell beds, and at all times afterwards. Accordingly, all of the First Defendant’s use of the Listings to advertise and sell beds is use in the course of trade of the sign BIRLEA, which is identical to the Mark, in respect of beds which are identical to the goods covered by the registration of the Mark and so are infringing;

iv) alternatively, if the sign Birlea was added to the Listings only at some later date (which the Claimant denies), the First Defendant has infringed the Mark by use of the Listings to sell its beds from that later date;

v) the Second Defendant is jointly and severally liable for any trade mark infringement of the First Defendant as joint tortfeasor.

5. The Defendants accept that the First Defendant used the Listings to advertise and sell beds. They do not deny that the sign complained of, if used, infringes the Mark. They do not challenge the validity of the Mark or seek the revocation of it. However, they dispute that the First Defendant ‘used’ the sign complained of for the purposes of Article 9(1)(a)/Article 9(2)(a) of 2009 EUTMR/EUTMR for the following reasons:
   

i) when the First Defendant first listed beds for sale under the Listings, the Listings did not include the name Birlea, rather they were ‘generic’ listings for a Torino daybed in cream or black;

ii) accordingly when they added the First Defendant as a seller to the Listings, they did so correctly as they were selling the identical goods to those listed on the Listings, which did not contain the Birlea name, so they cannot have infringed the Mark;

iii) the Defendants aver that the name Birlea was added to the Listings on around 20 March 2015. After this date, the Defendants’ case is that they did not become aware that the sign complained of had been added to the Listings until receipt of the Claimant’s letter before action on 6 May 2015, at which point they promptly stopped advertising and selling beds from the Listings. It is common ground that the last sale of beds by the First Defendant from the Listings was on 5 May 2015;

iv) The Defendants seek to make a novel argument: they accept that there is no intentional element of trade mark infringement under the 2009 EUTMR/EUTMR, but submit that despite this, they should only be liable for any infringement after the date on which they became aware of the presence of the sign Birlea on the Listings, as until they are aware of it, they have not ‘used’ it;

v) the Second Defendant denies any liability as a joint tortfeasor.

6. Accordingly the only areas of dispute that I have to determine are: (i) that of use of the sign complained of; and (ii) whether the Second Defendant is jointly liable for any infringement by the First Defendant. The Defendants now accept that if I find that the Birlea name was on the Listings at the time the First Defendant started using them to advertise and sell the beds, the First Defendant is liable for infringement of the Mark as claimed.
   
7. On the Defendants’ novel argument, the Claimant submits that as a matter of law there is no mental element, or requirement for knowledge of use of the sign complained of, for the purposes of trade mark infringement, so lack of knowledge affords no defence. In any event, it says that the Defendants knew of the use of the Birlea name on the Listings by a number of alternative means: (i) sales records created by Amazon and posted to the First Defendant’s seller account each time a sale was made from the Listings; (ii) emails sent by Amazon to the Second Defendant notifying him each time a sale was made from the Listings; (iii) an email sent by the Claimant to the First Defendant on 4 September 2014 complaining of their use of the Listings; and (iv) a letter before action sent by the Claimant to the First Defendant on 6 May 2015.
   
8. The Claimant seeks: (i) a pan-European injunction to restrain the Defendant from infringing the Mark; (ii) an order for delivery up or destruction by the Defendants of all infringing materials; (iii) at its election, an inquiry as to damages suffered by the Claimant, or an account of profits accrued by the Defendant arising out of such infringements; (iv) an order for the dissemination and publication of the judgment at the Defendants’ expense; and (v) costs and interest.
   
9. The Defendants seek dismissal of the claims against them. Alternatively, if the court finds that the First Defendant has infringed the Mark, the Second Defendant denies any liability as a joint tortfeasor and seeks dismissal of the claim against him.
   
10. HHJ Hacon by an order of 3 February 2017 directed a split trial of the matter and accordingly I am only concerned with liability in this judgment.
    

PROCEEDINGS

11. It is worth spending some time examining the development of the Defendants’ case through the pleadings and witness statements, as it has changed materially and significantly over time. As is usual in IPEC, a case management direction was given that pleadings should stand as evidence and each was signed with a statement of truth.
    
12. The Claim was originally issued as against the First Defendant only. Mr Jamil Baig, a director of the First Defendant, signed the Defence on 24 July 2015. It stated that:
    

i) the First Defendant had purchased beds directly from Profitlane Sdm Bhd in Malaysia, which stated that they were the same beds it supplied directly to the Claimant and were ‘ex-stock’. The First Defendant now accepts that is incorrect, and Profitlane was not a supplier to the Claimant.

ii) It bought three container-loads of beds being 1075 units in total, but did not sell all of them, as some were damaged or otherwise unsaleable. It could not say how many it had sold. The First Defendant now accepts that it bought six container-loads of Torino beds between June 2012 and July 2014, and sold 1696 units.

iii) The First Defendant listed them on Amazon, but did not advertise them as Birlea beds, as it was Amazon who tagged their beds to the existing Listings. The Second Defendant admits, and the First Defendant now accepts, that the Second Defendant added the beds to the Listings.

iv) In any event the First Defendant did not believe that the listing originally contained the name Birlea, but only referred to ‘Torino’ day beds. The Defendants have maintained this position.

v) The First Defendant admits that it supplied an ‘Essina’ bed to the Claimant when the Claimant placed a test purchase for a Torino bed, but “believes that this was an isolated incident and resulted from a “picking error” in the [First] Defendant’s warehouse.” Counsel for the Defendants in closing accepted that the evidence shows that a number of Essira beds were supplied in fulfilment of Torino orders.

13. The Second Defendant, as a director of the First Defendant, signed the First Defendant’s Replies to the Claimant’s Request for Further Information on 17 November 2015. The Claimant had asked for copies of all communications from Amazon which recorded or related to the Listings. The First Defendant answered: “The only communication the [First] Defendant received from Amazon was the standard pro forma email thanking the [First] Defendant for placing its entry and confirming that it had been successfully entered on the Amazon site. The [First] Defendant did not keep, and does not keep, copies of such emails”. The Defendants now accept that for each sale of a bed, Amazon sends two communications to the First Defendant: (i) a sales record to the First Defendant’s Amazon seller account; and (ii) an email notification of the sale to the Second Defendant’s email account. Accordingly for the 1696 sales of beds on the Listings now admitted by the Defendants, they would have received 3392 communications.
    
14. In the Second Defendant’s first witness statement of 27 April 2016, he identifies the D4G Listing and the D64 Listing as those he used to list the First Defendant’s beds for sale. He states in paragraph 25 that “the [First] Defendant has not sold any cream daybeds, and we did not advertise any for sale”. He goes on to refer to a printout which he states shows “all “traffic” on behalf of the [First] Defendant” for the D4G Listing, stating “It confirms that the entry is not found and as such that the [First] Defendant has never sold any cream [Torino] daybeds”. He repeats at paragraph 34 that “The [First] Defendant never offered any cream beds for sale under [the D4G Listing]”. The Defendants now accept that the printout referred to was for a search conducted by the Second Defendant in which he replaced each zero in the D4G Listing ASIN number with the letter ‘O’, which is why no results were returned. They now accept that: (i) the First Defendant sold 1145 cream beds on the D4G Listing; (ii) the Second Defendant knew that the First Defendant had sold cream beds at the time he made the first witness statement; and (iii) this witness statement was accordingly incorrect and misleading. The Second Defendant further stated that the First Defendant using the D64 Listing sold 303 black beds. The Defendants now accept this was incorrect, because they had searched under a limited date range from 14 February 2014 to 24 May 2016. They accept that the correct number of black units sold from the D64 Listing is 551.
    
15. Following a case management conference on 17 May 2016, HHJ Hacon permitted amendment of the Claim Form and Particulars of Claim to add Mr Raheel Baig as Second Defendant and to include a claim against him as an alleged joint tortfeasor with the First Defendant. He ordered the Defendants to file and serve a further response to the Claimant’s request for communications with Amazon. Both Mr Jamil Baig and the Second Defendant signed the Defendants’ Additional Replies to the Claimant’s Request for Further Information on 25 May 2016 and stated: “The Defendants received standard email notifications of orders received from customers from Amazon on the diverse dates when such orders were placed. The Defendants did not retain these at the time they were received. The Defendants have accessed their account and as such been able to obtain a historical list of the 293 order[s] received, full redacted copies of which accompany this reply.” In the Amended Defence of the First and Second Defendants, signed by both Messrs Baig on 17 June 2016, they attached a print-out from the First Defendant’s Amazon seller account evidencing those 293 orders. Again, the Defendants now accept that they received 1696 orders, not 293, and that the print-out related only to a sub-set of orders of black beds on the D64 Listing.
    
16. Following a meeting between the parties at the offices of the Defendant’s solicitors on 6 December 2016, where Mr Lester of the Claimant’s solicitors was permitted to carry out a search of the First Defendant’s Amazon seller account and found the 1696 sales of beds now accepted by the Defendants, the Second Defendant filed a second witness statement on 2 February 2017 in which he provided an explanation for why he had failed to find, and so disclose, the full sales history. This explanation was the subject of some cross-examination at trial. The First Defendant disclosed the Amazon sales records for each sale after this meeting. The email notifications sent by Amazon to the Second Defendant for each sale have never been disclosed and it is the Second Defendant’s evidence that they were lost when a server was replaced and no back-up kept.
    
17. The Defendants filed a Re-Amended Defence on 8 February 2017 in which they accepted they had sold 1145 cream beds and 551 black beds, being 1696 in total. It stated that the listing used to sell black beds was the D64 Listing, and the listing used to sell cream beds was a listing with an ASIN code B00E81WJOY (“the JOY ASIN”). I believe they now accept that the JOY ASIN was a parent ASIN for the D64 listing, and the cream beds were sold using the D4G Listing. If not, I find as a fact that they were, as is clear from the sales orders for cream beds which all provide the D4G Listing ASIN number.
    
18. The Second Defendant filed a third witness statement on 15 May 2017. In it he stated that whatever title is given to a sales listing on Amazon can be changed, but that the creator of the listing is the only one who can change it and that change is not notified to any other seller This is not disputed by the Claimants. He said “Further and fundamentally, once the change has been made this appears to apply retrospectively on Amazon so any information that is obtained, or copies of documents generated or provided, refer only to the name of the listing as it is in the present.” The Defendant no longer pursues this theory, as explained further in my consideration of issue (ii) below. The Second Defendant acknowledged the inaccurate and misleading statement in his first witness statement that the First Defendant had sold no cream beds from the Listings, and provided further explanations for why he had been unable to find and fully disclose sales using the First Defendant’s Amazon seller account. However he continues to state in the third witness statement that a search under the D4G Listing ASIN number produces no sales results, when Mr Lester of the Claimant’s solicitor was able to produce productive searches on that ASIN number on 6 December 2016.
    

ISSUES

19. The following issues for trial on liability were identified by HHJ Hacon in an Order of 3 February 2017 which followed a case management conference of the same date:
    

i) When did the Defendants first advertise Torino day beds (“the beds”) on Amazon?

ii) When the Defendants first listed beds on Amazon, did they do so under the Birlea name?

iii) Were the Defendants notified of the fact that beds were listed under the name ‘Birlea” as a consequence of

a) The order details that were provided by Amazon as a consequence of each sale of the beds complained of?

b) The receipt of and response to the email from Dan Ambalavanar of 4 September 2014?

c) The receipt of the letter?

iv) When were the Defendants notified that they were advertising beds under the name “Birlea”?

v) Did the Defendants continue to advertise beds under the name “Birlea” after they became aware of such use?

vi) Did the Defendants purchase beds from suppliers other than Profitlane Sdm Bhd?

vii) How many beds did the Defendants sell under the name “Birlea”?

viii) Did the First Defendant infringe the Claimant’s EUTM?

ix) If so, is the Second Defendant also liable for any such infringements?

20. Issues (vi) and (vii) above are no longer in dispute. The Claimant now accepts that the Defendants purchased all relevant beds complained of from Profitlane Sdm Bhd, and it is now agreed that the First Defendant sold 1696 beds through use of the Listings. The sales records for those sales have been disclosed and can be found in the trial bundles.
    
21. Mr Thomas St Quintin appears for the Claimant and Ms Charlotte Blythe for the Defendants. I thank them for their comprehensive skeleton arguments and attractive submissions.
    

WITNESSES

22. I heard the following witnesses at trial: Mr Daniel Ambalavanar, Mr James Sonley and Mr Stuart Lester for the Claimant; and the Second Defendant (known as Mr Raheel Baig) and Mr Mohammed Jamil Baig (known as Mr Jamil Baig) for the Defendants.
    
23. Mr Ambalavanar was the Next Day and E-Commerce Manager of the Claimant from August 2013 to November 2016. He had worked for the Claimant in other capacities since 2009. He moved onto other employment in a different field in November 2016 and so I consider him to be in some measure an independent witness. Mr Ambalavanar filed three witness statements dated 28 September 2015, 29 January 2016 and 5 June 2017. He was not responsible for creating the Listings, which he believes were created by his predecessor in the role of Next Day and E-Commerce Manager, Danni Sadler, but he was responsible for two rounds of changes to the Listings and he gives evidence about these changes and about the Claimant’s Amazon listings policies generally. His evidence was that he believes that the Claimant created the Listings, as only it has the authority to request a change to the listing titles, which is a right granted by Amazon only to the listing creator. He agrees with Mr Sonley that the Torino daybed Listings appear to have been available on Amazon since 11 October 2010, and is confident that the Listings contained the Birlea name in the title from at least 1 February 2013, and probably earlier. To support this, he relies on a search for ‘Birlea Torino’ on Amazon, which brings up each of the Listings with ‘1 February 2013’ against the product description. Mr Ambalavanar’s evidence is that he believes each of the Listings would already have included reference to the Birlea name at this time, but on 1 February 2013 the titles were amended to include more product information following a meeting with Amazon in which advice about optimising search rankings was given. He relies on a document produced by the Claimant’s Amazon Vendor customer contact, Salem Marafi, which is a monthly record of the titles of the Listings between February 2013 and January 2015. That shows that from February 2013 to April 2013 the listing titles were “Birlea Furniture Metal Torino Day Bed, 3ft, Black [or Cream]” and from May 2013 to January 2015 they were “Birlea Torino 3ft Daybed, Black [or Cream]”. In addition, in cross-examination he stated that in all his years of working for the Claimant he never saw an Amazon listing of a Birlea bed that did not have the name ‘Birlea’ in the title.
    
24. Mr Sonley is the Head of Online Sales and Marketplaces for Right Deals UK Limited, which is a customer of the Claimant. Right Deals UK Limited sells the Claimant’s products and other products on Amazon and on other online channels and has been an authorised seller of the Claimant’s Torino beds on Amazon via the Listings since 27 February 2014. Mr Sonley is, in my judgment, also an independent witness as he has no benefit to gain from this litigation. He filed three witness statements dated 29 January 2016, 5 June 2017 and 7 July 2017. Mr Sonley said that in his role he had sold over 20,000 product lines on Amazon with a sales value of about £5m per annum over 10 years and considered himself highly knowledgeable about Amazon’s listing practices. He exhibits screenshots which appear to show that the Listings were first available on Amazon on 11 October 2010. He disputes the theory of the Defendants that past sales orders are retrospectively amended by Amazon if the listing title is changed. In his witness statement he stated: “The title [of sales orders] is exactly as shown on the Amazon product listing page at the time of the order, as the data is pulled from the title on the product listing page. The title detailed in the Orders section is not updated retrospectively even if a title of a listing subsequently changes. It remains as it was when the order was made.” In cross-examination on this point he said that he was “100% confident” that the sales order confirmation generated by Amazon took the title and description of the product sold from the Amazon page as seen by the customer at the time the order was placed, with a 15 minute tolerance.
    
25. Mr Lester filed one witness statement dated 5 June 2017. He is a solicitor employed by Freeths LLP, the Claimant’s solicitors. He attended an inspection meeting as part of the disclosure process at the Defendant’s solicitors’ offices on 6 December 2016 and his witness statement describes this meeting. In particular he describes how the Second Defendant purported to search the ‘Reports’ section of the First Defendant’s Amazon seller account, to return no results. Mr Lester then asked for, and was given, control of the laptop and he searched the ‘Orders’ section of the First Defendant’s Amazon seller account to find 1696 sales of cream and black beds by the First Defendant via the Listings. Mr Lester’s evidence is that while watching the Second Defendant log into and navigate around the First Defendant’s Amazon seller account, he considered that he appeared adept at using the Amazon platform; that the ‘Orders’ section of the Amazon seller account, which the Second Defendant said he had never searched, was the most obvious and relevant one to search to locate data relating to the history of previous orders of the six main headings on the account; and that he believed that the Second Defendant had carried out previous searches with mistyped ASIN numbers (replacing zeroes with Os) and with limited date ranges in order to hide sales from the Claimant.
    
26. I consider all three of the Claimant’s witnesses to be good witnesses who provided straightforward, credible and reliable evidence honestly and carefully. I have no doubt that all of them came to court to assist it to the best of their abilities and recollection.
    
27. Mr Jamil Baig filed two witness statements dated 28 April 2016 and 9 May 2017. He is a director of the First Defendant and the brother of the Second Defendant. It became clear in cross-examination, and he accepted, that almost all the evidence he gave was not evidence from his own knowledge, but rather repetition of what his brother, the Second Defendant, had told him. Accordingly that evidence does not assist me and I give it no weight. In respect of the first-hand evidence he does give, it is of limited relevance. I do not believe he came to court to give dishonest evidence, but I cannot ignore that he signed, with a statement of truth, pleadings containing material inaccuracies. This inevitably affects my view of his overall credibility and I treat his evidence with some caution.
    
28. The Second Defendant filed four witness statements dated 29 April 2016, 2 February 2017, 15 May 2017 and 30 June 2017. Mr St Quintin submits that he was an appalling witness who lied multiple times in preparation for trial, including signing witness statements and pleadings containing lies with statements of truth, and who introduced more lies into his oral evidence in an attempt to explain away his earlier dishonesty. Ms Blythe submits that mistakes in the Second Defendant’s earlier witness statements were inadvertent, not deliberate, and he should be given credit for providing explanations for those mistakes in his later witness statements.
    
29. I found the Second Defendant to be a very unsatisfactory witness. I consider that his credibility is very materially damaged by the false statements made in his first witness statement that the First Defendant had not sold or advertised for sale any cream Torino beds on the D4G Listing, when he knew that it had both bought cream Torino beds, and sold them via the Listings. His explanation for giving this untruthful evidence was also unsatisfactory. If it was true that he couldn’t find evidence of the orders on the Amazon seller account, when he knew such sales had been made, the proper response was not to lie about it, but to search in a different way such that the evidence could be found, or to provide truthful evidence that he knew of cream bed sales but could not locate the sales evidence. His evidence that he thought perhaps those sales had been made from different listings I consider unbelievable. There is no evidence that he carried out any investigation into other listings to see if the beds had been sold there. I do not believe he can lose sight of over 1000 bed sales in this way. I also consider that he gave further false evidence in his third witness statement by repeatedly stating that a search under the D4G Listing ASIN number shows no evidence of sales of cream beds. Mr Lester says that it does, and he did it in the 6 December meeting, and I have the sales order confirmations before me to support that assertion. The Second Defendant watched him do it and accepted in cross-examination that Mr Lester’s account of that meeting was accurate. I accept Mr Lester’s evidence and find the Second Defendant’s evidence on this point to be untruthful.
    
30. I am also concerned about the fact that the searches of the First Defendant’s Amazon seller account that the Second Defendant initially carried out and disclosed to the Claimant were flawed in a number of different ways. I accept it is possible for someone to get confused between zeroes and Os when typing out an alphanumerical product code. However, the Second Defendant should have questioned why a search of the D4G Listing ASIN number returned no sales results when he searched cream beds, in light of his admission that he knew the First Defendant had sold some. The fact that he did not, and went on to carry out and disclose two more flawed searches (the search of the D64 Listing with a highly reduced date range, and a search described in Mr Lester’s first witness statement which on close inspection has had a bar-code product filter applied to it so that no results are found by a search of ASIN numbers), and the fact that he gave false evidence that the First Defendant had bought 3 container-loads of beds when it had bought 6, leads me to the conclusion on the balance of probabilities, in circumstances where his credibility has already been impaired by admittedly false statements, that such searches were part of a campaign of obfuscation entered into by the Second Defendant to hide from the Claimant the true extent of the First Defendant’s sales using the Listings. I accept Mr St Quintin’s submissions to that effect.
    
31. This conclusion is partly based on the Second Defendant’s performance in cross-examination, in which I found him similarly obfuscatory. For example, he was asked why he searched the D64 Listing with such a reduced date range. He said that was because Amazon had told him he could only search back for a maximum of two years. Mr St Quintin showed him that the date range he had chosen was from 14 February 2014 to 24 May 2016, i.e. more than two years and encompassing a period of over a year after the last sale of product by the First Defendant on the Listings. The Second Defendant accepted it was more than two years, but said that “it was done over the phone with the Amazon representative”. He was asked why he had searched from February 2014. He said “because that was when the Amazon listing was created, February 2013, so that’s when I searched”. When Mr St Quintin pointed out once again that he had searched from February 2014, not February 2013, he responded “Well, it could have been a mistake.”
    
32. Watching Mr St Quintin’s efforts to cross-examine the Second Defendant was like watching him try to nail jelly to a wall. I mean Mr St Quintin no disrespect by that analogy. The nature of the Second Defendant’s slippery gyrations in responding to questions was such that even Mr St Quintin’s skilful attempts to get him to provide (and stick to) straight answers, were doomed to failure. As the evidence of each of the Second Defendant’s witness statements was materially amended and contradicted by the next, so did he attempt to offer new evidence in cross-examination, and then resile from it, eventually admitting the possibility that almost all the evidence he had given was wrong. Even when Mr St Quintin put to him that he was prepared to say anything at all if he thought it might help his defence, the Second Defendant was unable to give a definite answer, saying “I don’t think it is true”. I think it is true. In my judgment the Second Defendant would say anything at all, without regard to truth, if he thought it would assist his defence. I treat his evidence with very great caution and do not accept it unless other credible evidence or the inherent probabilities support it.
    

THE LAW

33. The alleged infringements cover a time period in which the 2009 EUTMR was amended by Regulation 2015/2424 (the “Amending Regulation”) with effect from 23 March 2016, and then later codified as the EUTMR. The Amending Regulation caused a renumbering of Article 9 but no substantive changes to it. Accordingly the claim is brought under Article 9(1)(a) of the 2009 EUTMR in respect of alleged infringements up to 22 March 2016 and Article 9(2)(a) of 2009 EUTMR as amended by the Amending Regulation, and of the codified EUTMR in respect of later alleged infringements. Article 9 EUTMR provides so far as is relevant:
    

“Rights conferred by an EU trade mark

1. The registration of an EU trade mark shall confer on the proprietor exclusive rights therein.

2. Without prejudice to the rights of proprietors acquired before the filing date or the priority date of the EU trade mark, the proprietor of that EU 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 EU trade mark and is used in relation to goods or services which are identical with those for which the EU trade mark is registered;

(b) …

(c) …

3. The following, in particular, may be prohibited under paragraph 2:

(a) …

(b) offering the goods, putting them on the market, or stocking them for those purposes under the sign;

(c)…

(d)…

(e) using the sign on business papers and in advertising;

(f)…

Use of a sign in the course of trade

34. It is accepted by the Defendants that use by the First Defendant of the Birlea name on the Listings to sell the beds would be use in the course of trade of a sign identical with the Mark in relation to goods identical for those for which it is registered. The issue is whether the First Defendant has used the Birlea sign. Any use of a sign in the context of commercial activity with a view to economic advantage and not as a private matter, no matter how modest, will be sufficient to satisfy the requirement for ‘use in the course of trade’ (per C-206/01 Arsenal Football Club plc. v Matthew Reed EC:C:2002:651 at paragraph 40).
    
35. Use in relation to goods or services was considered by the CJEU in C-17/06 Céline SARL v Céline SA EU:C:2007:497 at paragraphs 22 – 23, where it was held that there is use “in relation to goods” where a third party affixes the sign constituting his company name, trade name or shop name to the goods which he markets and, even where the sign is not affixed, where the sign is used in such a way that a link is established between the sign which constitutes the company, trade or shop name of the third party and the goods marketed or the services provided by the third party.
36. The CJEU made specific comment on the use of signs in the context of online marketplaces in L’Oreal v eBay Case C-324/09, EU:C:2011:474, [2011] ETMR 52, at paragraphs [98] to [105]. In particular, at paragraphs 102 and 103 it stated:

102 If a sign identical with, or similar to, the proprietor’s trade mark is to be ‘used” within the meaning of art.5 of Directive 89/104 and art.9 of Regulation 40/94, by a third party, that implies, at the very least, that that third party uses the sign in its own commercial communication. Insofar as that third party provides a service consisting in enabling its customers to display on its website, in the course of their commercial activities such as their offers for sale, signs corresponding to trade marks, it does not itself use those signs within the meaning of that EU legislation (see to that effect, Google France [2010] ETMR 30 at [56] and [57]).

103 As was stated, inter alia by the UK Government and the Commission at the hearing and by the Advocate General at points AG119 and AG120 of his Opinion, it follows that the use of signs identical with or similar to trade marks in offers for sale displayed on an online marketplace is made by the sellers who are customers of the operator of that marketplace and not by that operator itself.”

37. Accordingly, in the context of registered trade marks, I accept Mr St Quintin’s submission that as a matter of law it is the seller on an online marketplace who makes use of the signs displayed on a product listing within it, because the marketplace provides a means for the seller to make its own commercial communications. There is no requirement for knowledge of use of the sign for infringement to be made out, only that there is in fact actual use in such a way that a link is established between the sign complained of and the goods marketed by the seller.
    

ANALYSIS OF ISSUES

Issue (i) – When did the Defendants first advertise Torino day beds (“the beds”) on Amazon?

38. The Defence does not provide these dates. Nor does the Response to the Claimant’s Request for Further Information, although the Claimant asked the question. The Second Defendant’s witness statements give different alternatives. In his first witness statement he says that the First Defendant first used the Listings in late September/Early October 2014. His second witness statement states that the First Defendant first used the D4G Listing in February 2013 and the D64 Listing in March 2013. His third witness statement puts both at March 2013. Mr St Quintin in closing asked me to find that the First Defendant advertised on the Listing shortly before the first sales, which were 27 February 2013 for the D4G Listing and early March 2013 for the D64 Listing. Ms Blythe in her closing submissions for the Defendants accepted that this was a reasonable inference for the court to make. I so find.
    

Issue (ii) – When the Defendants first listed beds on Amazon, did they do so under the Birlea name?

39. Ms Blythe submits for the Defendants that the only direct evidence I have on this point comes from the Second Defendant. She asks me to accept his evidence that at the time the First Defendant was added to the Listings, the Birlea name did not appear on either of them, which she submits is the Defendants’ consistent position through the Defence and its amendments, the Replies to the Claimant’s Request for Information, and the Second Defendant’s witness statements. However, I have already found that I cannot accept the Second Defendant’s evidence unless other credible evidence or the inherent probabilities support it.
    
40. The sales records for the very first beds sold under each of the D4G Listing and the D64 Listing both show the name Birlea in the title. The Claimant’s case is that those sales records reflect the titles of the Listings as at the date of sale, and accordingly the court can be confident that the name Birlea was on the Listings at least by the time of the first sales. It relies on the evidence of Mr Ambalavanar, Mr Sonley and the spreadsheet produced by Mr Marafi attached to the witness statement of Mr Ambalavanar to support its case. It would have wished also to rely on the sales notification emails to the Second Defendant, which it believes would also have shown the use of Birlea in the title of the notification, but these have not been disclosed by the Defendants. Ms Blythe asks me to draw an adverse inference from the fact that the Claimant has not called the creator of the Listings, believed to be Ms Danni Sadler, to give evidence. I have heard that Ms Sadler no longer works at the Claimant. In light of the extensive documentary evidence supporting the Claimant’s case, I do not draw that adverse inference. The sales records are, in my judgment, compelling, but the Defendants dispute the Claimant’s case.
    
41. I have mentioned that the Second Defendant, in his third witness statement, posited a theory that the title shown on an accessed or printed past sales record was the title on the relevant product listing at the time that the sales record was accessed and/or printed, not that which was shown on the listing at the time the sales order was placed by the customer. In other words, Amazon would retrospectively change the title on the sales record to reflect whatever the current product listing showed. The inferences that could be drawn from this theory, if true, were therefore: (i) that sales records accessed later could not evidence what the title of a Listing was at the time that an order was placed; and (ii) the fact that the first and all subsequent sales records contained the Birlea name was not probative of the Birlea name being on the Listings at the time such sales were made.
    
42. However, a full copy of all of the sales records for each sale by the First Defendant on the Listings was printed out on a single day, being 8 December 2016. This was two days after the 6 December 2016 meeting in which Mr Lester discovered that those sale records existed. Leafing through that file, it can be seen that the title of both Listings has changed more than once on the face of those sales records. If the Second Defendant’s theory on wholesale retrospective substitution of the listing title at the date of printing was correct, those sale records should only show two titles – the title of the D4G Listing as at 8 December 2016 and the title of the D64 listing as at 8 December 2016. They do not.
    
43. Accordingly, the Defendants have now abandoned that theory of wholesale retrospective title change. They now seek to argue that Amazon executes a partial retrospective change to the title of sales orders which are accessed or printed later, by using the sales title of the Listing at the time the relevant order was placed, but inserting as a prefix the trade mark from the ‘brand’ field, if that has changed over time. So if the D64 Listing title in March 2013 when an order was placed was “3ft Torino Bed in Black” but the brand field left unpopulated, and the D64 Listing was changed in, say, 2015 to populate the brand field ‘by Birlea’, then the same March 2013 sales order if printed in December 2016 would refer to “Birlea 3ft Torino Bed in Black”, with the prefix ‘Birlea’ automatically inserted from the newly populated brand field. As part of this theory, the Second Defendant also stated in cross-examination that he believed the Claimant did not populate the brand fields of the Listings until 20 March 2015. There is no written or documentary evidence to support this belief. There is only the Second Defendant’s evidence that the name Birlea was not to be found anywhere on the Listings at the time he added the First Defendant as a seller. However in cross-examination, the Second Defendant suggested for the first time that it was not possible to populate the brand field on an Amazon listing except by use of a registered trade mark, for which the trade mark registration number must be provided. In my judgment this was one of the more obfuscating and confusing parts of his evidence. After some pressing questions by Mr St Quintin, he resiled from this evidence and adopted instead Mr Sonley’s position that anything could be put in the brand field, or it could be left blank.
    
44. The Defendants rely on several emails from Amazon to support their partial retrospective brand insertion theory, of which the following are of most relevance:
    

i) an email of 20 April 2016 from ‘Rich’ of Amazon Seller Support. He says “As we have discussed the below listings are created by Amazon on the same date which is 2013-02-14. The title provided are the first titles used when the listings were created. [D4G Listing] Torino Metal Single Day Bed Cream [D64 Listing] TORINO SINGLE BLACK METAL DAY BED”;

ii) an email of 24 May 2016 in which ‘Precy’ of Amazon provided the same information and added “And this is the new update of this ASINs: for [the D4G Listing] the new title is “Birlea Torino Metal Single Day Bed Cream” and for [the D64 Listing] the new title is “Birlea Torino 3ft Single Metal Daybed, Black”;

iii) an email of 7 February 2017 from Kris Linton of Amazon, in which he stated that “the title for [the Listings] were updated with the prefix “Birlea” on March 20 2015″;

iv) an email of 20 June 2017 from an employee of Amazon called ‘Loyiso’. She says, “The order information will pull through the information on the product detail page regardless of when the order was placed”. Although this email makes no reference at all to the current brand field being used and pulled through as a prefix to the old title listing, as the Defendants’ theory requires, the Second Defendant stated in cross-examination, for the first time, that he had had several telephone conversations with Amazon employees who had confirmed that was the case. These have never been mentioned before and in my judgment it is implausible that he would have failed to mention such telephone conversations, which go to the root of the Defendants’ defence at trial, if they had happened. I have no doubt that his evidence on this point was untruthful.

45. These emails each, apparently, follow telephone conversations with the relevant Amazon employees. Each contains a piece of information with no context as to the conversation which preceded it or the question which was asked by the Second Defendant to which the answer is being provided by the Amazon employees. I also cannot know who these employees are, what level of seniority, experience or knowledge they have, and whether their information is likely to be accurate or not. The email from ‘Rich’ is particularly confusing, as it gives a creation date for the Listings of 14 February 2013 when: (i) other evidence from Amazon exhibited by Mr Sonley, and endorsed by Mr Ambalavanar, shows the Listings in use from 2010 and (ii) customer reviews on the Listings go back to 2012. One of those reviews specifically refers to Birlea within the text of the review. For those reasons I consider on the balance of probabilities that ‘Rich’s’ email is not accurate as to the date of creation of the Listings. I accept Mr St Quintin’s submission that his email cannot tell the whole story, and indeed I find that none of them can. To my mind they raise more questions than they answer.
    
46. Taking all of the evidence into account I can not accept the Defendants’ theory of a partial retrospective insertion of the Birlea sign into later accessed or produced sales records for the following reasons:
    

i) I do not accept the Defendants’ submission that Rich’s and Mr Linton’s emails are evidence that there was no Birlea prefix on the Listings at all before 20 March 2015:

a) Rich’s email merely states that at the time the Listings were first created there was no reference to Birlea. The Claimant accepts that this may possibly have been the case at the point of creation of the Listings, which it puts in 2010 or earlier, but the issue I have to decide is whether the name Birlea was used at the time that the First Defendant began to use the Listings in February and March 2013;

b) Mr Linton’s email refers to a change in prefix to Birlea. Mr Ambalavanar has explained the change that he made to the title of the Listings from referring to ‘Birlea Furniture’ to ‘Birlea’ on 20 March 2013. That could be described as a change in prefix to Birlea, and as such does not contradict the email from Kris Linton, in my judgment. Mr Ambalavanar’s evidence is corroborated by the sales orders from that time, in relation to which all sales before that date refer to ‘Birlea Furniture’ and all sales after that date refer to ‘Birlea’. In my judgment it is also corroborated to some extent by the spreadsheet produced by Mr Marafi showing the title of the Listings by month (attached to Mr Ambalavanar’s witness statement), which shows a change in prefix from Birlea Furniture to Birlea from March to April 2013.

ii) If the Defendants’ theory of partial retrospective insertion of the Birlea sign is correct, that would mean that the Listings before the prefix change in March 2013 began with the word ‘Furniture’, which would be nonsensical (e.g. ‘Furniture 3ft Torino Bed in Black’). The Second Defendant accepted that difficulty in cross-examination.

iii) If the Defendants’ theory of partial retrospective insertion is correct, it should also apply to the Defendants’ own brand listings. It can be seen from the screen grab of the Defendants’ seller page obtained by Mr Lester on 6 December 2016 that there are a number of products sold by the First Defendant with ‘bedzonline’ in the listing title and ‘by bedzonline’ in the brand field, but the relevant sales records refer only to the title with a single reference to bedzonline. If the brand was retrospectively inserted, one might expect to see it twice, i.e. ‘bedzonline bedzonline metal daybed’. Similarly, the First Defendant sells generic mattresses. For example, one listing is for a 4ft Double Memory Foam Mattress ‘by bedzonline’. The sales record might be expected to show ‘bedzonline 4ft Double Memory Foam Mattress’ but it does not. It merely shows the listing title with no reference to the brand. The Second Defendant in cross-examination accepted that these examples were anomalous to the Defendants’ case.

iv) I do not accept the interpretation that the Defendants seek to give the email from Loyiso. As I have already stated, I do not consider that Loyiso’s statement that: “The order information will pull through the information on the product detail page regardless of when the order was placed” can sensibly be interpreted to support the partial retrospective brand insertion theory. It doesn’t refer to the brand, or any partial insertion. It is sufficiently vague and ambiguous that I consider that the meaning of her statement can only be accurately determined by asking her. Of course, she was not available at trial to be asked.

v) Mr Ambalavanar’s evidence that he has never known a Birlea bed to be sold on Amazon without the use of ‘Birlea’ in the title, and Mr Sonley’s evidence that he was “100% confident” that the sales order confirmation generated by Amazon took the title and description of the product sold from the Amazon page as seen by the customer at the time the order was placed, provide some additional support for my finding, in my judgment. I put it no higher than that.

vi) There is no other evidence to support the theory, save the Second Defendant’s evidence that the Listings did not contain the name Birlea at the time the First Defendant began to use them. My concerns about the Second Defendant’s credibility mean that I cannot accept his evidence in light of the weight of evidence against it.

vii) Finally, the theory of partial retrospective brand insertion is, in my judgment, inherently implausible. The Defendants have suggested no rational explanation for why Amazon should carry out such a retrospective change. Conversely, Mr Sonley for the Claimant provides suggestions as to why Amazon would not do such a thing, stating: “Firstly it would be detrimental for auditing purposes. However perhaps more importantly from Amazon’s perspective is that it would require an enormous investment of time, cost and resources to rewrite historical data relating to potentially billions of records on a rolling basis. This would be near impossible for them to achieve.” This is mere supposition on his part, albeit informed supposition, but I accept these as valid considerations, which go to the inherent implausibility of the theory.

47. Taking all this into account I consider it more likely than not that the partial retrospective insertion theory has been constructed by the Defendants in order to seek to explain the otherwise inescapable fact that every single sales record, from the first sale of beds by the First Defendant using the Listings to the last, contains the name Birlea in the title. The reason that they do is, in my judgment, the obvious one: namely that the titles of the Listings used ‘Birlea’ at all times from the First Defendant’s first use of the Listings. I so find. In light of this finding I consider it is more likely than not that the brand field was also similarly populated ‘by Birlea’ at all relevant times.
    

Issue (iii) – Were the Defendants notified of the fact that beds were listed under the name ‘Birlea” as a consequence of: (a) the order details that were provided by Amazon as a consequence of each sale of the beds complained of; (b) the receipt of and response to the email from Dan Ambalavanar of 4 September 2014; or (c) the receipt of the letter before action?

48. It is not disputed that every sale of a bed by the First Defendant from the Listings from the first sales in February and March 2013 caused the prompt and automatic generation by Amazon of notifications in the form of: (i) a sales record to the First Defendant’s seller account and (ii) an email notifying the First Defendant of the sale sent to the Second Defendant’s email account. Nor is it disputed that both contain the title of the relevant Listing at the time of the sale. I have found that at all times those titles contained the name ‘Birlea’.
    

Issue (iv) – When were the Defendants notified that they were advertising beds under the name “Birlea”?

49. From receipt of the first sales record and email notifying the first sale of a bed by the First Defendant using the D4G Listing, namely 27 February 2013.
    

Issue (v) – Did the Defendants continue to advertise beds under the name “Birlea” after they became aware of such use?

50. Yes. It is not disputed that the use of the Listings by the First Defendant to advertise and sell beds continued until 5 May 2016.
    

Issue (viii) – Did the First Defendant infringe the Claimant’s EUTM?

51. Yes. The Defendants accept that this is the result of my finding that the Listings contained the sign Birlea at the time that the First Defendant first used the Listings. All use of the Listings by the First Defendant to offer for sale, advertise and sell beds after 23 April 2013 (the filing date of the Mark) infringes the Mark.
    

Issue (xi) – If so, is the Second Defendant also liable for any such infringements?

52. The law in relation to joint tortfeasance is that set out by the Supreme Court in the case of Fish & Fish Ltd v Sea Shepherd UK [2015] UKSC 10, [2015] AC 1229. The headnote provides the following summary:

“In order to be liable with a principal tortfeasor a defendant had to be proved to have combined with the principal tortfeasor to do, or to secure the doing of, acts which constituted the tort; that that required proof that the defendant had acted in a way which furthered the commission of the tort by the principal tortfeasor and that he had done so in pursuance of a common design to do, or to secure the doing of, the acts which constituted the tort; and that whether the matters relied on by a claimant had any significance to the commission of the tort would depend on the circumstances in each case.”

53. It was further helpfully summarised by HHJ Hacon in Vertical Leisure v Poleplus Limited [2015] EWHC 841 (IPEC) at para 66:
    

“I interpret this to mean that in order to fix an alleged joint tortfeasor with liability, it must be shown both that he actively co-operated to bring about the act of the primary tortfeasor and also that he intended that his co-operation would help to bring about that act (the act found to be tortious). Liability will always be subject to the threshold requirement that the alleged joint tortfeasor’s contribution to the act was more than de minimis.”

54. Whether the alleged joint tortfeasor has met the threshold of joint tortfeasance by his actions is acutely fact-sensitive. In this case Ms Blythe for the Defendants submits that:
    

i) at the time that the Second Defendant advertised the beds on Amazon using the Listings he was a mere employee of the First Defendant as he did not act as a director until 1 April 2014;

ii) as an employee he was authorised and instructed by the directors of the First Defendant to buy the beds and advertise them on the internet;

iii) even after he became a director he is only one of four directors and seven shareholders of the First Defendant; and

iv) as such he is not the controlling mind and spirit of the First Defendant.

55. The difficulty I have with these submissions is that they do not address the issues that I have to consider per Fish & Fish. There is no requirement for a joint tortfeasor to be a director of a corporate primary tortfeasor, let alone to be the controlling mind and spirit of it. The Second Defendant admits in his witness statements that he and his brother met with Profitlane representatives at a trade show in Malaysia where they were told about Torino daybeds for sale; that he telephoned that representative from the UK after their return and agreed a price to buy the beds; that his main role at the First Defendant is to deal with online sales and he oversees a team of ten people for that purpose; and despite that large team, it was he personally who put the beds onto the Amazon platform and attached them to the Listings for sale. In my judgment these factual circumstances fulfil the Fish & Fish requirements that (i) he acted in a way which furthered the commission of the tort by the principal tortfeasor and (ii) that he had done so in pursuance of a common design to do, or to secure the doing of, the acts which constituted the tort. I have already set out the leading role that the Second Defendant has played in these proceedings, including signing pleadings as a director of the First Defendant even before he was joined as a Second Defendant, and providing the majority of the evidence relied on by the First Defendant. Although these do not, of course, go to the commission of the tortious acts of infringement of the Mark, they do evidence, to my mind, the principal role that the Second Defendant played in their commission, which I find easily meets the threshold of ‘active-co-operation’ identified by HHJ Hacon in Vertical Leisure. In addition, the effect of my finding that the Listings contained the Birlea sign at all times when the First Defendant used them, means that I am satisfied that the Second Defendant had actual knowledge that the First Defendant’s beds were being sold by use of the sign complained of at all material times.
    
56. For those reasons I find that the Second Defendant is jointly liable for the infringing acts of the First Defendant as joint tortfeasor.
    

SUMMARY

57. To summarise:
    

i) The First Defendant is liable for infringement of the Mark under Article 9(1)(a)/Article 9(2)(a) of the 2009 EUTMR/EUTMR in relation to the advertisement, offer for sale and sales of beds using the Listings from 23 April 2013 onwards.

ii) The Second Defendant is jointly liable with the First Defendant as joint tortfeasor.

58. The judgment is listed for handing down at 3pm on 11 January 2018 at the Rolls Building. I will hear submissions on consequential orders, including directions on to a quantum trial, at the handing down.