David Stone (sitting as Deputy High Court Judge) :
1. “Bodycon” and “bandage” are two dress styles made popular by celebrities including Jennifer Lopez, Beyoncé and Kim Kardashian. As the name suggests, bodycon dresses emphasise body contours, highlighting the wearer’s bust and buttocks, and minimising the waist. Bandage dresses have a similar effect, being made of thick, stretchy material like bandages. The Claimants sell bodycon and bandage garments under the main brand House of CB with a sister brand Mistress Rocks, mostly online, but with some bricks and mortar stores in the United Kingdom and the USA. They say that the Defendants, competitors who sell bodycon and bandage garments online under the brand Oh Polly, have copied 91 of their garment designs, including dresses, jumpsuits and tops, thereby infringing their unregistered design rights under UK and EU law. Further, the Claimants say that the Defendants have copied their business model, social media, marketing, packaging and presentation (including using the same models and the same locations for photoshoots) such that consumers will be deceived into thinking that Oh Polly is a sister brand to House of CB, so as to amount to passing off.
2. The Defendants deny design infringement, saying that, whilst on some occasions, House of CB garments were “referenced” in the production of Oh Polly garments, they were not copied. Thus, the Court is asked to determine when “referencing” or “inspiration” - a common feature of the fashion industry - ends, and unlawful copying begins. On the passing off case, the Defendants admit that there are some similarities between the two businesses, but say those similarities are common amongst many online fashion businesses, and no consumers have thought or will think that Oh Polly is a sister brand to House of CB.
3. By and large, it is sufficient throughout this judgment to refer to the two sides, without specifying any particular claimant or defendant. For completeness, I note as follows:
i) The First Claimant is a Hong Kong-registered company that operates the websites at
www.houseofcb.com and
www.mistressrocks.com. In addition to its e-business, the First Claimant has also entered into agreements with franchisees, including the Third Claimant, which sell its garments through bricks and mortar stores.
ii) The Second Claimant is also a Hong Kong-registered company, which owns various registered trade marks (which are not in issue in these proceedings). It licenses those trade marks to the First Claimant.
iii) The Third Claimant operates bricks and mortar retail stores in the United Kingdom, with stores in London, Sheffield and Manchester.
iv) The First Defendant is a company registered in England and Wales - it carried on the Oh Polly business from 2015 until 2017, when the business was transferred to the Fourth Defendant. It now has a share of a warehouse and receives rent from the Fourth Defendant.
v) The Second Defendant, Ms Claire Henderson, is the CEO of the Oh Polly business.
vi) The Third Defendant, Dr Michael Branney, is the Managing Director of the Oh Polly business. Each of Ms Henderson and Dr Branney owns 50% of the First and Fourth Defendants. They have admitted for the purposes of these proceedings that they would be jointly liable with the First and Fourth Defendants for any acts of unregistered design right infringement and/or passing off.
vii) The Fourth Defendant is a company registered in Scotland - it carries on the Oh Polly business today, and has since 2017.
4. The progress of the matter from complaint to trial was not particularly happy. The First Claimant, through its then solicitors, first wrote to the First and Third Defendants on 19 April 2016, alleging passing off, and that eight Oh Polly garments (sold in various colourways) infringed its UK unregistered design rights (UKUDR) and Community unregistered design rights (CUDR). A further letter before action was sent on 25 November 2016, this time relying on 19 designs, none of which had been included in the initial letter. The Defendants did not respond to that letter.
5. Eighteen months later, on 6 June 2018, a further letter before action was sent, this time to the Second, Third and Fourth Defendants, listing 87 garments that were said to infringe the First Claimant’s UKUDR and CUDR. Only 9 of the 19 garments referred to in the letter of 25 November 2016 were included in this list. Without waiting for a response, the Claim Form in these proceedings was issued the next day.
6. The Defendants did not respond to the claim, and the Claimants issued an application for judgment in default. After a compromise, directions were agreed for the service of a Defence. There was a hearing on 21 November 2018 before Zacaroli J, who made orders to keep confidential the names of the designers of garments on which the Claimants relied. The Defence was finally filed on 25 January 2019, some six months following service of the Claim Form. The Claimants filed their Reply some five months later on 14 June 2019, together with draft Amended Particulars of Claim. That latter document abandoned design infringement in relation to 15 of the 87 pleaded garments, and added a further 23 new garments, four of which claims were abandoned prior to the formal application to amend.
7. The first CCMC took place before Deputy Master Nurse on 15 October 2019, who accepted on case management grounds that the case ought to be “streamlined” and listed this trial. The Amended Defence was served on 31 January 2020. A second CCMC took place before Deputy Master Nurse on 7 April 2020 at which streamlining was ordered - each side was directed to select 10 garments to be dealt with at this trial (along with the passing off claim), with the design infringement claims in relation to the remaining 71 garments stayed. Each side selected 10 garments on 13 May 2020, but the Claimants did not like the garments selected by the Defendants, and issued an application to have the trial of the Defendants’ 10 selected garments stayed. That application was heard by Deputy Master Lloyd on 20 August 2020 and dismissed with costs.
8. The selected garments for this trial each have a number, with C representing the Claimants’ garment in which UKUDR and CUDR are asserted, and D representing the Defendants’ garment which is alleged to infringe. The garments in this trial are: C2/D2, C3/D3, C4/D4, C7,/D7, C9/D9, C12/D12, C13/D13, C17/D17, C21/D21, C35/D35, C47/D47, C49/D49, C61/D61, C63/D63, C66/D66, C77/D77, C81/D81, C91/D91, C93/D93 and C102/D102. Images of each garment appear below. Each of the garments also has a name - in the case of the Claimants, usually a woman’s name, and in the case of the Defendants, usually a pun, innuendo or play on words. For clarity and simplicity, I have used the garment numbers rather than names.
9. Fact evidence was finally exchanged on 2 October 2020, just two months before the trial. I heard the Pre-Trial Review on 26 October 2020, and made orders for a Re-Amended Defence, a Re-Amended Reply and reply evidence. I also lifted Zacaroli J’s confidentiality order for reasons I gave at the time.
10. In the end, at this trial, the Court was asked to assess 20 garments for infringement, each in relation to both UKUDR and CUDR. In relation to CUDR, several of the garments were available in multiple colourways, each of which had to be assessed. The Defendants relied on 108 prior designs in an attempt to invalidate the Claimants’ UKUDR and CUDR. As will become apparent from the length of this judgment, that task involved comparing each of the Claimants’ 20 designs against the prior designs for validity under UKUDR and then under UCDR (the latter sometimes involving multiple colourways). It then involved comparing each of the Claimants’ 20 designs against each of the 20 allegedly infringing garments, in each case under the two different legal regimes (UKUDR and CUDR, with the latter involving multiple colourways). That involved something over 270 comparisons under two different legal regimes.
11. By the end of the trial, counsel for the parties agreed that the designs in issue fell within only three “buckets”, to which I return below. I am therefore moved to repeat what Henry Carr J said in Neptune (Europe) Limited v Devol Kitchens Limited [2017] EWHC 2172 (Pat) (Neptune), at paragraph 4:
“At the pre-trial review, I ordered that the issues of liability in relation to the unregistered design rights were to be tried based upon 3 designs selected by Neptune and 3 designs selected by DeVOL. With hindsight, it would have been better if I had limited the parties to a single design each, as the same issues could have been fully argued. In future (irrespective of whether the claim is part of the Shorter Trial Scheme) where multiple designs are in issue, it would be sensible to confine the liability trial to an appropriate, and limited, selection.”
12. If the garments before the Court in this trial are representative of the 91 garments in issue in the proceedings (and the Claimants say they are), then this trial could readily have been heard in relation to three garments only, so as to dispose of the remaining 88 garments by consent (or by relisting any garments that could not be agreed). This would have saved the parties considerable time and costs.
13. The trial was heard remotely because of Covid-19 precautions, using photographs of the various garments rather than the garments themselves. At the request of the Defendants, the Court reconvened following closing speeches for a viewing of the 20 garments relied on by the Claimants and the 20 garments alleged to infringe, with only counsel and the transcript writer present. Whilst unusual, this was a useful exercise and enabled me to compare the garments side by side, allowing counsel to draw to my attention similarities and differences. I also had the garments available to me whilst preparing this judgment. Whilst what was made available to me may have been the only articles the parties could obtain, in a case such as this in future it would be preferable for the physical examples provided to be in comparable sizes, and comparable colourways. There are additional challenges when comparing one example in size XS and one in size XL.
14. This judgment contains an unusually large number of pictures, aimed at reducing the word count necessary to describe each garment. I regret that a number of the images have cropped the model’s head and in using these, I mean no disrespect to the models: these were, unfortunately, the only images available to me. I also note that some of the garments were described as “nude” in colour. Criticisms of that usage have been made for at least 10 years. I have, however, adopted and capitalised the parties’ use.
15. I have excerpted below social media posts from members of the public who were not called to give evidence and likely do not even know that these proceedings are underway. To preserve their anonymity, I have not referenced their social media handles, but they were in evidence before me. When quoting from social media posts, I have not used “[sic]”: the text and emojis are reproduced exactly.
16. The trial occurred prior to the United Kingdom’s departure from the EU-wide design regime established by Regulation (EC) No 6/2002 of 12 December 2001 on Community designs (the Design Regulation), but this judgment is to be handed down following that exit. I was asked by counsel for both sides to decide the case on the basis of the law as it was at the time of the trial, and hence I refer to the Design Regulation, rather than to the relevantly identical UK law which replaced it at 11pm on 31 December 2020. It should also be noted that at the time of the trial, this court was sitting as a Community Design Court, but it had ceased to be such a court by the time of hand-down. As no pan-EU remedies were sought, this makes no difference, and the parties did not ask for hand-down to take place prior to 31 December 2020.
17. There were no major disputes between the parties as to the applicable law, so this case turns entirely on its facts.
18. Ms Anna Edwards-Stuart and Mr David Ivison appeared for the Claimants. Mr Chris Aikens appeared for the Defendants.
19. The Claimant relied on four witnesses, each of whom was cross-examined remotely at trial. Some witnesses gave evidence from outside the jurisdiction: I was assured by counsel for the parties that local laws had been complied with. I am grateful to the firms of legal advisors who hosted the witnesses to enable them to give their evidence remotely.
20. Ms Connor Walker founded what is now the House of CB business now owned by the Claimants when she was 17 years old and travelling in Asia with her parents. As Connor Walker’s cousin Kirsty Walker also gave evidence, I will use both women’s first names without thereby meaning any disrespect.
21. Connor Walker remains heavily involved in the business, and continues to design garments for the House of CB brand. She gave evidence of the development of the business from that owned by a teenager selling dresses on eBay to the multi-million Pound business it is today.
22. Counsel for the Defendants submitted that Connor Walker was not a helpful witness describing her as “visibly impatient”. He pointed to what he said were inaccuracies in her evidence and/or issues on which she failed to accept matters put to her, and submitted that I should approach her evidence with some caution. I disagree. Connor Walker was clearly trying her best to assist the court. She did come across as impatient with some of the allegations put to her, but no more so in my judgment than some-one who feels that her designs and her business model have been copied by a third party. I have carefully reviewed the five examples of alleged inaccuracies presented by counsel for the Defendants, and I do not consider that they demonstrate unhelpfulness or the unreasonable refusal to accept matters which he submitted. Further, none of the examples given could be said to go to any key issue in the case.
23. Ms Joanna Richards is Connor Walker’s mother and is actively involved in running the House of CB business. Ms Richards’ witness statement addressed the appearances and features of the House of CB and Oh Polly websites, and the models and packaging they use.
24. Counsel for the Defendants submitted that Ms Richards tended to argue the case, and that I ought to treat her evidence with caution. Again, I disagree. I found her to be an honest and straightforward witness. Whilst she occasionally came across as feeling aggrieved by the Defendants’ actions, that cannot be held against her. She accepted under cross-examination inaccuracies as they were put to her.
25. Ms Georgina Douek is a designer employed by Sirens Design. Sirens Design is a design agency set up and owned by the Claimants to design garments for House of CB and Mistress Rocks: I return to it below. Ms Douek gave evidence that she was the designer of five of the garments in issue at this trial (C4, C12, C17, C61 and C93). She was also involved in the design of C21. Her statement also described the general design process at Sirens Design. No criticism was made of the way in which Ms Douek gave her evidence.
26. Ms Kirsty Walker is Ms Richards’ niece and Connor Walker’s cousin. She is a Warehouse Supervisor employed by Stocks Away Limited since 2014. She supervises a team who dispatch House of CB garments to consumers and deal with returned goods. No criticism was made of the way in which she gave her evidence.
27. Counsel for the Defendants criticised the Claimants for the witnesses they did not call:
i) the four other designers involved in the design of garments in issue in this trial: Rheanna Donaldson, Julia Kasper, Justin Ruddle and Alcy Lynch; and
ii) any individual who had actually been misled or confused that Oh Polly is a sister brand of House of CB.
I return to these criticisms below.
28. The Defendants served witness statements from eight witnesses, six of whom were cross-examined.
29. Ms Claire Henderson is the Second Defendant and CEO of the Fourth Defendant. Her brother Mr Joe Henderson did not give evidence, but he is relevant to the proceedings. I will refer to her as Ms Henderson and to him as Mr Henderson.
30. Ms Henderson gave evidence that she designed 17 of the 20 Oh Polly garments in issue in this trial. Her two witness statements set out her “design process” generally, as well as specifically in relation to the 17 designs.
31. Counsel for the Defendants accepted that “on some occasions” Ms Henderson “could have answered the specific questions put to her in a more straightforward and direct way and that she had a tendency to repeat herself”.
32. Counsel for the Claimants went further than that: she described Ms Henderson’s oral evidence as “a disaster”. It was, she said, impossible to get a straight answer from Ms Henderson in response to any question which she perceived as being problematic for the Defendants’ case. I accept that submission. Ms Henderson’s answers were rambling, obfuscatory and at times incoherent. She repeated some answers many times, even when not relevant to the question and having several times been requested by me to answer the question put to her. On this basis alone, I would have approached her evidence with some caution.
33. However, I have come to the conclusion, regrettably, that I must accept counsel for the Claimants’ submission that Ms Henderson has lied to the court. I therefore need to set out three issues in detail.
The Dropbox Exhibits
34. In her first witness statement, Ms Henderson gave evidence that between August 2015 and January 2018, she was designing 40 to 50 garments per month. She maintained an archive of images of tens of thousands of third party garment designs (potentially as many as 50,000) that she had downloaded from the Internet. Ms Henderson gave written evidence that to come up with an Oh Polly garment, she would research trends and then “conceptualise… the appearance of my own design in my head”.
35. I set out in full paragraph 37 from Ms Henderson’s first witness statement:
“Say there was a particular trend, for example Kim Kardashian wore a lace up dress which started to trend, I would have an image of the sort of design I wanted to produce as my own version of that popular or trending style. At this point, I might go through my saved pictures and set up a new folder on my laptop and put into that folder all the images of dresses that were very similar, that had a particular colour or maybe that had a little design feature that I thought would work well with the lace up design I had in my head. Other times, it was quicker to do an internet search to find an image similar to the concept I had in my head for illustration purposes, rather than go back through images I’ve already saved. Instead of printing material out, my moodboard is a folder of images. From there, I use the images as a reference point for the design of the garment I had conceptualised in my head and which I wanted to produce.”
36. Ms Henderson referred to these mood board folders in her discussion of the design process for each of the particular garments in issue. For example, in relation to D4 she said (at paragraph 51 of her first witness statement):
“In 2018, bralettes were really in style and the single-shoulder, knotted effect appeared to be a trend, particularly for swimwear, as evidenced by the images I had saved and are visible at CLH16.”
37. Exhibit CLH16 is headed “Bralettes and Shoulder Knots” and shows a list of links to images held in Dropbox as part of Ms Henderson’s archive. The dates on which those Dropbox images were saved are given - in CLH16, the last date given is 3 June 2016 (D4 was conceived in March 2018). There are also a series of images set out in CLH16. I will refer to CLH16 and the similar exhibits for different garments as the Dropbox Exhibits.
38. In relation to D7, having referred to images of Liz Hurley and Mariah Carey in strappy dresses, Ms Henderson wrote at paragraph 53 of her first witness statement:
“Both garments are fairly basic, as evidenced by the research I had done and the images I had saved, which are visible at CLH18.”
39. Dropbox Exhibit CLH18 is similar in presentation to CLH16. It is headed “Neckline and Strappy Back Dresses”, and lists two Dropbox addresses, one saved on 6 April 2016 and one saved on 21 April 2016. Neither shows Liz Hurley or Mariah Carey. D7 was conceived in October 2017.
40. Ms Henderson’s first witness statement made similar claims in relation to D2, D9, D12, D13, D17, D21, D35, D47, D49, D61, D66, D77, D81 and D91, for each of which she provided a Dropbox Exhibit in a relevantly identical format and with similar content. A Dropbox Exhibit was therefore provided by Ms Henderson in relation to 16 of the 17 designs in issue before me for which Ms Henderson was identified as the designer: there is no Dropbox Exhibit for D63.
41. Thus, as I understood Ms Henderson’s written evidence prior to the start of the trial, it was that she created digital mood boards - and they were represented by the Dropbox Exhibits. As it turned out, and as Ms Henderson’s oral testimony stood at the conclusion of her second day of cross-examination, the Dropbox Exhibits were not that at all. Rather, they were documents created for the purposes of these proceedings, which recorded garments that Ms Henderson had found in her archive on a review conducted whilst preparing her written evidence. They captured garments which she may have looked at - or at least, documents which looked like the garments in issue in this trial. An excerpt from her cross-examination is set out below:
“Q. And if you scroll down you can see those two images there which, presumably, correspond to the two Dropbox links?
A. Yes, I believe so.
Q. Ms. Henderson, do you say that in the course of preparing, of creating the design for [D7] you actually looked at either of the two dresses in the top image or the dress on the bottom image?
A. My Lord, I think I would have been familiar with them, yes, because usually my process, as I have set out numerous times now, when I am looking at creating a new design, I will also, let us just say, for example, that I have the Mariah Carey image and the Liz Hurley image and I like those and that is a design that I want; I will still refer to my Dropbox to see if there is any features that maybe in my Dropbox I have saved that I like. So I will refer to those, yes.
Q. Ms. Henderson, you did not answer my question.
A. Sorry, what was the question?
Q. My question was, in the course of designing [D7] ----
A. Okay.
Q. ---- did you, as a matter of fact, go back to your Dropbox and look at any of the dresses that we see on page I1300 or I1302, "yes" or "no"?
A. I believe I would have, yes, my Lord. I believe I would have.
Q. I am not asking you what you believe. Did you, Ms. Henderson, "yes" or "no"?
A. I believe, based on the process that I used to work, I would have seen these, yes. I do not believe that the process to create this garment would be any different to the process that I used to create every garment. So with that regard, yes, I would have seen these images. To the best of my knowledge, yes, I believe I would have seen these images.
Q. The answer, Ms. Henderson, is you do not know?
A. It is not the answer, my Lord. If it was the answer I would be honest and that is what I would say.
Q. You do not know because you have just said you "believe". You do not say you know, you do not remember?
A. I believe I would have because that is the process that I use for every single garment that I produce. That is the purpose of me having my Dropbox and having all the folders. It is so that I can refer back to them on a frequent basis which is what I do. It is what I did all day every day.
Q. I will ask you one more time, Ms. Henderson.
A. Okay.
Q. Are you confident, as a matter of fact, that you looked at each of these dresses in the course of designing D7?
A. I am confident, yes.
Q. Are you confident yes or no?
A. Yes, I am confident, my Lord. I am confident based on how I design a garment. I am confident. That is how I design a garment. I have something in mind, even if I have a specific, even if my idea and I have an illustration and it is very clear, I will always, always, always go back to my Dropbox, because there may be another add-on feature in my Dropbox that will help enhance the design. I refer back to my Dropbox for colours. I am always within those folders, always.”
42. At the end of her cross-examination, I asked Ms Henderson some questions about one of the Dropbox Exhibits:
“Q. Did you prepare that document?
A. I put the images together. I had phone calls with [Dr Branney] over the course of about one week and we put the images together. So it was myself who went through my Dropbox because, obviously, [Dr Branney] is not so aware and then he would put he had put them together for the lawyers, I believe, yes.
Q. Roughly when did that happen?
A. I think it was about June time, June/July, yes, maybe like June when I, before I left, so probably about June, maybe.
Q. All right, and that is true for each of the documents you were taken to that looks like this, so each of the lists of computer codes?
A. We went through all of the designs, my Lord, yes. It took quite a while. But in putting the document together that was [Dr Branney] who had actually put it together, if you know what I mean. I did not sit here putting it together. I just went through on phone calls and confirmed the images with him and then sometimes he would show me an image just to confirm if it was from my laptop.”
43. Having seen Ms Henderson give her oral evidence and having reviewed each of the Dropbox Exhibits, they are, in my judgment, documents put together for the purposes of these proceedings, as Ms Henderson accepted by the end of her testimony. They were created in June or July of 2020 when Ms Henderson was preparing her evidence for this trial. They were created by Ms Henderson and Dr Branney. I have no confidence at all that these are images of garments Ms Henderson referred to in the creation of the garments in issue in this trial. Ms Henderson’s own evidence was that she was designing 40 to 50 garments a month in the relevant time frame. She had tens of thousands (possibly as many as 50,000) images of garments in her Dropbox. It is simply not possible that she was able to reconstruct the list of images from which she took inspiration, nor did she, under cross-examination, make that case. They are therefore not the “new folders” or “moodboards” that Ms Henderson referred to in paragraph 37 of her first witness statement. Therefore, even if Ms Henderson’s first witness statement is not technically inaccurate (and I consider that it is), it certainly gives a misleading picture of what the Dropbox Exhibits represent.
44. That misleading impression is amplified by the Re-Amended Defence and the Amended Annexes to the Defence. The Re-Amended Defence notes at paragraph 38(a) that Annexes D1 to D106 contain “Particulars of the design of each of the Oh Polly garments alleged to infringe, namely the dates of design, who created it, their employment status, and images of any third party, Oh Polly or House of CB designs referenced during the design process that the Defendants have been able to provide” (emphasis added). This document was signed by Ms Henderson and by Dr Branney under a statement of truth on 28 October 2020, after they had prepared the Dropbox Exhibits, and shortly before the trial.
45. For each design, each Amended Annex has a Section 5 titled Particulars of Defendants’ Design. That section includes: 1. Date of design; 2. Designer/status; and 3. Designs referred to. Under the last of those headings, the images taken from the Dropbox Exhibits are set out. This is, in my judgment, an express representation that the designs shown were actually referred to by the named designer at the time of the listed “date design conceived” and prior to the listed date the design was finalised. In every case, that representation is incorrect for the reasons I have given. Ms Henderson was the named designer in relation to the 17 garments: she, and only she, knew authoritatively the third party garments to which she had referred.
46. There are several other reasons for suspecting the Dropbox Exhibits:
i) There is a Dropbox Exhibit for design D35. The Defendants admitted that D35 was copied from the Claimants’ C35 garment, and that no other garments were “referenced” by Ms Henderson when designing D35. Yet a Dropbox Exhibit exists, allegedly setting out the other documents referred to. Like others of the Dropbox Exhibits, that for D35 includes images for garments that are very dissimilar to the garment Ms Henderson said she then designed. For example, shown below are images of C35, D35 and some of the images shown in the Dropbox Exhibit for D35:
C35 D35
Excerpted images from the Dropbox Exhibit for D35
In cross-examination, Ms Henderson first said that she had “referenced” the images in the Dropbox Exhibit for D35, before later admitting that she had not “referenced” these images at all. Given the similarities between D35 and C35 (they are close to identical) and the very great differences between D35 and the images shown in the Dropbox Exhibit for D35, her later admission, which is consistent with the pleadings, seems significantly more likely, and I so find;
ii) The computer file references shown in the Dropbox Exhibits identify different folders and sub-folders which are not named after the Defendants’ garments or features of those garments;
iii) The dates on which the computer files were saved span a period of many months and hence do not suggest that they were accessed and saved over a short space of time during what Ms Henderson described as her research;
iv) The dates on which the computer files were saved are in all cases months, and in some cases years, prior to the pleaded date of creation of the Defendants’ garments;
v) Some of the Dropbox Exhibits included images which did not reflect the features of the garments for which they were alleged to have been research, or, indeed, the heading of the document. For example, the Dropbox Exhibit for D49, a satin mini dress, does not contain any images of satin mini dresses, even though it is headed “Satin mini dressed [sic]”. Instead, it contains images of tops, pantsuits and two dresses to the knee and below the knee, in a variety of fabrics; and
vi) The Defendants now admit that they referenced specific garments (some from House of CB, some from third parties) in relation to each of the Defendants’ garments, but those referenced garments do not appear in the Dropbox Exhibits.
47. Following cross-examination, the Defendants persisted with the argument that Ms Henderson created digital mood boards from which she was inspired, including mentioning as much in their counsel’s closing skeleton. In my judgment, there is no reliable evidence before me that Ms Henderson created mood boards, digital or otherwise, for the garments in this trial which she was said to have designed.
48. Ms Henderson knew that the Dropbox Exhibits were documents created for the purposes of these proceedings. She presented them, and allowed them to be presented, as “digital mood boards” that inspired her at the time she created the relevant designs, right up until her counsel’s closing speech. Her counsel accepted that it was “unfortunate that [Ms Henderson’s] precise methodology was not spelt out in [her] witness statement”. I agree. It was both unfortunate and misleading.
The Amended Defence
49. Counsel for the Claimants raised a point in relation to the pleadings that also goes to the reliability of Ms Henderson’s evidence. Ms Henderson signed the Defence under a statement of truth in early 2019. That document included the following statement:
“Save that garment C35 was referenced in the creation of the design for garment D35, the Defendants created the designs for the Oh Polly Garments independently and without copying any of the designs relied on by the Claimants.”
50. As will become apparent, the Defendants have now conceded that the Claimants’ garments were referenced by Ms Henderson on at least 9 occasions in relation to the 20 garments in issue in this trial. The Defence was later amended and re-amended, such that the Re-Amended Defence before me reads as follows:
“Whilst in some instances a House of CB garment was referenced in the creation of the design for the Oh Poly garment, in all cases other than D35, for which the allegation of copying is admitted, the Defendants created the designs for the Oh Polly Garments without copying any of the designs relied on by the Claimants.”
51. The Defence was therefore false at the time Ms Henderson signed it. Counsel for the Claimants submitted that Ms Henderson lied when she signed the Defence, and I accept that submission. Again, she had designed the garments in issue and if she was unable to recall her inspiration/references, she had access to her records (including the Defendants’ internal record keeping database, called Trello, to which I return below) and should have consulted them.
52. I therefore consider Ms Henderson also to have lied when she signed the original Defence.
Ms Henderson’s Design Process
53. It was common ground that Ms Henderson cannot draw. She also cannot use any of the available computer assisted design programs. Her evidence was that she designed by looking at trends from images she had collected in her Dropbox, designed the details of each garment in her head, and then looked for images of existing garments which she could use to show the factory what it was she wanted made. There are no documents which (and no other witnesses who) back up this aspect of her account of her design process.
54. As set out above, Ms Henderson admitted under cross-examination that the Dropbox Exhibits were post facto documents created for the purposes of this litigation. She was unable to identify clearly in relation to any of the Defendants’ 17 garments she designed the actual third party images she had looked at in advance of her designing, and there was no evidence, other than her say so, that she had conceived of the full design in her head prior to going looking for an image or images which would enable her to demonstrate to the factory what she wanted. Instead, the written record shows that in each of the 17 cases, Ms Henderson identified at least one garment (being a House of CB garment or a third party garment), an image of which she emailed to Dr Branney and/or Mr Henderson or otherwise provided directly to the factory. I consider her version of her design process to be a fabrication, concocted to get around the very real difficulty that Ms Henderson took images of the Claimants’ garments (and third party garments) and sent them to the factory to be made up.
55. I therefore, regrettably, accept counsel for the Claimants’ submission that Ms Henderson’s evidence cannot be trusted. In my judgment, she set out to deceive the Claimants and the Court in relation to her overall design process, and the design of the particular garments before me. In her oral testimony, she gave whatever answer she felt best at that moment to assist her case. That criticism also extends to her written testimony. I have therefore discounted her evidence completely unless it is corroborated by another witness or the documents before me.
56. Dr Michael Branney is the Third Defendant and Managing Director of the Fourth Defendant. Dr Branney is largely responsible for running the operations of the Oh Polly business. He manages the staff and the website. His first witness statement set out some of the background to the Oh Polly business, its structure and focus, its aims, and various aspects of the running of the business. Dr Branney’s first witness statement also addressed the allegations of similarities between the Oh Polly business and the House of CB business. His second witness statement responded to the written evidence of Connor Walker, Ms Douek, Ms Richards and Kirsty Walker.
57. Counsel for the Claimants criticised Dr Branney as an argumentative and difficult witness, prepared to advocate for himself and the Defendants. There is some force in those criticisms. More seriously, counsel for the Claimants also pointed to Dr Branney’s dishonesty in relation to the original Defence, and in particular the passage I have excerpted above in relation to Ms Henderson’s evidence, suggesting that he was “willing to tell lies to further his and the other Defendants’ interests”.
58. Prior to signing the Defence, Dr Branney had conducted a review of the Defendants’ internal records, including the Trello system to which I have already referred, and to which I return in more detail below. At the time of signing the Defence, Dr Branney “could not identify a House of CB image in the Trello card[s]”. He accepted under cross-examination that the Defence was “wrong”. But his explanations for how he failed to recognise the House of CB garments on at least nine occasions were unconvincing.
59. Dr Branney was taken to several Trello records in cross-examination, and asked to compare the images there with the images of the Claimants’ garments which had been annexed to the Particulars of Claim. I set out below the image or images from the Trello records which Dr Branney was shown, and the image from the Particulars of Claim that Dr Branney was shown:
Images from Trello cards Images from Particulars of Claim
60. I accept counsel for the Claimants’ submission that “a perfunctory glance reveals the garments in these images to be the same”. Dr Branney was unable to explain how he missed these three images of the Claimants’ garments when he signed the Defence and I do not accept the explanations he gave. The Defendants took over 6 months after receiving the Particulars of Claim to file their Defence, and had ample opportunity to get this right. However, I consider Dr Branney’s failing here to be maladroit, rather than malevolent. He had not designed the garments in issue - Ms Henderson had. He was working from documents and from Ms Henderson’s explanations. He ought to have been more careful before signing the statement of truth, but I do not consider that this means I should discount his evidence completely.
61. Counsel for the Claimants described Dr Branney’s evidence as that “of a person willing to say whatever needed to be said to achieve his objective.” Having watched him give evidence, I consider that Dr Branney was an active advocate for the Defendants. I have therefore approached his evidence with caution, but I reject counsel for the Claimants’ submission that I should dismiss it altogether.
62. Ms Alexandra McShane is employed by the Fourth Defendant as a Customer Services Manager. She described in her witness statement the Oh Polly customer services team and returns process, including describing items from other retailers returned wrongly to Oh Polly. She was not cross-examined, and I therefore accept her evidence.
63. Mr Ramsay Bell is Oh Polly’s Head of Development. He described in his witness statement the history of the Oh Polly website, and gave examples of third party websites with features similar to those relied on by the Claimants for their passing off case. He was not cross-examined, and I therefore accept his evidence.
64. Ms Arlene McNab is employed by the Fourth Defendant as an E-Commerce Manager. Her witness statement primarily described the rebranding process that Oh Polly went through in early 2018. In cross-examination, Ms McNab’s attention was drawn to some errors in her evidence. Counsel for the Claimants criticised the way in which Ms McNab’s written evidence had been prepared, but not the way in which she gave her oral testimony, which I therefore accept.
65. Ms Amy Johnston is a Senior Designer employed by the Fourth Defendant. She was the first specialist designer employed in the Oh Polly business (in November 2017). Her evidence covered the design of two of the Oh Polly garments in this trial (D3 and D93). Counsel for the Claimants accepted that Ms Johnston answered fairly the questions put to her in cross-examination, but she submitted that, in fact, Ms Johnston was “often little more than Ms Henderson’s amanuensis, using an iPad app to sketch out ideas supplied to her by Ms Henderson”. I return to this issue below in relation to D3 and D93.
66. Ms Megan Mitchell is a Creative Brand Specialist at Oh Polly. Her written evidence addressed Oh Polly’s social media activities and other aspects of Oh Polly’s marketing strategy. Counsel for the Claimants accepted that Ms Mitchell had given her oral evidence in a straightforward way, but criticised the preparation of her witness statement. For example, Ms Mitchell’s witness statement said that she had been “referred to the Particulars of Claim”, when in fact she had never seen the document. Counsel submitted that Ms Mitchell was “not particularly well-engaged with the production of her witness statement”. I accept that submission. However, whilst regrettable, I do not consider it makes any difference to the matters I need to decide.
67. Ms Shauni Connell is a Senior Womenswear Designer employed by the Fourth Defendant. Her witness statement described her creation of the design for D102. Ms Connell explained in a straightforward way the inspiration for her design for D102. Counsel for the Claimants submitted that D102 ended up closer to Ms Johnston’s sketches than Ms Connell’s, but accepted that her evidence was given honestly.
68. I understand the following account of the background facts not to be contested.
69. In 2009, Connor Walker began purchasing bandage dresses in China whilst accompanying her parents on business trips. She brought the dresses back to the UK and sold them through eBay. She called this business Celeb Boutique. In 2010, Connor Walker built the website at
www.celebboutique.com to move her business away from eBay. Through that website, she sold bandage and bodycon dresses. Connor Walker used social media to promote sales. In 2011, she started giving dresses to celebrities including reality television stars such as the cast of The Only Way is Essex. Later that year, the First Claimant was incorporated to manage the Celeb Boutique business.
70. Until 2012, the garments for sale through the Celeb Boutique business were purchased from wholesalers. Connor Walker now accepts that some of those garments were copies of third party designers, including a number of dresses copied from designs by Hervé Léger.
71. In 2012, Connor Walker moved Celeb Boutique towards designing its own garments. Siren Branding Limited was set up for that purpose. The design business of Siren Branding Limited was transferred to Sirens Design Limited in 2014: as it does not matter for the purposes of this case, I refer to both entities as Sirens Design. All rights in designs created by Sirens Design are owned by the First Claimant. Eventually, Celeb Boutique opened its own factory to manufacture the garments designed by Sirens Design on its behalf.
72. In July 2013, Celeb Boutique opened its first bricks and mortar store at Westfield Stratford City shopping centre in London.
73. Also in 2013, Connor Walker decided to rebrand the Celeb Boutique business as House of CB. “House of” was taken from the French term “Maison de” which is used by traditional couture houses. CB, the initials of Celeb Boutique, maintained some brand continuity. A new logo and a new website were developed, with
www.houseofcb.com going live in July 2014. The physical stores were rebranded at the same time. Following the rebrand, turnover increased significantly, as did the number of celebrities wearing the brand’s clothing (including Jennifer Lopez and members of the Kardashian family). The business expanded to the USA, and a physical store was opened in Los Angeles in June 2015 for which Khloe Kardashian hosted the opening party.
74. In April 2014, Connor Walker launched a sister brand: Mistress Rocks. Mistress Rocks is aimed at a slightly younger customer than House of CB. Mistress Rocks garments are also designed by employees of Sirens Design.
75. The clothing business that became Oh Polly also started out on eBay, with Ms Henderson selling some of her own clothing and other items while still at university. In 2013, Ms Henderson and Dr Branney started using the name Polly Couture for that business. It was a success. The First Defendant was incorporated in 2013 to conduct the eBay business.
76. In 2015, Ms Henderson and Dr Branney decided to set up their own business. They adopted the name Oh Polly for that business, and launched a website at
www.ohpolly.com in August 2015. The Fourth Defendant took over the business in June 2017. Ms Henderson’s evidence was that at the start of the Oh Polly business, she designed all the garments sold by the business, designing 40 to 50 garments per month. Ms Johnston joined Oh Polly as a designer in 2017. Oh Polly garments are made in factories in Bangladesh and China.
77. The Defendants used a software program to assist with the manufacture of garments, called Trello. Within the Trello system, there are “cards” which track the development of a garment from design through to the end of manufacture. Trello records for each of the Defendants’ garments were in evidence before me. In each case, the Trello records included images uploaded to Trello by Ms Henderson, Dr Branney and Mr Henderson. They also included images of various fittings for the garments, as well as instructions for the factory to make changes. The Defendants accepted that some of the Trello records were not complete - that is, there could have been additional images or instructions given and/or received which have not been recorded.
78. I need to say something about mood boards. The Claimants’ evidence was that they created mood boards for each collection by collating images from magazines and the Internet, fabric swatches and the like, and pinning them to foam boards fixed to the wall of the design studio. These mood boards set the tone for the garments then to be designed. This is common practice in the fashion industry. An image of Connor Walker at one of the Claimants’ mood boards is shown here:
79. The Claimants’ practice was for mood boards to remain in place for approximately six weeks whilst garments were designed and fitted. At the end of that time, the images and fabric swatches, etc, would be removed and thrown out.
80. Counsel for the Defendants was highly critical of the Claimants’ failure to preserve and disclose the appearance of the mood boards used in the design of the garments in issue in these proceedings. Even if it was not practicable to preserve the boards themselves (because they were fixed to the wall of the studio, and were needed for the next collection), he said photographs should have been taken of them, and the actual images/fabric swatches retained and disclosed.
81. Counsel for the Claimants’ answer was a simple one - the obligation on the Claimants was to preserve disclosable documents “as soon as litigation is contemplated”, and they have done so. The Claimants, she said, can only contemplate litigation in respect of a design after it has come to their attention that the Defendants have released a similar garment - by then, months, sometimes years or more, have passed - and hence the mood boards have already been disbanded. Further, counsel for the Claimants submitted:
“For their allegation of deliberate destruction of disclosable documents to be sustainable, it would have to be the Defendants’ position that the Claimants are obliged to ensure that their design companies preserve all of their mood boards for an indefinite period - even though they have no use for them - in case at some point in the future they come to suspect that the Defendants have copied another of their garments. That is clearly wrong.”
82. I agree with that submission. Whilst litigation against the Defendants was contemplated since 2016, and initiated in 2018, that does not place on a party an obligation to preserve all documents - only disclosable documents. A document will be disclosable only if it is relevant to proceedings. A mood board may be relevant to any infringed garments designed following consultation of the mood board, but, as set out above, that was not known at the time the mood boards in issue were taken down. The Claimants have in the last few years created thousands of garment designs - only 20 were in issue before me. By the time those 20 garments were identified (because of the allegation that the Defendants had copied them), the mood boards were already long gone. I do not consider that an allegation of infringement of UKUDR (or indeed CUDR) carries with it an obligation immediately to preserve all documents created by the claimant (or putative claimant).
83. I should add for completeness that I do not interpret the disclosure rules as imposing an obligation to create documents - only to preserve existing documents. I therefore cannot readily see how the Claimants can be criticised for not photographing the mood boards in the absence of a duty to preserve them.
84. Finally, although Connor Walker was skilfully cross-examined in relation to the mood boards, that line of enquiry went nowhere: Connor Walker was clear that there was no untoward motivation in removing the mood boards. Further, both Connor Walker and Ms Douek denied that the Claimants’ garments were copied from garments shown on the mood boards, and they were not moved from that position in cross-examination.
85. I understood the parties to be in agreement on the principles to be applied, subject to the small number of controversies I refer to below.
86. Sections 213 and 226 of the Copyright Designs and Patents Act 1988 (CDPA) provide as follows:
“213.- Design right.
(1) Design right is a property right which subsists in accordance with this Part in an original design.
(2) In this Part “design” means the design of the shape or configuration (whether internal or external) of the whole or part of an article.
…
(4) A design is not “original” for the purposes of this Part if it is commonplace in a qualifying country in the design field in question at the time of its creation; and “qualifying country” has the meaning given in section 217(3).
…
226.- Primary infringement of design right
(1) The owner of design right in a design has the exclusive right to reproduce the design for commercial purposes—
(a) by making articles to that design, or
(b) by making a design document recording the design for the purpose of enabling such articles to be made.
(2) Reproduction of a design by making articles to the design means copying the design so as to produce articles exactly or substantially to that design, and references in this Part to making articles to a design shall be construed accordingly.
(3) Design right is infringed by a person who without the licence of the design right owner does, or authorises another to do, anything which by virtue of this section is the exclusive right of the design right owner.
(4) For the purposes of this section reproduction may be direct or indirect, and it is immaterial whether any intervening acts themselves infringe the design right.”
Subsistence
87. Subsistence of UKUDR therefore requires:
i) That the design be “original”;
ii) That the designer or his/her employer is a “qualifying person”, or that articles made to the design were first marketed in a way which qualifies them for protection;
iii) That the design not be excluded from protection (“must fit, “must match” etc);
iv) That the design has been recorded in a design document or an article has been made to the design; and
v) That the term of the design has not expired.
88. The Defendants seriously contested only the first of these - originality. To qualify for UKUDR, a design must be original in two senses:
i) The “copyright sense”; and
ii) Not commonplace within the meaning of section 213(4) of the CDPA.
89. Arnold J (as he then was) addressed originality in the copyright sense in Whitby Specialist Vehicles Limited v Yorkshire Specialist Vehicles and Ors [2014] EWHC 4242 (Pat) at paragraph 43:
“In order for design right to subsist, a design must be “original” in the copyright sense of originating with the author, and not being copied by the author from another: see Farmers Build Ltd v Carrier Bulk Materials Handling Ltd [1999] RPC 461 at 475, 482. In Magmatic v PMS at [84] I expressed the view that the test is whether sufficient skill, effort and aesthetic judgment has been expended on the new design to make it original. During the course of argument in the present case, the question was raised whether “original” should be interpreted in the same manner as the CJEU has interpreted the requirement for originality in the context of copyright, that is to say, as requiring creativity on the part of the designer: see C-429/08 Football Association Premier League Ltd v QC Leisure [2011] ECR I9083, Case C-145/10 Painer v Standard Verlags GmBH [2011] ECR I-12533 and Case C-604/10 Football Dataco Ltd v Yahoo! UK Ltd [EU:C:2012:115]. I shall assume, without deciding, that this is the correct approach.”
90. The parties were agreed that the outcome of this case did not depend on a distinction between the two approaches. As the parties referred to this as “originality in the copyright sense” I have adopted that expression. I have also used the expression “slavishly copied” to mean a design created without originality in the copyright sense.
91. In addition to being original, to qualify for UKUDR, a design must also not be “commonplace in a qualifying country in the design field in question at the time of its creation”. In Neptune, Henry Carr J said (at paragraphs 59 and 60):
“59. In Ocular Sciences [1997] RPC 289 Laddie J explained that the commonplace exclusion applies to “any design which is trite, trivial, common-or-garden, hackneyed or of the type which would excite no peculiar attention in those in the relevant art”. The analysis must be conducted by reference to material “shown to be current in the thinking of designers in the field at the time of creation of the designs”, per Jacob LJ in Lambretta Clothing v Teddy Smith Ltd [2004] EWCA Civ 886; [2005] RPC 6 at [56].
60. Following the amendment to s.213(2), it is more difficult for the claimant to define the shape of a design at a higher level of abstraction than its physical manifestation in the relevant article. As explained by Arnold J in the Whitby case at [45], this makes it harder for the claimant to prove infringement, and also makes it harder for the defendant to prove that the design is commonplace. Nonetheless, the commonplace exclusion remains a useful cross-check on the breadth of a claim to infringement - the more generalised the definition of the design relied upon, the more likely it is to encompass designs which would “excite no peculiar attention in those in the relevant art.””
92. I was also referred to the useful summary of the task facing a defendant set out by HHJ Hacon in Action Storage Systems Limited v G-Force Europe and Anor [2016] EWHC 3151 (IPEC) at paragraph 37:
“(1) A defendant alleging that a design is commonplace should plead the significant features of the design as he contends them to be, the prior art relied on in which those features are said to be found and the date from which each cited item of prior art was available to designers in the relevant design field.
(2) Prior art which renders a design commonplace will not be obscure. The evidential burden rests on the defendant to show that it is not.
(3) A design will be commonplace if it is shown to have been current in the thinking of designers in the field in question at the time of creation of the design, see Lambretta Clothing Co Ltd v Teddy Smith (UK) Ltd [2005] RPC 6 at [56]. Another way of looking at this is that a commonplace design will be one which is trite, trivial, common-or-garden, hackneyed or of the type which would excite no particular attention in those in the relevant design field, see Ocular Sciences Ltd v Aspect Vision Care Ltd [1997] RPC 289, at p.429, approved in Farmers Build Ltd v Carier Bulk Materials Handling Ltd [1999] RPC 13, at pp.477 and 479. A third way of characterising a commonplace design is that it will be ready to hand, not matter that has to be hunted for and found at the last minute, see Ultraframe (UK) Ltd v Eurocell Building Plastics Ltd [2005] EWCA Civ 761; [2005] RPC 36, at [60].
(4) The design field in question is that with which a notional designer of the article in issue is familiar, see Lambretta Clothing at [45].
(5) A design made up of features which individually are commonplace is not necessarily itself commonplace. A new combination of run-of-the-mill features may not be commonplace. See Ocular Sciences at p.429, approved by the Court of Appeal in Farmers Build at p.476 and in Ultraframe at [64].
(6) If the designer of the accused article has expended sufficient skill and labour to make his design original (in the copyright sense) over a single piece of commonplace prior art, he is liable also to have succeeded in creating a design that is not rendered commonplace by that prior art.”
Infringement
93. In relation to infringement, two connected issues require consideration:
i) What is meant by reproduction in section 226(2) of the CDPA?
ii) When is an article produced exactly or substantially to a design within the meaning of section 226(2) of the CDPA?
94. As section 226(2) of the CDPA makes clear, reproduction requires copying. Copying is not defined in the CDPA. The Defendants accepted that a causal connection is required, but submitted that that is not, of itself, sufficient. Rather, counsel for the Defendants submitted that copying is only established where (1) the earlier design is the only cause of the design of the defendant’s article; and (2) the defendant had not already independently conceived of the idea of a design with the majority of the features present in the earlier design before they came across the earlier design. I do not consider that this definition assists me: indeed, I consider it flies in the face of common sense, and conflates two separate but connected elements of section 226(2). To take the two parts of the proffered definition in turn: first, an infringing design can be copied from two earlier designs. Counsel for the Claimants gave the example of a top and a pair of trousers which are separate garments, but which a defendant reproduces, stitched together, to create a jumpsuit. Both the top and the trousers have been copied. The next question is whether the jumpsuit has been made substantially to the design of either the top or the trousers - but both have been copied. It would deprive the CDPA of any force if a copying defendant is excused because s/he copied from more than one source.
95. The second part of the Defendants’ proffered definition is also unhelpful, and is aimed squarely at what are said to be the facts of this case. In most cases, the allegation that the junior product was completely and meticulously designed in the head of its creator prior to seeing the allegedly copied design will be a question of proof, so it does not sit well within a definition of copying. It is also not something that can be independently verified: of course, if Ms Henderson had created a drawing or other record of her thoughts prior to her accessing third party designs, then that could suggest no copying. I add that the definition proposed by the Defendants gets more and more improbable the more complex the article. Many of the garments in issue in this case are not simple dresses. Many have striking and unusual features. It strains belief to suggest that a defendant could have designed all those features in her head, before going looking for third party examples that reproduce most if not all of the features. I therefore reject the Defendants’ definition of copying: this second element may be pertinent to the facts of a given case on the issue of proof, but it is not appropriate to include it in the definition of copying.
96. The Claimants’ counsel put copying like this:
“Our position is that a simple causal link is enough to establish copying. If the claimant’s design has contributed to the defendant’s creation of its design there is copying. Whether or not that amounts to infringement will depend on whether or not - assessed quantitatively and qualitatively - what has been reproduced is sufficient”.
97. I accept that definition as adequate for the purposes of this case.
98. Section 226(2) also requires that the allegedly infringing design be made “exactly or substantially” to the design. This encompasses both exact replicas, and also reproductions which, though not exact, nevertheless create an article that is “substantially” to the design. In Neptune, Henry Carr J said at paragraphs 49 to 53:
“49. The correct approach to considering whether an allegedly infringing article is produced exactly or substantially to the design was set out by Aldous J in C&H Engineering v F Klucznik & Sons Ltd (No.1) [1992] FSR 421 at p 428:
“Under section 226 there will only be infringement if the design is copied so as to produce articles exactly or substantially to the design. Thus, the test for infringement requires the alleged infringing article or articles be compared with the document or article embodying the design. Thereafter the court must decide whether copying took place and, if so, whether the alleged infringing article is made exactly to the design or substantially to that design. Whether or not the alleged infringing article is made substantially to the plaintiff’s design must be an objective test to be decided through the eyes of the person to whom the design is directed.”
50. Neptune submits that this approach is over-simplistic, and that guidance can be found by analogy to infringement by reproduction of a substantial part of a copyright work. It relies upon a well-known passage in Lord Hoffman’s speech in Newspaper Licensing Agency Ltd v Marks & Spencer Plc [2003] 1 AC 551 at pp.559–560, which explains that for the purpose of assessing whether a substantial part has been reproduced in a copyright claim, the Court should consider whether the original skill and labour, which was the reason for conferring copyright protection, has been copied:
“The House of Lords decided in Ladbroke (Football) Ltd v William Hill (Football) Ltd [1964] 1 WLR 273 that the question of substantiality is a matter of quality rather than quantity … But what quality is one looking for? That question, as it seems to me, must be answered by reference to the reason why the work is given copyright protection. In literary copyright, for example, copyright is conferred (irrespective of literary merit) upon an original literary work. It follows that the quality relevant for the purposes of substantiality is the literary originality of that which has been copied. In the case of an artistic work, it is the artistic originality of that which has been copied.”
51. Mr Cuddigan submits that, by parity of reasoning, protection is conferred upon designs because of the originality of their shapes and/or configurations. Where there is copying of ideas expressed in a design which, in their conjoined expression, have involved original design skill and labour, and/or are the expression of the intellectual creation of their author, the resulting article will be substantially made to the design.
52. However, an attempt to equate infringement of UK unregistered design right with infringement of copyright was specifically rejected by the Court of Appeal in Wooley v A Jewellers [2002] EWCA Civ 1119; [2003] FSR 15, where it was held at [19] that:
“19 … As Aldous J observed in the passage I have set out in the Klucznik case, there is a difference between an enquiry into whether the item copied forms a substantial part of the copyright work and an enquiry whether the whole design containing the element which has been copied is substantially the same design as that which enjoys design right protection. The enquiry which the judge carried out was that set out in paragraph 119 of his judgment. At no stage did the judge refer to the different test applicable to design right infringement. On that test, it may not be enough to copy a part, even a substantial part. Regard has to be had to the overall design which enjoys design right. Here the judge was diverted to certain difficult questions arising as to substantiality in copyright infringement which may have no relevance to design right infringement.”
53. In contrast to copyright, it is not an infringement of a UK unregistered design to reproduce “a substantial part” of a design. The importance of this distinction may be illustrated by the facts of the present case. Apart from the features which Neptune has excluded, it relies upon the entirety of each of the articles of furniture which is said to embody the designs in issue, and does not rely upon parts or combinations of parts of such articles. Therefore, it is necessary to consider the differences as well as the similarities between Chichester and Shaker products. It will not be enough to show that a particular feature or combination of features (which in a copyright claim might constitute a substantial part) has been copied. Nor will it be enough to show that Neptune’s key features have been copied, since those features, whether alone or in combination, have not been pleaded as a design right.”
99. I have adopted this approach in my analysis of whether the Defendants’ garments were “made exactly or substantially” to the Claimants’ designs.
100. Of course, it will be necessary to examine individually each of the 20 garment designs relevant for this trial. There were, however, some general issues that applied to all or many of the designs which it is convenient to discuss first.
101. The Defendants accepted that if all 20 designs in issue in this trial were created by the designers identified by the Claimants, then as employees of Sirens Design, Sirens Design was the first owner of any UKUDR and that any such rights had been validly assigned to the First Claimant. However, in their pleadings, the Defendants did not admit that all of the 20 designs were created by the designers identified by the Claimants.
102. Ms Douek gave evidence, unchallenged by the Defendants, that she created five of the 20 garments in issue. For the remaining 15 designs, the Claimants’ evidence was that of Connor Walker, who stated that the designs were created by the named designer. Counsel for the Defendants challenged this evidence, submitting that the Claimants had not satisfied the burden on them to establish that each of the 15 garments was designed by an employee of Sirens Design, or, indeed, the named individuals. The Defendants put forward no evidence to the contrary, and Connor Walker was not moved from her position in cross-examination. An explanation was provided as to why the designers other than Ms Douek had not been called to give evidence: Julia Kasper could not be contacted; Rheanna Donaldson has been in dispute with the Claimants over copyright; Justin Ruddle has set up in competition to the Claimants and Alcy Lynch is similarly no longer employed by Sirens Design.
103. In my judgment, the Defendants’ submission goes nowhere, and I reject it. I accept Connor Walker’s evidence that the 15 garments were designed as she said. It is not incumbent on a UKUDR claimant to provide testimony from the designer of each design, so long as it has evidence of who created the relevant design.
104. Counsel for the Defendants rightly pointed out that the Claimants bear the burden of proof on this issue: they must establish that their designs have not been copied from anyone else. But here, again, the Defendants submitted that the Claimants’ assertion in the pleadings (supported by a statement of truth) was insufficient to meet that burden. First, it was said that the House of CB designers may have copied from the garments shown on the mood boards. I have set out above my comments in relation to the controversy over the Claimants’ mood boards: it goes nowhere. Connor Walker and Ms Douek each gave evidence that Sirens Design designers did not copy the garments shown on the mood boards, from which they were not shifted under cross-examination. Second, counsel for the Defendants submitted that the Claimants should have called evidence from each designer. As I have set out above, an explanation was given for why that had not occurred and I accept that explanation. Additionally, although this was already a lengthy trial, it would have been made longer by the cross-examination of a further four witnesses. Third, counsel for the Defendants submitted that the garments were similar enough to the prior designs identified by the Defendants to raise an inference of copying. As I set out below, I do not consider any of the Claimants’ designs to be sufficiently close to the identified prior designs to give rise to an inference that they were copied. There was here what counsel for the Claimants described as an inconsistency in the Defendants’ position (and which the Defendants described as a squeeze). This was put in the following terms in counsel for the Defendants’ closing skeleton argument:
“We recognise that potential inconsistency, but let us be crystal clear: it is a squeeze on subsistence and infringement. [The Defendants] say the evidence is now clear that there is no material difference between the design process that [the Defendants] follow and the design process that [the Claimants] follow: both are following trends and casting very wide nets for inspiration. They create either physical or digital mood boards from which the trends are visible and from which the designers take ideas and even design features. If that is by its nature copying, then [the Claimants] have no design rights. If it is not copying, then rights do subsist in [the Claimants’] designs but [the Defendants] have not infringed them because they have not copied the designs and the right is sufficiently narrow that the Defendants’ designs fall outside them.”
105. The challenge with this submission is its opening assumption - that there is no material difference between the Claimants’ and the Respondents’ design processes. As will be apparent from this judgment, that position is inconsistent with my findings on the very different ways in which the House of CB and 17 of the Oh Polly garments in issue in this trial were designed. I therefore reject the squeeze on the facts of this case. Further, I reject the squeeze as a question of law, because the test for subsistence of UKUDR is different from the test for infringement. For subsistence, the design cannot have been slavishly copied (ie, it must be the result of the author’s own skill) and it cannot be commonplace (trite, hackneyed etc). The test for infringement is different - the infringing design must have been copied, and it must be made exactly or substantially to the claimed design. I can well understand that a squeeze exists in CUDR, where the infringement test is the flipside of the validity test, but I do not interpret UKUDR as producing a squeeze of that nature.
106. A further point was made that earlier House of CB garments not in issue in these proceedings had been copied from third party designs (I mention above that some were copied from designs by Hervé Léger). Connor Walker owned up to this - but said that any copying stopped when House of CB employed its own designers to design original garments. Dr Branney put together a schedule (which was filed as part of his reply evidence) comparing House of CB garments with third party garments: I was asked to draw the conclusion that if House of CB had copied those designs, I should question whether the 20 pleaded designs in this case were original (in the copyright sense). Connor Walker gave sworn evidence that House of CB’s design strategy had changed some years earlier, and she was not shaken on that position in cross-examination. I therefore reject this submission.
107. The Defendants have provided prior designs against which I am asked to assess originality in the copyright sense, and commonplaceness. I set out below my conclusions in relation to each garment and each prior design. But I do not consider that there is an overall pleading point, evidentiary burden point, disclosure point or squeeze that renders the Claimants’ designs unoriginal in the copyright sense, or commonplace.
108. Significant energy was exerted pre-trial in relation to the 101 prior designs that were said to show that the Claimants’ 20 pleaded designs were commonplace. However, in his closing skeleton argument, counsel for the Defendants dealt with this issue in half a sentence. The rest of that sentence read:
“…but if the court holds that none of the [Claimants’] designs is commonplace over any of the prior art, nonetheless, [the Defendants’] reliance on that prior art in support of their case on commonplace is a useful cross-check on the breadth of [the Claimants’] claim to infringement.”
109. In his oral closing submissions, counsel for the Defendants said this:
“The consideration of originality and commonplace in the context of UKUDR frames the infringement analysis and that has always been our primary goal.”
110. Counsel for the Claimants rejected those statements on the basis that they do not accurately describe the law. I have rejected above the notion of a squeeze on both the facts of this case, and on the law. I have found (below) that none of the Claimants’ designs lacks originality (in the copyright sense) or was commonplace on the basis of the prior designs before the court. In most cases, the prior designs pleaded (on both issues) were very different indeed from the Claimants’ designs, and the pleaded case of lack of originality (in the copyright sense) seemed to me to be hopeless. Similarly, the garments submitted to show that the Claimants’ designs were commonplace were often very different indeed and it struck me as odd that many of these were pleaded in the first place.
111. Because I have found on the basis of the garments pleaded that they do not invalidate the Claimants’ UKUDR, I do not need to decide whether the prior designs were disclosed at the time pleaded by the Defendants and/or how well they were known. I simply record here that it was the Claimants’ position that it could not verify many of the disclosures that were relied on, and that none of the pleaded prior designs was very well-known among designers or customers at the relevant times.
112. I start with the 17 garments before me which the Defendants plead were designed by Ms Henderson.
113. The Defendants now admit that a House of CB or Mistress Rocks garment was referred to in the design process of 9 of the 20 garments in issue in this trial. However, the Defendants submitted that in every case other than C35:
“there has not been copying either because the idea for the garment was conceived before the House of CB / Mistress Rocks garment was referred to or one or more other garments were also referred in the process, or both.”
114. As set out above, I have rejected both these arguments on definitional bases. I also reject the former on the facts before me: as set out above, the only evidence before the Court that the “idea” for the garment was already conceived is that given by Ms Henderson herself, and I have rejected her uncorroborated evidence.
115. By way of example only, take D2. Ms Henderson found a picture of C2 on social media and took a screen shot. That screen shot was uploaded to Trello for the Defendants’ manufacturer to work from. There were no other drawings, patterns or plans. There was a third party garment also shown on the Trello records, but, as I have said, copying more than one design is still copying.
116. In relation to the other eleven garments where the Defendants did not admit referencing one of the Claimants’ garments, the Claimants’ position was that I should infer copying on the balance of probabilities because of:
i) The similarities between the designs in issue suggested that copying had taken place;
ii) Ms Henderson’s approach to design - that is, identifying images of garments that she liked and providing these to the factory to be reproduced;
iii) The incomplete nature of the Trello records and the likelihood that Ms Henderson passed images directly to the manufacturers; and
iv) The likelihood that the House of CB garment alleged to have been copied existed in Ms Henderson’s archive.
117. I discuss the first of these in my analysis of each of the garments, set out below. However, on the facts of this case, absent striking closeness between the designs in relation to features that were not well known, it does not seem to me that the remaining three points put forward by the Claimants are sufficient to prove copying, or to give rise to a presumption which the Defendants are required to rebut. Further, in nearly all the cases where a House of CB/Mistress Rocks garment was not referred to in the design process, there is an identified third party garment which, from the similarity of the garments and from the Trello records, appears to be the basis for the Defendants’ design. That is entirely consistent with my findings in relation to Ms Henderson’s “design process” - only here, instead of copying one of the Claimants’ garments, she has copied a third party garment. That does not necessarily mean that the Claimants’ garment was not also copied, of course, but in the absence of any evidence at all to support the copying of particular garments of the Claimants (including, for example, striking similarities between unusual aspects of the Claimants’ designs), it seems to me most likely that the situation is better explained by the Defendants having copied a third party garment, as opposed to the more roundabout route proposed by counsel for the Claimants.
118. As set out above, I was able to compare images of the relevant garments produced by the Claimants and by the Defendants, and, following the close of oral submissions, was shown each garment by counsel for the Defendants, who alerted me specifically to the differences identified by the Defendants. I also had the actual garments during the preparation of this judgment.
119. The relevant articles of the Design Regulation are as follows:
“Article 3
Definitions
For the purposes of this Regulation:
(a) "design" means the appearance of the whole or a part of a product resulting from the features of, in particular, the lines, contours, colours, shape, texture and/or materials of the product itself and/or its ornamentation;
…
Article 4
Requirements for protection
1. A design shall be protected by a Community design to the extent that it is new and has individual character.
…
Article 6
Individual character
1. A design shall be considered to have individual character if the overall impression it produces on the informed user differs from the overall impression produced on such a user by any design which has been made available to the public:
in the case of an unregistered Community design, before the date on which the design for which protection is claimed has first been made available to the public;
…
2. In assessing individual character, the degree of freedom of the designer in developing the design shall be taken into consideration.
Article 7
Disclosure
1. For the purpose of applying Articles 5 and 6, a design shall be deemed to have been made available to the public if it has been published following registration or otherwise, or exhibited, used in trade or otherwise disclosed, before the date referred to in Articles 5(1)(a) and 6(1)(a) or in Articles 5(1)(b) and 6(1)(b), as the case may be, except where these events could not reasonably have become known in the normal course of business to the circles specialised in the sector concerned, operating within the Community. The design shall not, however, be deemed to have been made available to the public for the sole reason that it has been disclosed to a third person under explicit or implicit conditions of confidentiality.
…
Article 10
Scope of protection
1. The scope of the protection conferred by a Community design shall include any design which does not produce on the informed user a different overall impression.
2. In assessing the scope of protection, the degree of freedom of the designer in developing his design shall be taken into consideration.
…
Article 11
Commencement and term of protection of the unregistered Community design
1. A design which meets the requirements under Section 1 shall be protected by an unregistered Community design for a period of three years as from the date on which the design was first made available to the public within the Community.
2. For the purpose of paragraph 1, a design shall be deemed to have been made available to the public within the Community if it has been published, exhibited, used in trade or otherwise disclosed in such a way that, in the normal course of business, these events could reasonably have become known to the circles specialised in the sector concerned, operating within the Community. The design shall not, however, be deemed to have been made available to the public for the sole reason that it has been disclosed to a third person under explicit or implicit conditions of confidentiality.
…
Article 19
Rights conferred by the Community design
1. A registered Community design shall confer on its holder the exclusive right to use it and to prevent any third party not having his consent from using it. The aforementioned use shall cover, in particular, the making, offering, putting on the market, importing, exporting or using of a product in which the design is incorporated or to which it is applied, or stocking such a product for those purposes.
2. An unregistered Community design shall, however, confer on its holder the right to prevent the acts referred to in paragraph 1 only if the contested use results from copying the protected design.
The contested use shall not be deemed to result from copying the protected design if it results from an independent work of creation by a designer who may be reasonably thought not to be familiar with the design made available to the public by the holder.”
120. Thus, relevantly, CUDR subsists in the appearance of the whole or part of a product resulting from the features of, in particular, the lines, contours, colours, shape, texture and/or materials of the product itself or its ornamentation. Subsistence requires that:
i) the design has been made available to the public in the Community;
ii) the design is new and has individual character;
iii) the design is not excluded from the scope of CUDR (“must fit”, “technical function”, “public policy” etc); and
iv) the three year term of the right has not expired.
121. The substantive dispute in this case was in relation to individual character - do the designs of the 20 garments produce a different overall impression on the informed user than the 108 prior designs identified by the Defendants?
122. The law on individual character was considered in detail in Cantel Medical (UK) Limited v ARC Medical Design Limited [2018] EWHC 345 (Pat) (Cantel) at paragraphs 169 to 182. Relying on the judgment of the General Court in Case T-525/13 H&M Hennes & Mauritz BV & Co KG v OHIM EU:T:2015:617; [2015] ECDR 20 (H&M Hennes) HHJ Hacon (sitting as a Judge of the Patents Court) provided a convenient six step summary at paragraphs 181 and 182:
“181. I here adapt the four-stages prescribed by the General Court in H&M Hennes for assessing the individual character of a Community design to the comparison of an RCD with an accused design, adding other matters relevant to the present case. The court must:
(1) Decide the sector to which the products in which the designs are intended to be incorporated or to which they are intended to be applied belong;
(2) Identify the informed user and having done so decide (a) the degree of the informed user’s awareness of the prior art and (b) the level of attention paid by the informed user in the comparison, direct if possible, of the designs;
(3) Decide the designer’s degree of freedom in developing his design;
(4) Assess the outcome of the comparison between the RCD and the contested design, taking into account (a) the sector in question, (b) the designer’s degree of freedom, and (c) the overall impressions produced by the designs on the informed user, who will have in mind any earlier design which has been made available to the public.
182. To this I would add:
(5) Features of the designs which are solely dictated by technical function are to be ignored in the comparison.
(6) The informed user may in some cases discriminate between elements of the respective designs, attaching different degrees of importance to similarities or differences. This can depend on the practical significance of the relevant part of the product, the extent to which it would be seen in use, or on other matters.”
123. In relation to the informed user, HHJ Birss QC (sitting as a Judge of the Patents Court) (as he then was) set out a helpful summary of the informed user in Samsung Electronics (UK) Limited v Apple Inc [2012] EWHC 1882 (Pat) (Samsung v Apple) which was expressly approved by the Court of Appeal at [2012] EWCA Civ 1339 at paragraph 10 (per Sir Robin Jacob, with whom Longmore LJ and Kitchin LJ (as he then was) agreed). His Honour referred to three cases decided by the Court of Justice and the General Court: Case C-281/10 P PepsiCo Inc v Grupo Promer Mon-Graphic SA [2012] FSR 5 (PepsiCo); Case T-9/07 Grupo Promer Mon Graphic SA v OHIM EU:T:2010:96 [2010] ECDR 7 (Grupo Promer); and Case T-153/08 Shenzhen Taiden Industrial Co Ltd v OHIM EU:T:2010:248 (Shenzhen):
“Samsung submitted that the following summary characterises the informed user. I accept it and have added cross-references to the cases mentioned:
i) He (or she) is a user of the product in which the design is intended to be incorporated, not a designer, technical expert, manufacturer or seller (PepsiCo paragraph 54 referring to Grupo Promer paragraph 62; Shenzhen paragraph 46);
ii) However, unlike the average consumer of trade mark law, he is particularly observant (PepsiCo paragraph 53);
iii) He has knowledge of the design corpus and of the design features normally included in the designs existing in the sector concerned (PepsiCo paragraph 59 and also paragraph 54 referring to Grupo Promer paragraph 62);
iv) He is interested in the products concerned and shows a relatively high degree of attention when he uses them (PepsiCo paragraph 59);
v) He conducts a direct comparison of the designs in issue unless there are specific circumstances or the devices have certain characteristics which make it impractical or uncommon to do so (PepsiCo paragraph 55).
I would add that the informed user neither (a) merely perceives the designs as a whole and does not analyse details, nor (b) observes in detail minimal differences which may exist (PepsiCo paragraph 59).”
124. Counsel for the Defendants also referred me to a decision of the General Court in Case T-90/16 Thomas Murphy v EUIPO at paragraph 43:
“…the individual character of a design results from an overall impression of difference, or lack of ‘déjà vu’, from the point of view of an informed user in relation to any previous presence in the design corpus, without taking account of any differences that are insufficiently significant to affect that overall impression, even though they may be more than insignificant details, but taking into account of differences that are sufficiently marked so as to produce dissimilar overall impressions (judgment of 29 October 2015, Roca Sanitario v OHIM — Villeroy & Boch (Single control handle faucet), T-334/14, not published, EU:T:2015:817, paragraph 16).”
125. Briefly, I want to say something about the use of the expression “déjà vu”. Whilst it has achieved some prominence in decisions of the Invalidity Division and Boards of Appeal at the EUIPO, and has been recognised by the General Court, I do not find it helpful. The difficulty with the expression, when used in English, is that it is not precise enough. The French expression literally means “already seen”. When used in English, the expression has a wider meaning - a feeling of having already experienced the present situation, or “the strange feeling that in some way you have already experienced what is happening now.” None of these definitions sits well with the task assigned to the informed user.
126. In PepsiCo, the Court of Justice set out the task for the informed user in coming to an overall impression - I have set out above HHJ Hacon’s six-stage test from Cantel. It is a complex, multifactorial assessment. It is not simply a question of having seen the design before, and certainly not of a “strange feeling”. Further, the notion of déjà vu is more akin to imperfect recollection from trade mark law (rejected by the Court of Justice for design law), and places the informed user too close to the reasonably circumspect consumer from trade mark law. This may result in findings that insufficient differences will be noticed, leading to too few designs being valid, but those that are valid having too wide a scope of protection.
127. The Court of Justice has never approved the expression in the context of design law: it is unhelpful, and should be abandoned.
128. The test for infringement of CUDR is the flipside of that aspect of the validity test that relates to individual character. The law is therefore the same as set out above. In addition, a claimant must prove copying. The Court of Justice addressed copying in C-479/12 H Gautzsch Großhandel GmbH v Münchener Boulevard Möbel Joseph Duna GmbH [2014] RPC 28 at [37] to [44], concluding:
(a) the onus of proving copying lies on the holder of the design right in issue;
(b) the onus of proving independent creation lies on the alleged infringer; and
(c) the onus of proving copying may be lightened or reversed, in accordance with the rules of national law, where it is otherwise likely to be impossible or excessively difficult to prove copying.
129. As set out above, I do not consider that any of the 20 designs is invalid on the basis of not being original in the copyright sense, or being commonplace. It does not follow, however, that each of the 20 designs creates a different overall impression on the informed user than the prior designs relied on. Rather, I must assess each of the designs against the prior designs relied on and reach a conclusion as to whether or not the two designs produce on the informed user a different overall impression.
130. Applying the six-stage test set out by HHJ Hacon in Cantel:
(1) Decide the sector to which the products in which the designs are intended to be incorporated or to which they are intended to be applied belong:
The sector here is women’s clothing;
(2) Identify the informed user and having done so decide (a) the degree of the informed user’s awareness of the prior art and (b) the level of attention paid by the informed user in the comparison, direct if possible, of the designs;
The informed user is a user of women’s clothing, who has the characteristics identified by HHJ Birss QC (as he then was) in Samsung v Apple. The informed user in this case is aware of bodycon and bandage dress fashions, including garments worn by celebrities and shown on social media. The informed user would compare the designs directly.
(3) Decide the designer’s degree of freedom in developing the design;
A garment has to be designed to be worn and to fit the body - but within those constraints, the designer has almost complete design freedom, as is apparent from the array of differences between the designs at issue in these proceedings.
(5) Features of the designs which are solely dictated by technical function are to be ignored in the comparison.
It was not contended that any feature of any of the designs was solely dictated by technical function.
131. In relation to step 6 as enumerated by HHJ Hacon in Cantel, it is clear that some aspects of a garment will strike the informed user as more important. In this case, all the dresses were primarily photographed from the front: whilst photographs of the back aspect were included on both sides’ websites, they were not the primary image. From that, I take it that the front of each garment will be of primary importance to the informed user in most cases. Differences on the back of the garment will still be noticed, but they will not be given the importance of the front of the garment.
132. I also need to say something about fabrics. The Oh Polly business is entirely online. Whilst the House of CB business does have some bricks and mortar stores, the bulk of its business is also online. That means that the majority of purchasing decisions are being made based on the images of the garments shown on the websites. Of course, the evidence also demonstrated that something like 30% of online sales are returned. Consumers will return a garment they have purchased on-line if, for example, they do not like the texture or fit. But 70% of sales were not returned, from which I conclude that the experience of the garment on receipt matched the expectation from viewing the garment on the website. It seems to me that what matters is what the dress looks like when worn. Thus, where the images with which I have been provided (which were taken from the parties’ websites) show very similar looking dresses, it will matter less to the informed user that, on a hanger, the dresses do not look as similar.
133. I return to step 4 and the remainder of step 6 of HHJ Hacon’s six-stage test in my analysis of each of the 20 designs, below.
134. The Defendants submitted that their case on ownership is the same as in relation to ownership of UKUDR. My conclusions are therefore the same: I reject the Defendants’ submissions.
135. The Defendants did not make any additional arguments in relation to copying for CUDR. My conclusions on copying are therefore the same as for UKUDR.
136. There was argument before me as to whether the infringement test for UKUDR was more or less stringent than that for CUDR. That is a sterile debate. The test of “produced exactly or substantially to the design” is a different test from “does not produce on the informed user a different overall impression”. I have therefore applied both tests separately in relation to each of the relevant garments.
137. C2 was designed by Justin Ruddle on or about 14 July 2017, and first displayed on House of CB social media accounts on 13 January 2018. Ms Henderson saw C2 on YouTube, and took a screen shot, which she saved to her laptop. Ms Henderson provided email instructions to Mr Henderson with images of C2 and a third party dress called the Meshki dress. Mr Henderson used software to remove the buckle shown on the C2 image, in accordance with Ms Henderson’s instructions. D2 was first offered for sale on the Oh Polly website on 19 March 2018.
C2 D2
138. The significant features of C2 were said to be:
“This is a ruched body hugging dress. The dress has horizontal ruching all over. The top has an off shoulder design with wide ruched straps that meet in the middle across the chest. These straps continue across the bust [although the Defendants submitted that the straps do not continue across the bust. They are broken up with a buckle], leaving a triangular shaped cut out to the centre bust [although the Defendants submitted that the cut out is more “dome” shape than triangular]. The waist sits high and the ruching comes from both side seams. The skirt sits just above the knee. The back is off shoulder [and the Defendants noted that C2 has a back seem]. This item comes in Wine silky jersey fabric.”
139. The features of colours and materials relate to CUDR only.
140. In relation to originality in the copyright sense (UKUDR) and the closest prior design (CUDR), the Defendants relied on a dress designed by Cushnie et Ochs, shown here worn by Karlie Kloss in an image taken from The Daily Mail Online dated 22 June 2016:
141. In addition, the Defendants relied on a further six dresses to suggest that C2 was commonplace.
142. Having reviewed the images, in my judgment, C2 was not slavishly copied from the Cushnie et Ochs dress and does not produce the same overall impression on the informed user. The Cushnie et Ochs dress is not a ruched dress - the fabric has a flat and smooth finish. It is not bodycon. The straps, which comprise flat panels of tapered fabric, start as a wide strap below the bust and pass over the bust tapering to a very thin strap passing over the opposite arm - the straps overlay in the middle. The dress sits below the knee. The underbust cut out is smaller than C2, exposing less of the underbust. In addition in relation to CUDR, the dress is white and has a flat, smooth finish.
143. I have also reviewed the other dresses which the Defendants consider not to be as close to C2. None of them makes C2 commonplace - they are all different designs.
144. C2 is not invalid in relation to UKUDR or CUDR.
145. The Defendants’ case is that, in 2018, underbust cut-out dresses and Bardot (off the shoulder) styles were on trend. Ms Henderson saw C2 on YouTube, and took a screen shot, but wanted to keep the Bardot strip of fabric across the bust, rather than interrupt it with a buckle. Ms Henderson says she wanted to create something similar to the Cushnie et Ochs dress (which was from 2 years earlier), but in a ruched fabric. This is not what the documentary record shows, and so I discount her evidence of her intention. The documentary record shows that she sent an email on 14 January 2018, the day after the launch of C2, attaching images of C2 and the Meshki ruched dress to Dr Branney and Mr Henderson, giving instructions in her email to ignore the buckle over the bust, and asking Dr Branney/Mr Henderson to edit it out before he uploaded the image of C2 to Trello. She did not send or refer to the Cushnie et Ochs dress.
146. The Meshki dress is shown here:
147. I have no hesitation in finding that D2 was copied from C2. As is clear from her email and the Trello records, Ms Henderson downloaded an image of C2 from YouTube, and instructed that it be sent to the factory to be reproduced.
148. The main difference between C2 and D2 is the buckle. The Defendants also pointed to the fact that D2 has no cut out under the arm, some of the ruching is done differently, D2 has no seam or zip on the back panel, and the Bardot panel on the rear of D2 is differently constructed. The Defendants also point to the different colours (wine vs dark rose) as a further difference for the purposes of CUDR.
149. In my judgment, none of these differences individually or collectively is sufficient to render D2 other than an article made substantially to the design of C2. Similarly, the differences would not strike the informed user so as to produce a different overall impression. In coming to those conclusions, I have undertaken a detailed side by side analysis of the dresses themselves.
150. D2 infringes the UKUDR and CUDR in C2.
151. C3 was designed by Justin Ruddle on or about 26 June 2017, and first offered for sale on the Mistress Rocks website in mocha on 20 November 2017 and in red on 29 January 2018. On 15 January 2018, Ms Henderson found images of third party garments (not including C3), annotated them, and sent them to Mr Henderson. The next day, Ms Johnston added to Trello her sketch based on the images provided by Ms Henderson. D3 was first offered for sale on the Oh Polly website on 23 March 2018.
C3 D3
152. The significant features of C3 were said to be:
“This is a smooth strappy body hugging maxi dress. The front has a scoop plunge neck, it has skinny shoulder straps and a backless cut [although the Defendants submitted that the back is high waisted rather than backless]. The straps extend around the shoulder and join under the arm [the Defendants said at the bust]. The skirt of the dress sits at the waist at the back and it has a back vent. This item is sold in Deep Mocha and Red silky jersey.”
153. The features of colours and materials relate to CUDR only.
154. In relation to originality in the copyright sense (UKUDR) the Defendants relied on a Calvin Klein dress worn by Kate Moss to the 1995 Met Gala Dinner. In relation to the closest prior design (CUDR), the Defendants relied on an Abyss by Abby dress. Both dresses are shown here:
Calvin Klein Abyss by Abby
155. In addition, the Defendants relied on a further two dresses to suggest that C3 was commonplace.
156. Having reviewed the images, in my judgment, C3 was not slavishly copied from the Calvin Klein dress, and does not create the same overall impression on the informed user as the Abyss by Abby dress. The Calvin Klein dress is a loose-fitting maxi gown. It is not body hugging on either the hips or the legs - it is not stretchy, and has a column fit. It has flat, square-shaped straps sown into the seams of the bust line and a straight across cut neckline. The dress appears to be cut on the bias. The Abyss by Abby dress is a floor-length fishtail gown, in which the skirt flares out mid-thigh. The gown is sufficiently long that it pools on the floor. In addition in relation to CUDR, the dress is red (and so differs from the deep mocha version of C3). Given those findings, I do not need to determine whether these dresses were disclosed on the dates asserted by the Defendants.
157. I have also reviewed the other dresses which the Defendants consider not to be as close to C3. None of them makes C3 commonplace - they are all different designs.
158. C3 is not invalid in relation to UKUDR or CUDR.
159. The Defendants’ case is that there is nothing in the documentary record to suggest that C3 was ever referred to by the Defendants, let alone copied. It was not put to Ms Johnston that she was mistaken in her written evidence when she said that she did not refer to C3 in her design process. Whilst Ms Henderson had provided some images prior to Ms Johnston designing D3, those images did not include C3, and it was not put to Ms Henderson that she had referenced C3. I do not consider that C3 and D3 are close enough for an inference of copying to require rebutting. I accept the Defendants’ submission that C3 is a generic style of dress, with various prior designs encapsulating a number of the features. C3 and D3 also differ, for example in the length (and the slit up the back), the neckline and the placement of the straps (on C3, the straps join at the side bust, whereas on D3, they join at the lower back). Perhaps the red and deep mocha colours were unusual (for the purposes of CUDR), but D3 was never sold in those colourways.
160. I therefore do not consider that Ms Johnston or Ms Henderson copied C3. Further, D3 is not an article made substantially to the design of C3 (UKUDR) nor does it produce on the informed user the same overall impression (CUDR). The UKUDR and CUDR infringement cases therefore fail in relation to C3.
161. C4 was designed by Georgina Douek on or about 3 February 2017, and first offered for sale on the House of CB website in cream on 7 August 2017. Shortly thereafter, Ms Henderson saw C4 on Pinterest, and copied the image. The C4 image was uploaded to Trello, along with an image of a top called Marysia. D4 was finalised on 12 April 2018 and offered for sale on the Oh Polly website from 10 May 2018. C4 was available in cream, burgundy and black. D4 was available in cream and sage.
C4 D4
162. The significant features of C4 were said to be:
“This item is a one shoulder knotted bandeau top [although the Defendants said C4 is not knotted, it is stitched]. It is constructed to sit around the bust with one shoulder strap on the left shoulder that connects to the front of the bandeau. The bandeau knots at the centre and the fabric continues down to hang vertically under the left bust ending at the waistline [the Defendants said the hips]. It has one thick band that sits across the back [the Defendants said C4 has a central back zip]. This item is available in cream, burgundy and black and is a knit fabric [the Defendants said it has distinctive ribbing].”
163. The features of colours and materials relate to CUDR only.
164. In relation to originality in the copyright sense (UKUDR), the Defendants relied on two prior designs, a Jacquemus design from the Fall 2016 collection and a top from the Spring 2013 collection of Maison Martin Margiela. Both are shown below. In relation to the closest prior design (CUDR) the Defendants relied only on the Margiela garment:
Jacquemus Margiela
165. In addition, the Defendants relied on a further two garments to suggest that C2 was commonplace.
166. Having reviewed the images, in my judgment, C4 was not slavishly copied from the Jacquemus or Margiela garments, and does not create the same overall impression on the informed user as the Margiela garment. The Jacquemard garment is a tartan top comprising a length of fabric that has been tied around the model. The bandeau shape is twisted at the centre bust - one wide end extends from the twist up and over the wearer’s right shoulder. The cloth covers the left shoulder completely. A short tail of fabric sits under the centre bust. The Margiela garment is a twist bandeau top, with a deep bandeau shape that extends from the underarm down to the waist. It twists at the centre bust in a wide and deep twist, and has a short tail of fabric. The upper tie end is stiffened and stands up in front of the left shoulder in a fan-like shape.
167. I have also reviewed the other garments which the Defendants consider not to be as close to C4. None of them makes C4 commonplace - they are all different designs.
168. C4 is not invalid in relation to UKUDR or CUDR.
169. The Defendants’ case is that, in 2018, bralettes were in style and the single-shoulder, knotted effect appeared to be a particular trend. Ms Henderson found an image of C4 on Pinterest. An image of C4 was uploaded to Trello, along with an image of the Marysia garment, shown here:
170. Whilst Ms Henderson says that she saw the Marysia garment before seeing the C4 garment, there is no independent evidence to support that assertion, and so I discount it. Both images were uploaded to Trello with design instructions added using Microsoft Paint. The Defendants’ case is that D4 was not copied from C4 because Ms Henderson uploaded two reference garments and gave design instructions by reference to both of them. For the reasons set out above, I reject the argument that there is no copying where two or more prior designs are copied. I have no hesitation in concluding that D4 was copied from C4.
171. The main differences between C4 and D4 are as follows:
i) D4 is knotted differently;
ii) D4 has elastic in the shoulder strap and horizontal ruching;
iii) The shoulder strap on C4 runs to a three-quarter position on the wearer’s back whereas the shoulder strap on D4 runs to a full side position on the wearer’s back; and
iv) D4 has stretchy fabric whereas C4 (which is ribbed) has no give/stretch.
172. In my judgment, none of these differences individually or collectively is sufficient to render D4 other than an article made substantially to the design of C4. Similarly, the differences would not strike the informed user so as to produce a different overall impression. Similarly, the differences between the cream versions would not strike the informed user so as to produce a difference overall impression. C4 was available in cream, burgundy and black. D4 was available in cream and sage. In relation to CUDR, D4 in cream infringes C4 in cream, but the sage colour would strike the informed user as different - it would produce on that informed user a different overall impression.
173. D4 infringes UKUDR and CUDR (in cream only) in C4.
174. C7 was designed by Alcy Lynch on or about 13 September 2016, and first offered for sale on the House of CB website on 10 April 2017. Ms Henderson found some old images online of Liz Hurley and Mariah Carey in strappy dresses, and wanted a simple strappy dress for Oh Polly. She annotated both images in Microsoft Paint and emailed the images and instructions to Dr Branney and Mr Henderson. Mr Henderson uploaded them to Trello. D7 was first made available for sale on the Oh Polly website on 20 December 2017. D7 was sold in Nude and olive green.
C7 D7
175. The significant features of C7 were said to be:
“This is a strappy back knee length dress [although the Defendants said it finishes above the knee]. It has a smooth plain finish. It has a bib front and slim spaghetti straps. The straps cross over at the back and tie at the back. The skirt is knee length. The skirt falls to a slight V-shape at the back centre waist below the ending of the straps. This item is sold in Nude. Presented in a smooth stretch light crepe [although the Defendants said it is not stretchy].”
176. The features of colour and materials relate to CUDR only.
177. In relation to originality in the copyright sense (UKUDR) the Defendants relied on a Marc Jacobs dress from his 1996 Spring/Summer collection. In relation to the closest prior design (CUDR), the Defendants relied on a 2016 areyouami dress. Both garments are shown here:
Marc Jacobs areyouami
178. In addition, the Defendants relied on a further dress to suggest that C7 was commonplace.
179. Having reviewed the images, in my judgment, C7 was not slavishly copied from the Marc Jacobs dress, and does not create the same overall impression on the informed user as the areyouami dress. The main similarity between C7 and the Marc Jacobs dress is that both are strappy, backless, knee-length dresses. The Mark Jacobs dress has pockets, and long straps that tie around the waist at the front. The areyouami dress is a mini dress. It has a very narrow cut across the front so that the side body and side bust are visible. It has one strap that extends from the front of the bodice around the arm and fastens to the top part of a triangular shaped strap that extends from the side bust of the garment. Another strap is fastened to the bottom of a triangular shaped strap. The skirt sits high and straight across at the back waist. The skirt is very short, ending under the buttocks. Given my findings, it is not necessary for me to determine whether the areyouami dress was disclosed on the date asserted by the Defendants.
180. I have also reviewed the other dress which the Defendants consider not to be as close to C7. It does not make C7 commonplace - it is a different design.
181. C7 is not invalid in relation to UKUDR or CUDR.
182. The Defendants’ case is that there is nothing in the documentary record to suggest that C7 was ever referred to by the Defendants, let alone copied. Rather, the documentary record shows Ms Henderson referencing two other strappy dresses as worn by Liz Hurley and Mariah Carey, shown here:
Liz Hurley Mariah Carey
183. I also do not consider that C7 and D7 are close enough for an inference of copying to require rebutting. I accept the Defendants’ submission that C7 is a generic style of strappy dress, with various prior designs encapsulating a number of the features. C7 and D7 also differ, for example in the length, the neckline and the placement of the straps. In addition (for the purposes of CUDR), D7 was also sold in olive green.
184. I therefore do not consider that Ms Henderson copied C7: she copied the Liz Hurley and Mariah Carey strappy dresses. D7 is not an article made substantially to the design of C7 (UKUDR) and does not produce on the informed user the same overall impression (CUDR). The UKUDR and CUDR infringement cases therefore fail in relation to C7.
185. C9 was designed by Rheanna Donaldson on or about 4 March 2016, and first offered for sale on the House of CB website on 26 September 2016. On 9 June 2017, Ms Henderson found some (non-House of CB) images online and sent them to Dr Branney and Mr Henderson, together with directions for length and general shape. D9 was first offered for sale on the Oh Polly website on 3 August 2017.
C9 D9
186. The significant features of C9 were said to be:
“This is a high necked long sleeve mini dress. It has a round neck and a deep waistband at the front. It has two slits running down the front of the skirt from the bottom of the waistband to the hem at the position of the centre of the thighs. There are six rectangular loops sewn into the edges of the slits and a ribbon with metal ends is laced through these loops with the ends of the ribbon hanging down below the hem. At the front of the top there is a central panel of fabric which ends at the top of the slits [the Defendants said it stops at the waist band]. The side panels loop under the arms and continue at the back of the top and down the side of skirt. There is a zip down the entire length of the centre back of the dress. It has narrow panels of fabric either side of the centre back at the top, a deep waistband and the skirt also has narrow panels of fabric either side of the centre back. Presented in a stretch crepe. This garment is sold in black [although the Defendants said that C9 is shown in navy on the House of CB website and in the sample provided].”
187. The features of colours and materials relate to CUDR only.
188. In relation to originality in the copyright sense (UKUDR) the Defendants relied on a garment worn by Gigi Hadid as shown in The Hollywood Reporter on 20 June 2016, although the Claimants say that that claim had already been abandoned by the Defendants, in which case, the assertion fails. However, in case I am wrong in that, I will consider the submission as if it had not been abandoned. In relation to the closest prior design (CUDR), the Defendants relied on a 2016 Sublime Havoc dress which Ms Henderson had saved to her Dropbox in 2016. Both dresses are shown here:
Gigi Hadid Sublime Havoc
189. In addition, the Defendants relied on a further six dresses to suggest that C9 was commonplace.
190. Having reviewed the images, in my judgment, C9 was not slavishly copied from the Gigi Hadid garment, and does not create the same overall impression on the informed user as the Sublime Havoc garment. The Gigi Hadid garment is a zip-up playsuit. It has a bodycon fit and long sleeves. It has a centre zip that extends from the neckline to below the waist. The legs are short and end at the top of the thigh. There is a lace up design to the front side of each thigh. The lace up is in a zigzag pattern. The sleeves have embroidery. The Sublime Havoc dress is a sleeveless mini dress. The top has a deep plunge neckline with a lace up detail to the centre bust that narrows as it extends down to the under bust. The bust is visible through the lace up detailing.
191. I have also reviewed the other dresses which the Defendants consider not to be as close to C9. They do not make C9 commonplace - they are different designs.
192. C9 is not invalid in relation to UKUDR or CUDR.
193. The Defendants’ case is that there is nothing in the documentary record to suggest that C9 was ever referred to by the Defendants, let alone copied. Rather, the documentary record shows Ms Henderson copying two other dresses, images of which she annotated and sent to Dr Branney and Mr Henderson. They are shown here:
194. Mr Henderson uploaded the images to Trello. I do not consider that C9 and D9 are close enough for an inference of copying to require rebutting. There are similarities between the dresses, but there are also significant differences, and the similarities are repeated elsewhere amongst the prior designs. The differences between C9 and D9 include the neckline, the way the loops for the lacing are made, the different fabric for the laces, the toggles for the laces, and, importantly, the additional pieces of material in C9 that hang below the lacing.
195. I therefore do not consider that Ms Henderson copied C9. D9 is not an article made substantially to the design of C9 (UKUDR) nor does it produce on the informed user the same overall impression (CUDR). The UKUDR and CUDR infringement cases therefore fail in relation to C9.
196. C12 was designed by Georgina Douek based on an earlier jumpsuit design which she used as a starting point to design C12 - it had a similar bib front. C12 was designed on or about 18 November 2016. Changes to the design were made, and the garment was first offered for sale on the House of CB website on 29 May 2017. Just seven days later, on 5 June 2017, Ms Henderson sent an email with design instructions, attaching images of two House of CB garments. Mr Henderson created the Trello record that same day. D12 was first offered for sale on the Oh Polly website on 15 August 2017.
C12 D12
197. The significant features of C12 were said to be:
“This is a sleeveless playsuit with a bib front with a straight across neck, wide straps that pass over the shoulders and cross the back passing through loops at the bust line of the bib and meeting at the centre of the back with a metal buckle. It has a wide waistband and the shorts have wide legs with a zip down the centre back seam and deep darts from the waistband at the back of the shorts. This garment is sold in white stretch lightweight crepe [although the Defendants said the sample was 100% polyester and had no stretch].”
198. The features of colours and materials relate to CUDR only.
199. In relation to originality in the copyright sense (UKUDR), the Defendants relied on a 2015 Jollyvision garment. In relation to the closest prior design (CUDR) the Defendants relied on the 2016 Zachary garment. Both are shown here:
Jollyvision Zachary
200. In addition, the Defendants relied on a further two garments to suggest that C12 was commonplace.
201. Having reviewed the images, in my judgment, C12 was not slavishly copied from the Jollyvision garment, and does not create the same overall impression on the informed user as the Zachary garment. Both the Jollyvision and the Zachary garments are dresses, not playsuits. They are nothing alike.
202. I have also reviewed the other garments which the Defendants consider not to be as close to C12. None of them makes C12 commonplace - they are all different designs.
203. C12 is not invalid in relation to UKUDR or CUDR.
204. The Defendants’ case is that on 23 April 2017, Ms Henderson uploaded reference images to the Trello record for a different Oh Polly dress (Style 1057) - none of these was of a House of CB garment. Style 1057 is shown here:
205. This was 6 weeks before C12 was released. Style 1057 embraced cut outs and strap features which were trending that summer, as were garments which showed the side bust. Later, Ms Henderson uploaded an image of C12 and a further House of CB garment (shown below) to Paint and included some design instructions.
206. She referred to an Oh Polly garment for the colour, and to a further Oh Polly garment for the length. The Defendants’ case is that D12 was not copied from C12 because (a) Ms Henderson had already made a start with Style 1057; (b) two garments were referenced; (c) instructions were given to change the C12 design; and (d) other Oh Polly garments were referenced. I reject (a) because, again, there is no corroborating evidence that D12 was based on Style 1057 and style 1057 is obviously different. For the reasons set out above, I also reject arguments (b) to (d). I have no hesitation in concluding that D12 was copied from C12.
207. Additional differences between C12 and D12 were said to be as follows:
i) C12 has a prominent side bust, but D12 has no side bust; and
ii) D12 is made of stretchy fabric, whereas there is no stretch at all in the sample of C12.
208. To my mind, the main difference is in the legs of the garment - Ms Douek conceded that her design was aimed at being more of a lingerie style: D12 is more bodycon. That said, in my judgment, the differences pointed out by the Defendants’ counsel are insufficient, individually or collectively, to render D12 other than an article made substantially to the design of C12. Similarly, the differences (including the difference in colour) would not strike the informed user so as to produce a different overall impression. Some of the differences between C12 and D12 are more apparent in the actual garments on a hanger, but others (such as the cut of the legs) are not. As worn, they are very similar indeed.
209. D12 infringes UKUDR and CUDR in C12.
210. C13 was designed by Justin Ruddle on or about 4 April 2017. The garment was first offered for sale on the Mistress Rocks website on 28 August 2017. Just 13 days later, on 10 September 2017, Ms Henderson sent an email with design instructions, attaching an image of C13 and an AYM Studio garment. D13 was offered for sale on the Oh Polly website on 19 October 2017.
C13 D13
211. These photographs appear to suggest that the hemline of D13 is different to that of C13 - it is not. Counsel for the Defendants confirmed to me during the review of the physical garments that the apparent difference is a result of how the dresses have been photographed.
212. The significant features of C13 were said to be:
“This item is a long sleeved ruched mini dress. This dress has long sleeves and a low scoop neck ending in a soft V-shape. The front is ruched from the middle and has 2 rows of eyelets running down the centre from the V of the neckline to the hem. There is a lace threaded through the eyelets in a criss-cross fashion. The back of the dress is plain. The skirt sits mid thigh[.] This item is sold in Black stretchy mesh chiffon fabric [although the Defendants said it is chiffon, not mesh].”
213. The features of colours and materials relate to CUDR only.
214. In relation to originality in the copyright sense (UKUDR), the Defendants relied on a John Galliano dress worn by Kim Kardashian to a 2016 awards show, shown below, although the Claimants said that that claim had already been abandoned by the Defendants, in which case, the submission fails. However, in case I am wrong in that, I will consider the submission as if it had not been abandoned. In relation to the closest prior design (CUDR) the Defendants relied on a 2017 Aym.studio garment, also shown below:
John Galliano Aym.studio
215. In addition, the Defendants relied on a further four garments to suggest that C13 was commonplace.
216. Having reviewed the images, in my judgment, C13 was not slavishly copied from the Galliano dress, and does not create the same overall impression on the informed user as the Aym.studio garment. The Galliano dress is a long sleeved dress made from sheer fabric. It has a wide off shoulder cut with extra-long sleeves that gather up along the length of the arm. The garment has a central panel made from more sheer fabric with a centre drawstring pull feature that enables the hem to be pulled shorter. The long ties from the drawstring dangle below the hem and between the legs, down to the mid-calf. The Aym.studio dress is a long sleeved off the shoulder dress, with the shoulders cut very wide so as to create a very wide V-shaped neckline. The centre front has a panel with lacing details that extends vertically down to the hem. The lace up ties at the centre bust can be pulled tight or left loose. The bust is pulled together and done up at the front with the aid of the ties. The loops are sewn into the seams of the panel. The back has a wide and low V-shape.
217. I have also reviewed the other garments which the Defendants consider not to be as close to C13. None of them makes C13 commonplace - they are all different designs.
218. C13 is not invalid in relation to UKUDR or CUDR.
219. The Defendants’ case is that Ms Henderson started with the vintage Galliano dress worn by Kim Kardashian, shown above. As I have said, I am unable to accept Ms Henderson’s evidence unless corroborated, and here it is not. There is documentary support for the fact that Ms Henderson emailed three images of C13 and an image of the Aym.studio dress to Dr Branney and Mr Henderson with some design instructions. These were uploaded to Trello. The Galliano dress was neither emailed nor uploaded. Counsel for the Defendants submitted that Ms Henderson did not copy C13 because her idea for D13 started with the Galliano dress, and she wanted to do something closer to that dress than to C13. As I have said, I do not accept Ms Henderson’s evidence where it is not otherwise corroborated. Further, her position is inconsistent with two facts for which there is clear evidence: (a) she did not send the Galliano dress to Dr Branney and Mr Henderson nor upload it to Trello; and (b) D13 looks much more like C13 than it looks like the Galliano dress. It also seems significantly more likely that Ms Henderson was inspired by C13 which was launched 13 days earlier on a website she was monitoring closely than she was by a dress worn at an awards ceremony the previous year. I therefore reject her version of events. Further, it is said that the “key” off the shoulder feature was not copied from C13, but was taken from the Aym.studio dress. There is no corroborative evidence for this. I also doubt that this is a “key” feature of D13, because the images with which I have been provided all show the model wearing D13 on the shoulders. I reject the Defendants’ submissions, and have no hesitation in concluding that D13 was copied from C13.
220. The further differences between C13 and D13 are said to be as follows:
i) D13 has a deep V-shaped neck, rather than a scoop shape, with the neckline going below the bust;
ii) The ruching on D13 differs; and
iii) D13 was specifically cut to have a defined asymmetric hem, whereas in its final form C13 has a flat hem.
221. Having inspected both garments, I reject that third submission. As counsel for the Defendants confirmed, the asymmetry is in how it has been photographed, not how it has been cut.
222. C13 and D13 are very similar indeed, as is obvious from the images set out above. The differences pointed out by the Defendants’ counsel are insufficient, individually or collectively, to render D13 other than an article made substantially to the design of C13. Similarly, the differences would not strike the informed user so as to produce a different overall impression. My detailed side by side comparison of the garments confirms this position.
223. D13 infringes UKUDR and CUDR in C13.
224. C17 was designed by Georgina Douek on or about 26 April 2016 and first offered for sale on the House of CB website on 10 October 2016. On 1 April 2017, Ms Henderson provided email instructions for D17, including an image of a Murmur dress. An image of C17 was added to the Trello record for D17 on 24 April 2017. D17 was first offered for sale on the Oh Polly website on 25 May 2017.
C17 D17
225. The significant features of C17 were said to be:
“This is a sleeveless below the knee dress with a [the Defendants said wired] bustier effect top with thin straps which are connected to the body of the dress with a metal loop. Both sides of the dress are cut away from top to bottom and the front and back panels of the dress are connected with horizontal narrow strips of fabric. The first strip of fabric is connected to the top of the front panel at the side of the bra top and continues across the back joining at the centre of the back with a metal adjuster. The other strips of fabric joining the front and back panels are connected by a metal loop at the front and the back and have a metal adjuster in the centre. The main front panel of the dress is made of one piece of fabric which curves in at the waist. The back of the dress is made of two sections of fabric which also curve in at the waist with a centre back seam and an invisible zip down that seam. This garment is sold in blush stretchy lightweight crepe.”
226. The features of colours and materials relate to CUDR only.
227. In relation to originality in the copyright sense (UKUDR), the Defendants relied on a Vogue image of a Cushnie et Ochs dress. For the closest prior design (CUDR), the Defendants relied on a Blac Chyna dress published in The Irish Mirror in April 2016. Both dresses are shown here:
Cushnie et Ochs Blac Chyna
228. In addition, the Defendants relied on a further dress to suggest that C17 was commonplace.
229. Having reviewed the images, in my judgment, C17 was not slavishly copied from the Cushnie et Ochs dress, and does not create the same overall impression on the informed user as the Blac Chyna dress. The Cushnie et Ochs dress is a racer neck midi dress with a loose shift shape. It is open on the sides from the underarm to the hem - the open sides are secured by a row of buttons on each side to each of which is attached an elastic band. The buttons extend from under the underarms down to the top of each thigh. It is yellow. The Blac Chyna dress is made of bandage fabric. It is floor length. It has unusual shoulder straps that radiate from the centre of the bust, and a round cut out shape to expose the centre bust. The sides of the dress are not cut out. It is red.
230. I have also reviewed the other dress which the Defendants consider not to be as close to C17. It does not make C17 commonplace - it is a different design.
231. C17 is not invalid in relation to UKUDR or CUDR.
232. The Defendants’ case is that D17 is based on a Murmur mesh dress, shown here:
233. An image of the Murmur dress was sent by Ms Henderson as the basis for D17. The next day, she sent a second email headed “CHANGE OF DESIGN”, adding “no mesh please, so just the cut out sides all the way down (showing skin in between) and absolutely no mesh”. The Trello record was created on 5 April 2017. It was not until 24 April 2017 that C17 was introduced, with Dr Branney adding a jpeg image to the Trello record, shown here:
234. By this time, the fit samples had already been developed in the factory. The design was finalised shortly thereafter, on 29 April 2017.
235. Not without hesitation, I have come to the conclusion that D17 was not copied from C17. The buckle arrangement on the side of C17 was copied, but by then D17 had largely been designed, based on the Murmur dress without the mesh. Dresses with side cut outs had been available for some time: Liz Hurley wore such a dress to accompany Hugh Grant to the opening of Four Weddings and a Funeral in 1994. Had the Claimants pleaded only the cut out part of C17, the result may have been different. But I do not consider that it can be said that the Defendants have copied C17 as a whole.
236. There are a number of differences between C17 and D17: (i) D17 has a waistband whereas C17 does not; (ii) D17 has a dart at the bust whereas C17’s bust is achieved through wiring; and the cut out side of D17 does not continue to the top of the dress under the arm on each side. I am conscious that the images of C17 and D17 look very similar, but the differences are more noticeable in the garments themselves.
237. In my judgment, these differences are sufficient collectively to render D17 not an article made substantially to the design of C17. Similarly, the differences would strike the informed user so as to produce a different overall impression.
238. D17 does not infringe UKUDR or CUDR in C17.
239. C17 was designed by Julia Kasper on or about 4 August 2014 and first offered for sale on the House of CB website in black on 3 November 2014, in white on 13 January 2015, in wine on 20 November 2017, in tan on 6 August 2018 and in gold on 29 October 2018. The wine, tan and gold versions were designed jointly by Georgina Douek and Justin Ruddle. On 21 April 2016, Mr Ansar Ullah at Oh Polly’s factory in Bangladesh attached images of a Meshki top to the Trello record - possibly Ms Henderson gave them to him on the basis that she was in Bangladesh at the time. Further references were added to an earlier Oh Polly style 283 and an ASOS dress. D21 was offered for sale on 20 June 2016 in blue, pink and black.
C21 D21
240. The significant features of C21 were said to be:
“This is a strapless close fitting dress with sweetheart shaped neck at the top of the front of the dress [although the Defendants said it is actually curved, not sweetheart] and a deep strip of fabric running around the top [although the Defendants said this is actually a fold over of fabric with a vertical bust seam], a fluted hem made of horizontal strips of fabric and a zip at the centre back seam with two deep strips of fabric running down the centre back. This design is sold in white, black, wine, tan and gold. It is made of stretchy bandage fabric.”
241. The features of colours and materials relate to CUDR only.
242. In relation to originality in the copyright sense (UKUDR), the Defendants relied on a Hervé Léger dress taken from a 2011 Vogue magazine. For the closest prior design (CUDR), the Defendants relied on a Christian Siriano dress shown in a 2013 edition of Vogue magazine. Both are shown here:
Hervé Léger Christian Siriano
243. In addition, the Defendants relied on a further four dresses to suggest that C21 was commonplace.
244. Having reviewed the images, in my judgment, C21 was not slavishly copied from the Hervé Legér dress, and does not create the same overall impression on the informed user as the Christian Siriano dress. The Hervé Legér dress has shoulder straps. It has a V-shaped neckline and open work to the front waist area. It has a skirt with a pleated and fluted hem that ends at about the knee. It is made from a metallic bandage fabric with jacquard pattern. It has beaded trims running the length of the straps, neckline and down the panelling of the dress. The Christian Siriano dress is a red strapless dress with front panels that narrow at the waist. It has a skirt with a wide fluted hem made from a single piece of fabric. The dress is fitted with a narrow belt at the waist.
245. I have also reviewed the other dresses which the Defendants consider not to be as close to C21. They do not make C21 commonplace - they are different designs.
246. C21 is not invalid in relation to UKUDR or CUDR.
247. The Defendants’ case is that D21 was based on a Meshki top, as shown on the Trello record, and shown here:
248. An earlier Oh Polly design was then referred to. The first sample was made with sleeves, as shown in the Meshki top, but these were later dropped. An ASOS dress was later referred to. There is no basis for inferring that C21 was referred to.
249. In my judgment, D21 was not copied from C21. As the Defendants point out, closefitting strapless dresses with a fishtail or fluted hem were popular at the time. C21 and D21 are not similar enough to create a presumption of copying that needs answering, and the Trello records provide an explanation for the genesis of D21.
250. As I have said, there are a number of differences between C21 and D21: (i) C21 has a curved neckline - D21 has a sweetheart neckline; (ii) D21 does not have a waistband; (iii) the fluted hems are quite different; and (iv) D21 has piping running the full length of the garment.
251. In my judgment, these differences are sufficient collectively to render D21 not an article made substantially to the design of C21. Similarly, the differences would strike the informed user so as to produce a different overall impression.
252. D21 does not infringe UKUDR or CUDR in C21.
253. C35 was designed by Rheanna Donaldson on or about 26 August 2016. The garment was first offered for sale on the House of CB website on 10 April 2017. Four months later, on 17 August 2017, Ms Henderson sent an email with design instructions, attaching an image of C35. D35 was offered for sale on the Oh Polly website on 26 October 2017.
C35 D35
254. The significant features of C35 were said to be:
“This is a short long sleeved dress. It has a round neck and the front is made of 3 panels of fabric with a wider panel running down the centre and narrow panels at the sides. It has two V-shaped cut outs either side of the neck which finish at a point just above each breast. The sides of the cut outs are punctured with four metal eyelets and a narrow strip of fabric is laced through the eyelets. There is deep waist band across the side front panels and the back of the dress. It has a high neck at the back and a zip running down the entire length of the centre back. This garment is sold in baby pink stretch crepe.”
255. The features of colours and materials relate to CUDR only.
256. In relation to originality in the copyright sense (UKUDR), the Defendants relied on a dress worn by Jennifer Lopez as shown on a blog in September 2015. In relation to the closest prior design (CUDR) the Defendants relied only on the Tigermist dress from prior to May 2016. Both are shown below:
Jennifer Lopez Tigermist
257. In addition, the Defendants relied on a further garment to suggest that C35 was commonplace.
258. Having reviewed the images, in my judgment, C35 was not slavishly copied from the Jennifer Lopez dress, and does not create the same overall impression on the informed user as the Tigermist dress. The Jennifer Lopez garment has diagonal cut outs on either side running from underneath the underarms up to the yoke. The cut outs are secured with two buttons. The dress has two tone coloured panelling - with the sides and shoulders in a different colour to the front and sleeves of the dress (this is difficult to see in the image above). The Tigermist dress is a round neck long-sleeved mini dress. It has a high round neckline with cut out panels to each side of the neck that curve outwards in a half moon shape and extend to the underarm. The cut-out panels are of a uniform width for the length of the cut out. The dress is also cut right around the waist, forming a top and skirt secured together by lacing details all around the waist.
259. I have also reviewed the other garment which the Defendants consider not to be as close to C35. It does not make C35 commonplace - it is a different design.
260. C35 is not invalid in relation to UKUDR or CUDR.
261. The Defendants’ case is that Ms Henderson copied C35, but that copying was not “deliberate”. It is difficult to square that comment with reality, and I reject it: Ms Henderson emailed an image of C35 to Dr Branney and Mr Henderson with the instructions “Mini dress as attached”. Further, the Defendants submit that the image uploaded was only of the front of the dress. Hence, it was submitted that key features at the back could not have been copied. I do not consider that that matters - here, all the key features of the dress are at the front - the back is standard.
262. No further prominent differences between C35 and D35 were enumerated.
263. The two garments are very similar indeed, as is obvious from the images set out above. I have also compared the dresses side by side. D35 is an article made substantially to the design of C35. Similarly, D35 produces on the informed user the same overall impression as C35.
264. D35 infringes UKUDR and CUDR in C35.
265. C47 was designed by Julia Kasper on or about 6 May 2015 and first offered for sale on the House of CB website on 22 February 2016. On 25 December 2016, Ms Henderson sent an email with design instructions for D47 to Dr Branney, who, the same day, uploaded to Trello a Microsoft Paint-annotated image of a Gina Gado dress. D47 was first offered for sale on the Oh Polly website (in grey and Nude) on 14 February 2017.
C47 D47
266. The significant features of C47 were said to be:
“This is a draped wrap dress. This dress has a draped wrap design. The front consists of two individual pieces of fabric that extend from the shoulder to the deep waistband to cover each part of the bust. Running down the centre of the bust are two strings that extend up from the centre of the waistband between the breasts and fasten round the neck. The skirt has a crossover slit that extends down from the base of the waistband on one side and opens to show the thigh. The back has an invisible zip. This item is sold in Grey silky jersey. [The Defendants said that the Claimants have not claimed as a significant feature the ruched waistband that runs around the garment].”
267. The features of colours and materials relate to CUDR only.
268. In relation to originality in the copyright sense (UKUDR) and for the closest prior design (CUDR), the Defendants relied on a Mamas Boutique dress published on the fashion house’s Instagram account in 2015, shown here:
269. In addition, the Defendants relied on a further four dresses to suggest that C47 was commonplace.
270. Having reviewed the images, I cannot conclude that C47 was slavishly copied from the Mamas Boutique dress or that it produces the same overall impression on the informed user. Like C47, the Mamas Boutique dress has a plunging neckline with a centre strap detail that joins to a V-shape at the centre waistband. The draped fabric that covers the bust appears similar. The Mamas Boutique skirt crosses over at the right hip and is split at the right thigh, whereas C47 is the reverse - but this could have been reversed in the photography of the Mamas Boutique dress. However, the greatest difficulty I have is that very little of the Mamas Boutique dress can be seen in the image I have been given, such that it is not possible to know what is going on at the back of the dress, or even what length the dress is (it appears to be longer).
271. I have also reviewed the other dresses which the Defendants consider not to be as close to C47. They do not make C47 commonplace - they are different designs.
272. C47 is not invalid in relation to UKUDR or CUDR.
273. The Defendants’ case is that D47 is based on earlier Oh Polly designs called Flashback and Drape Expectations. However, the written record shows only the design instructions sent to Dr Branney on Christmas Day 2016, including the Gina Gado dress uploaded to Trello. There is no basis for inferring that C21 was referred to. The Gina Gado dress is shown here:
274. In my judgment, D47 was not copied from C47. As the Defendants point out, draped wrap-around dresses with a slit at the thigh were popular at the time. C47 and D47 are not similar enough to create a presumption of copying that needs answering, and the Trello records provide an explanation for the genesis of D47. It seems much more likely to me that D47 was copied from the Gina Gado dress.
275. As I have said, there are a number of differences between C47 and D47: (i) in D47, the bust panels overlap at the front to form a V-shape; (ii) in D47, the neck strap is braided and forms a loop to pass over the head, rather than ties to be tied behind the neck as in C47; (iii) the dresses are different lengths; and (iv) D47 has a tie waist belt which ties at the side.
276. In my judgment, these differences are sufficient collectively to render D47 not an article made substantially to the design of C47. Similarly, the differences would strike the informed user so as to produce a different overall impression.
277. D47 does not infringe UKUDR or CUDR in C47.
278. C49 was designed by Justin Ruddle on or about 26 April 2016 and first offered for sale on the House of CB website on 14 November 2016. On 10 January 2017, Dr Branney uploaded to Trello Ms Henderson’s Microsoft Paint images of a Missguided/PLT garment, with annotations. A colour reference image was also attached. D49 was first offered for sale on the Oh Polly website (in teal and bronze) on 20 February 2017.
C49 D49
279. The significant features of C49 were said to be:
“This is a short backless dress with a halter neck top. The dress has a plunging neckline to the waist. The top left hand front panel of the dress is a strip of fabric which goes up to a point just below the shoulder which is connected to 2 narrow tapes which join together at shoulder level. The top right hand panel of the front of the dress also goes up to a point finishing just above the bust which is connected to a single tape. The tapes connected to the top of the front panels are tied at the back of the neck and hang down to just above waist level. The front panels finish at the waist but give the appearance of crossing at the waist to form a large knot continuing down the sides of the front of the skirt front at hip level. There is a zip in the centre back seam of the skirt. The skirt has no visible waistband at the back and a deep dart on each side from the waist to create a tight fit. This garment is sold in deep green satin.”
280. The features of colours and materials relate to CUDR only.
281. In relation to originality in the copyright sense (UKUDR) the Defendants relied on an Alexander Vauthier dress from his Spring 2014 collection, although the Claimants said that that claim had already been abandoned by the Defendants, in which case, the invalidity submission fails. However, in case I am wrong in that, I will consider the submission as if it had not been abandoned. For the closest prior design (CUDR) the Defendants relied on a Mauri and Eve dress from 2016. Both are shown here. It was not clear to me which of the seven garments shown in the right hand two photographs was the Mauri and Eve dress being referred to - but it does not matter:
Alexander Vauthier Mauri and Eve
282. In addition, the Defendants relied on a further two dresses to suggest that C49 was commonplace.
283. Having reviewed the images, I do not consider that C49 was copied from the Alexander Vauthier dress or that it produces the same overall impression on the informed user as any of the Mauri and Eve dresses shown. The Alexander Vauthier dress is a loose draped handkerchief mini dress, with a draped waterfall effect to the skirt that sits asymmetrically at the hemline. The waistband is loose and angles diagonally down from the right hip. The right shoulder cape has a waterfall effect. None of the Mauri and Eve dresses shown in the exhibit is anywhere close to C49.
284. I have also reviewed the other dresses which the Defendants consider not to be as close to C49. They do not make C49 commonplace - they are different designs.
285. C49 is not invalid in relation to UKUDR or CUDR.
286. The Defendants’ case is that D49 is based on a Missguided/PLT garment, an image of which was uploaded to Trello. There is no basis for inferring that C49 was referred to.
287. The Missguided/PLT garment is shown here:
288. In my judgment, D49 was not copied from C49. As the Defendants point out, short dresses with plunging necklines were popular at the relevant date. In that context, C49 and D49 are not similar enough to create a presumption of copying that needs answering, and the Trello records provide an explanation for the genesis of D49: it was likely copied from the Missguided/PLT garment.
289. As I have said, there are a number of differences between C49 and D49: (i) C49 has asymmetric panels covering the bust, with tapes tied at the back of the neck whereas D49 is closer to a standard halter top shape, with no asymmetry; (ii) the knotting of the fabric at the front of the dress is different, with D49 having cut out panels which expose the skin of the waist - this cut out continues around to the rear of the dress; and (iii) there is a seam at the front of the skirt.
290. In my judgment, these differences are sufficient collectively to render D49 not an article made substantially to the design of C49. Similarly, the differences would strike the informed user so as to produce a different overall impression.
291. D49 does not infringe UKUDR or CUDR in C49.
292. C61 was designed by Georgina Douek on or about 30 May 2016. She had recently returned from Berlin where she had observed use of latex in clothing. The garment was first offered for sale on the House of CB website on 12 June 2017. Two months later, on 9 August 2017, Ms Henderson sent an email with design instructions, attaching an image of C61. D61 was first offered for sale on the Oh Polly website on 18 October 2017 in brown and white.
C61 D61
293. The significant features of C61 were said to be:
“This is a tight fitting, above the knee, strapless dress. It has a straight neckline. The front of the top of the dress is made of 3 panels with a wider central panel and two narrow panels at the sides and the front of the skirt is also made of 3 panels with a matching wider panel in the middle which curves out at the hips and in again towards the hem. There is no waistband. Strips of fabric run down the seams down the front of the dress and these are punctured with metal eyelets and a thin strip of fabric is laced through the eyelets. The top of the back of the dress is made of 2 panels of fabric. There is a zip running down the entire centre back of the dress. The back of the skirt is made of four panels of fabric and strips of fabric run down the seams as in the front which are also punctured with metal eyelets and a narrow strip of fabric is laced through the eyelets. This garment is sold in black latex.”
294. The features of colours and materials relate to CUDR only.
295. All the prior designs relied on by the Defendants post-date the creation date for D61. The invalidity arguments for this design therefore fail.
296. C61 is not invalid in relation to UKUDR or CUDR.
297. The Defendants’ case is that lace-up garments were already trending and selling well for Oh Polly. Ms Henderson said that she had already designed Style 1026, an earlier example of a lace-up design with a strapless neckline and dual lace up feature on the front panel and rear full body zip. As set out previously, I am unable to accept Ms Henderson’s evidence that she based the design of D61 on Style 1026 without corroborative evidence, of which there was none. On 9 August 2017, Ms Henderson emailed Dr Branney and Mr Henderson an image of C61 (which she had labelled “sample3”) for use by the factory. I reject her evidence that she had already designed D61 in her head. I also reject the submission that her reference to another Oh Polly dress for the length of the hem meant that she did not copy C61. Similarly, the reference to another Oh Polly dress when the sample was being amended does not mean that C61 was not copied. I also reject the argument that the back of the dress could not have been copied because the uploaded image of C61 only shows the front of the dress. I have no hesitation in concluding that D61 was copied from C61.
298. The further differences between C61 and D61 are said to be as follows:
i) D61 has an empire waist line, whereas C61 has a standard waistline - for my part, I could not discern this apparent difference in either the images or on the samples I was shown.
ii) D61 has lace running up the length of the body on the back of the dress whereas the lacing stops at the waist in C61.
iii) D61 has nylon shoestring lace threaded through the eyelets (relevant to CUDR only).
iv) The lace in D61 is in contrasting colour to that of the dress fabric (relevant to CUDR only).
v) D61 is made of unstretchy latex, whereas C61 is made of very stretchy suedette.
299. Having inspected both garments, I reject those submissions. The two garments are very similar indeed, as is obvious from the images above. The differences pointed out by the Defendants’ counsel are insufficient, individually or collectively, to render D61 other than an article made substantially to the design of C61. Similarly, the differences would not strike the informed user so as to produce a different overall impression. I accept that the brown version of D61 infringes CUDR in C61, but the white version would create a different overall impression on the informed user, and so does not infringe CUDR in C61.
300. D61 infringes UKUDR in C61. The brown version of D61 infringes CUDR in C61 but the white version does not.
301. C63 was designed by Rheanna Donaldson on or about 9 August 2016 and first offered for sale on the House of CB website on 30 January 2017. Three months later, on 22 April 2017, Ms Henderson sent instructions by email to Dr Branney and Mr Henderson for D63, referring to an earlier Oh Polly design (Deep and Bounds) along with an image of a suede dress with a lace-up waist feature. D63 was first offered for sale on the Oh Polly website on 14 June 2017.
C63 D63
302. The significant features of D63 were said to be:
“This is a sleeveless dress with a scooped neck at the front and a deep scooped neck at the back. There is a slit down the centre front of the dress at the bust which stops just above the waistline and a slit at the centre front of the skirt. There is no waist band at the bottom of the centre front panel of the top of the dress but a waist band around the remainder of the dress. There is a zip running down the centre back seam [although the Defendants submitted that it is a hidden metal zip]. This garment is sold in red stretchy crepe.”
303. The features of colours and materials relate to CUDR only.
304. By the end of the trial, the Defendants no longer advanced a case of lack of originality in the copyright sense (UKUDR). The Defendants relied on an Emprada dress published on Instagram on 2 August 2016 as the closest prior design (CUDR), shown here:
305. No commonplace argument was made - rather, the Defendants submitted that, if valid, any rights in C63 were extremely narrow.
306. Having reviewed the image of the Emprada dress, I do not consider that C63 produces the same overall impression on the informed user as that design. Whilst both C63 and the Emprada dress have a split to the centre bust, there are multiple differences. The Emprada dress has a square neckline that is cut straight across the bust. The bust is not visible through the centre bust incision. The back is cut straight across.
307. C63 is not invalid in relation to UKUDR or CUDR.
308. The Defendants’ case is that D63 is based on an early Oh Polly garment, the Deep and Bounds style. This was sent by Ms Henderson on 22 April 2017 to Dr Branney and Mr Henderson - Mr Henderson uploaded the image to Trello. There is no basis for inferring that C63 was referred to.
309. In my judgment, D63 was not copied from C63. As the Defendants point out, short dresses with a V-shaped cut out at the centre bust were popular at the time. In that context, C63 and D63 are not similar enough to create a presumption of copying that needs answering, and the Trello records provide an explanation for the development of D63: it was copied from the earlier Oh Polly garment called Deep and Bounds.
310. As I have said, there are a number of differences between C63 and D63: (i) D63 has a square neck and spaghetti straps; (ii) the back is also square, rather than scooped; and (iii) there is no slit in the centre of the skirt. They are different dresses.
311. In my judgment, these differences are sufficient collectively to render D63 not an article made substantially to the design of C63. Similarly, the differences would strike the informed user so as to produce a different overall impression.
312. D63 does not infringe UKUDR or CUDR in C63.
313. C66 was designed by Rheanna Donaldson on or about 26 May 2015 and first offered for sale on the House of CB website on 23 November 2015. On 17 January 2017, Ms Henderson sent instructions by email to Dr Branney for D66, attaching annotated images of non-House of CB garments. Dr Branney uploaded the annotated images to Trello. D66 was first offered for sale on the Oh Polly website on 16 March 2017.
C66 D66
314. The significant features of C66 were said to be:
“This is a sleeveless, backless jumpsuit with wide leg flared trousers. The front of the top has a wrap across, halter neck effect leaving a triangle shape cut out above the waistband. The back of the jump suit has a scooped back with a zip at the centre back seam. This garment is sold in peach stretchy sequinned fabric.”
315. The features of colours and materials relate to CUDR only.
316. By the end of the trial, the Defendants no longer advanced a case of lack of originality in the copyright sense (UKUDR). The Defendants relied on a Rare London garment published on Facebook on 16 May 2014 as the closest prior design (CUDR), shown here:
317. The Defendants relied on a further 3 garments to show that C66 was commonplace.
318. Having reviewed the image of the Rare London garment, I do not consider that it produces the same overall impression on the informed user as C66. Both C66 and the Rare London garment are white jumpsuits with a cross-over halter neckline that exposes the lower bust. But the waist of the Rare London garment sits higher, and the legs are straight, with a raised seam that runs the length of the legs, and, importantly, it is sequinned. These differences will strike the informed user so as to produce a different overall impression.
319. There is nothing in the evidence to suggest that the design of C66 was commonplace at the relevant dates.
320. C66 is not invalid in relation to CUDR or UKUDR.
321. The Defendants’ case is that D66 is based on the instructions and images sent to Dr Branney by Ms Henderson on 17 January 2017. Neither of those images is C66. There is a written record of both Ms Henderson’s email, and the upload to Trello. The Trello record also shows that the design was amended during the fitting stage, from two straps around the waist to one. There is no basis for inferring that C66 was referred to.
322. In my judgment, D66 was not copied from C66. Both wide-legged jumpsuits and the cross-wrap feature were popular at the time. The Defendants have put forward written evidence of their design history. C66 and D66 are not similar enough to create a presumption of copying that needs answering, and the Trello records provide an explanation for the development of D66.
323. There are a number of differences between C66 and D66, the most obvious of which is that the top of D66 is put together by the wearer by wrapping a long piece of fabric, and tying it at the back, whereas C66 is sewn in position: the appearance of the top section therefore differs at both the front and back. The differences were even more obvious when comparing the actual garments.
324. In my judgment, these differences are sufficient collectively to render D66 not an article made substantially to the design of C66. Similarly, the differences would strike the informed user so as to produce a different overall impression.
325. D66 does not infringe UKUDR or CUDR in C66.
326. C77 was designed by Rheanna Donaldson on or about 5 May 2015 and first offered for sale on the House of CB website in white, grey, khaki, black and Nude on 9 May 2016, in tan and taupe on 3 September 2017 and in burgundy on 4 September 2017. On 19 January 2017, Ms Henderson sent instructions by email to Dr Branney for D77, attaching an image from Oh My Gosh clothing. Dr Branney forwarded Ms Henderson’s email to the factory in Bangladesh. The image was also uploaded to Trello. D77 was first offered for sale on the Oh Polly website in stone, cream, chocolate and new blue on 28 March 2017.
C77 D77
327. The significant features of C77 were said to be:
“This item is a vest shaped bodysuit. It has a sleeveless cut and shoulder straps. It is closefitting with a body hugging fit. It has a scoop neckline, a lower scoop back and a thong shape to the briefs. It has a pull on design without any fastenings. This item is sold in Black, White, Tan, Taupe, Grey, Khaki, Nude and Burgundy smooth knit stretchy fabric.”
328. The features of colours and materials relate to CUDR only.
329. By the end of the trial, the Defendants no longer advanced a case of lack of originality in the copyright sense (UKUDR). The Defendants relied on an image from The Daily Mail Online dated 16 April 2016 as the closest prior design (CUDR), shown here:
330. The Defendants relied on a further 2 garments to show that C77 was commonplace, arguing that “it is a standard tight-fitting bodysuit, equivalent to a T-shirt for a man”.
331. The Daily Mail image was published on 16 April 2016, that is, after C77 was designed, and shortly before it was made available to the public. It is therefore relevant for CUDR and not UKUDR. The difficulty with the image is that it does not show the garment below the waist - so it is not possible to tell if it is a thong or not. The fact that C77 is a thong will strike the informed user. Without demonstrating that that feature appears in the Daily Mail image, it is not possible for that image to produce on the informed user the same overall impression as C77. I therefore reject the Defendants’ case for invalidity of the CUDR.
332. Similarly, the prior designs adduced to demonstrate that the design of C77 was commonplace also fail to show the lower third of the garment in every case but one, and that image does not show the rear of the garment. The Defendants’ case for invalidity of C77 on the ground that it is commonplace also fails - it is not sufficient to say that there is nothing original about the garment: that must be proved.
333. In my judgment, C77 is not invalid in relation to CUDR and UKUDR.
334. The Defendants’ case is that D77 is based on the instructions and an image of an Oh My Gosh garment sent to Dr Branney by Ms Henderson on 19 January 2017. The Oh My Gosh garment is shown here.
335. That email was forwarded to the factory in Bangladesh. There is a written record of both Ms Henderson’s email, and the upload to Trello. There is no basis for inferring that C77 was referred to.
336. In my judgment, D77 was not copied from C77. Bodysuits were popular at the time. The Defendants have put forward written evidence of their design history. Whilst C77 and D77 appear to be very similar, the top of the bodysuit is a standard product, as demonstrated by the prior designs before the Court. The Trello records provide an explanation for the development of D77.
337. Had I found that D77 had been copied from C77, I would also have found infringement of UKUDR and CUDR. D77 is very similar to C77, such that it is an article made substantially to the design. It also produces on the informed user the same overall impression.
338. D77 does not infringe UKUDR or CUDR in C77.
339. C81 was designed by Rheanna Donaldson on or about 5 September 2014 and first offered for sale on the House of CB website in white and Nude on 28 March 2015 and in black on 29 March 2015. On 30 January 2017, Ms Henderson sent instructions by email to Dr Branney for D81, attaching images of three dresses (not including C81). A Trello record was created, and Mr Henderson uploaded the three images to Trello. D81 was first offered for sale on the Oh Polly website in black, silver and white on 11 March 2017.
C81 D81
340. The significant features of C77 were said to be:
“This dress has a close fitting smooth body hugging fit. The bust line is cut across the bust with a slight curve up towards the centre bust. It has very thin narrow shoulder straps [although the Defendants submitted that they are connected to the body of the slip with buckles on both sides, so as to be adjustable] and the hemline sits mid thigh. It has a pull on design with no zip. This dress is sold in Nude, Black and White silky jersey elastane.”
341. The features of colours and materials relate to CUDR only.
342. In relation to lack of originality in the copyright sense (UKUDR), the Defendants relied on an Hervé Léger dress. The Defendants submitted that the closest prior design (CUDR) was a dress worn by Jennifer Aniston shown in Getty Images images from July 1997. Both are shown here:
Hervé Léger Jennifer Aniston
343. The Defendants relied on a further 4 garments to show that C81 was commonplace, describing it as “an utterly generic design”.
344. Having reviewed the images provided, in my judgment, C81 was not slavishly copied from the Hervé Léger dress, and C81 produces on the informed user a different overall impression from the Jennifer Aniston dress. The Hervé Léger dress is a bandage mini dress. It is constructed by stitching small horizontal strips of multi coloured bandage fabric together. The neckline is tight on the mid bust, with the shoulder straps spaced on the outer edge of the shoulders. The Jennifer Aniston dress is a slip dress with a figure skimming fit and spaghetti straps. It is not bodycon.
345. Similarly, the prior designs adduced to demonstrate that the design of C81 was commonplace do no such thing - they are all very different dresses. I note for completeness that C81 is an undergarment (it is described as “Elastine Underdress Body Slip”) whereas all the prior designs and D81 are not. The Defendants’ case for invalidity of C81 on the ground that it is commonplace also fails. C81 may well be “utterly generic”, but no evidence has been adduced which proves that.
346. In my judgment, C81 is not invalid in relation to CUDR and UKUDR.
347. The Defendants’ case is that Ms Henderson emailed her instructions for D81 to Dr Branney on 30 January 2017, attaching three images which she had found. Those images are shown here:
348. That email and the Trello record were before the Court. There is no basis for inferring that C81 was referred to.
349. In my judgment, D81 was not copied from C81. C81 is an underdress - D81 is a dress which includes a choker element around the neck. The Defendants have put forward written evidence of their design history. Whilst C81 and D81 appear in the photographs to look similar, there are differences - D81 has a choker; D81 does not have metal buckles on the front of the strap; D81 does not have bust darts; D81 has a zip at the back and D81 is made from heavy-knit lurex fabric. The differences are sufficient not to create a presumption of copying that needs answering, and the Trello records provide an explanation for the development of D81 in any event.
350. These differences are also enough to render D81 not an article made substantially to the design of C81. Similarly, these differences would strike the informed user so as to produce a different overall impression.
351. D81 does not infringe UKUDR or CUDR in C81.
352. C91 was designed by Justin Ruddle on or about 5 May 2017. The garment was first offered for sale on the House of CB website on 20 November 2017. On 5 March 2018, images (including C91) found by Ms Henderson were uploaded to Trello. D91 was finalised on 9 May 2018 and made available in sage, hot pink, cream and black.
C91 D91
353. The significant features of C91 were said to be:
“This item is a jumpsuit with a deep plunge neckline. It has an extreme wide leg design. The bust has a deep plunge that meets the waistband. The waistband is wider at the centre and tapers more narrowly towards the sides [although the Defendants rejected this]. The legs have an extreme wide leg cut. The bust has seaming either side that starts at the waistband rising to mid bust. The trousers have a pleat at the front centre of each leg extending from the waistband down each thigh. The back has a backless design that is open to the back waistband. The straps extend over the shoulders down to the waistband. This item is available in Black. [The Defendants also pointed to a hidden zip.]”
354. The colour feature relates to CUDR only.
355. In relation to originality in the copyright sense (UKUDR), the Defendants relied on a green Cushnie et Ochs jumpsuit from Fall 2017. In relation to the closest prior design (CUDR) the Defendants relied on a design worn by Jessie J as shown in The Daily Mail on 16 January 2015. Both are shown here:
Cushnie et Ochs Jessie J
356. In addition, the Defendants relied on a further four garments to suggest that C91 was commonplace.
357. Having reviewed the images, in my judgment, C91 was not slavishly copied from the green Cushnie et Ochs garment, and does not create the same overall impression on the informed user as the Jessie J garment. The green Cushnie et Ochs garment is a jumpsuit with a plunging neckline and a wide V-shape bust. The bust is fully supported due to the under-bust band extending from immediately below the bustline to the waist. It has a diagonal skinny strap that runs from the wearer’s right shoulder down to the centre left bust. The sides are fully covered. It has a full covered back up to the neckline and a zip that runs up to the neckline. The legs are straight. The Jessie J garment is a halter neck jumpsuit. It has two bust panels that end at a sharp V at the centre front waist. The bust panels narrow towards the neckline and fasten around the back of the neck. The back is completely open with no visible straps. It has a tight bodycon fit especially around the hip and buttock area. The legs are tight fitting through the thighs and they flare slightly at the hem.
358. I have also reviewed the other garments which the Defendants consider not to be as close to C91. They do not render C91 commonplace - they are different designs.
359. C91 is not invalid in relation to UKUDR or CUDR.
360. The Defendants’ case is that Oh Polly had already launched its Just Say Flow jumpsuit, which had been inspired by a Na-kd garment. Just Say Flow sold well. The main change to create D91 was to make the trouser leg wider, there being a trend for wide-legged trousers identified by Ms Henderson. Ms Henderson said that, having identified the trend, she went on-line to try to find an image of the sort of wide-legged feature she had in mind - she found an image of C91, but did not like the colour. She found two other references for colour, and passed all three images on to Mr Henderson to upload to Trello. As set out previously, I am unable to accept Ms Henderson’s evidence unless it is supported by other evidence. Here it is not. The documentary record shows Ms Henderson sending C91 to Mr Henderson. As previously stated, I do not accept the Defendants’ argument that Ms Henderson designed D91 in her head, and then went looking for images to demonstrate what she was seeking to convey.
361. I have no hesitation in concluding that D91 was copied from C91.
362. The further differences between C91 and D91 are said to be as follows:
i) D91 has darts in the lower legs which only run as far as the hips;
ii) C91 had the hem finished blind stitch whereas D91 does not; and
iii) D91 is made of a light polyester, whereas C91 is made from a heavy bandage fabric.
363. Having inspected both garments, I reject those submissions. The two garments are very similar indeed, as is obvious from the images set out above. The differences pointed out by the Defendants’ counsel are insufficient, individually or collectively, to render D91 other than an article made substantially to the design of C91. However, the case under UCDR differs slightly: C91 is black, and the Claimants claim the colour as a feature of the design. They could have chosen not to. As black is claimed as a key feature of the design, I conclude that the black version of D91 infringes, but the cream, sage and hot pink versions do not: they produce a different overall impression on the informed user.
364. D91 infringes UKUDR in C91 and the black version of D91 infringes CUDR in C91. The cream, sage and hot pink versions of D91 do not infringe CUDR in C91.
365. C93 was designed by Georgina Douek on or about 25 August 2017 and first offered for sale on the House of CB website on 26 March 2018. On 10 April 2018, Ms Johnston uploaded to Trello images sent to her by Ms Henderson. Of the two garments, one is an earlier Oh Polly design, and the other has not been identified (but it is in any event, not House of CB). Ms Johnston created some sketches which were uploaded to Trello on 12 April 2018. Further sketches by Ms Johnston were uploaded on 17 April 2018 - she also suggested some changes, and uploaded images of further fit samples. D93 was first offered for sale on the Oh Polly website on 20 August 2018.
C93 D93
366. The significant features of C93 were said to be:
“This item is a jumpsuit. It has an extreme plunge neckline. The shoulder straps are wide bands that extend over the shoulders, down the front of each breast and then join to the waistband. The bust is somewhat exposed to the centre front and side. It has a band that runs horizontally under the bust line and around the body. The jumpsuit is tight fitting to the body. The plunge neckline extends down to meet the waistline. The waistline is defined. The legs are tight fitting and tapered to the ankle. This item is available in Rust. [The Defendants submitted that D93 has a closed back with a hidden zip].”
367. The colour feature relates to CUDR only.
368. In relation to lack of originality in the copyright sense (UKUDR) and as the closest prior design (CUDR), the Defendants relied on a Cushnie et Ochs garment from its Pre-Fall 2015 Ready to Wear collection, shown here:
369. The Defendants relied on a further 3 garments to show that C93 was commonplace.
370. Having reviewed the image provided to the Court, in my judgment, C93 was not slavishly copied from the Cushnie et Ochs jumpsuit, and C93 produces on the informed user a different overall impression from the Cushnie et Ochs jumpsuit. The Cushnie et Ochs jumpsuit is a jumpsuit with a plunge neckline. The bust panels end at a sharp V-shape at the centre front waist. It has a very deep waistband. It has a deep bandeau that sits horizontally around the bust area. The legs are straight, and fit loosely. The jumpsuit is purple.
371. Similarly, the prior designs adduced to demonstrate that the design of C93 was commonplace do no such thing - they are all very different jumpsuits.
372. In my judgment, C93 is not invalid in relation to CUDR and UKUDR.
373. The Defendants’ case is that Ms Johnston’s evidence on copying was not challenged. Rather, Ms Johnston designed D93 having looked at images that Ms Henderson had provided, none of which was a House of CB garment. The images and the Trello record were before the Court. There is no basis for inferring that C93 was referred to.
374. In my judgment, D93 was not copied from C93. This is one of the few examples before the Court where Ms Henderson’s involvement in the design was limited. I have accepted Ms Johnston’s evidence, including the evidence as to how she designed D93. The differences between C93 and D93 are also sufficient so as not to create a presumption of copying that needs answering, and Ms Johnston and the Trello records provide an explanation for the development of D93 in any event.
375. The many differences are also enough to render D93 not an article made substantially to the design of C93: (i) the neck straps on D93 create a halter effect; (ii) there are two thin under-bust horizontal straps, not one thick one; (iii) the legs are tight fitting in a heavy bandage material; and (iv) the back of the garments are completely different. These differences would also strike the informed user so as to produce a different overall impression (with the additional difference of the strikingly different colours). The differences were even more obvious in the actual garments themselves.
376. D93 does not infringe UKUDR or CUDR in C93.
377. C102 was designed by Alcy Lynch on or about 20 July 2018, and was first offered for sale on the House of CB website on 4 March 2019. The next day, Ms Henderson sent an email to Ms Connell and Ms Johnston saying that she wanted Oh Polly to develop a one shoulder mini dress. She went on: “The recent House of CB one is a copy of Alexander Vauthier but I’ve been seeing similar styles to this everywhere”. She included a link to the Vauthier dress available on the retail site Farfetch. Ms Henderson added “Obviously, ensure it looks different”. Ms Connell found C102 almost immediately. She found 13 other designs, although C102 is the only dress shown in white. She produced 3 different designs for D102 using a computer aided design program. At the same time, Ms Johnston also developed two sketches, which she emailed to Ms Henderson. When the Trello record was created, an image of the Vauthier dress was added. There were discussions about adding a belt. D102 went through several changes during the fitting process. D102 was first offered for sale on the Oh Polly website on 22 May 2019.
C102 D102
378. The significant features of C102 were said to be:
“This item is a one shouldered bodycon dress. It is made from white stretch satin. It has one long sleeve to the left hand side. It has a deep plunge neckline that ends in a V to the centre bust. It has a triangular shaped cup, without sleeve, to the right hand side. It has a tight bodycon fit. It is made from a stretch satin with a slight sheen. The hem ends mid thigh. This item is available in White Stretch Satin. [The Defendants submitted that the sample of C102 is not a bodycon fit and is not stretchy. Further, they pointed to a zip and seam down the entire length of the back of the garment, and padding in the shoulder].”
379. The features of colour and material relate to CUDR only.
380. In relation to originality in the copyright sense (UKUDR), the Defendants relied on a Vauthier dress worn by Elizabeth Gillies in an episode of Dynasty which aired on 1 November 2017 and an Anthony Vaccarello dress from his Fall 2012 collection. As the closest prior design (CUDR), the Defendants relied only on the Vaccarello dress. Both dresses are shown here:
Vauthier Vaccarello
381. In addition, the Defendants relied on a further three dresses to suggest that C102 was commonplace.
382. Having reviewed the images, in my judgment, C102 was not slavishly copied from the Vauthier or Vaccarello dresses, and does not create the same overall impression on the informed user as the Vaccarello dress. The Vauthier dress is a black bow-waisted blazer dress. It has a tailored blazer cut and an asymmetric hemline. It appears to have a wrap-over panel to the front, and a fold-over lapel on one side of the neckline. It has an oversized bow wrap at the waist. The Vaccarello dress is a single lapel blazer dress. The garment has one long sleeve. The asymmetric blazer is cut diagonally across the body. The hem is also cut diagonally. The blazer is attached to a bustier at the side waist, with the bustier extending from the underarm to the waist. Skin is visible under the bustier. The skirt is tight and sits mid-thigh.
383. I have also reviewed the other dresses which the Defendants consider not to be as close to C102. None of them makes C102 commonplace - they are all different designs.
384. C102 is not invalid in relation to UKUDR or CUDR.
385. The Defendants’ case is that Ms Henderson provided the initial instructions in an email to Ms Connell and Ms Johnston, along with an image of the Vauthier dress. Ms Henderson also mentioned a House of CB dress which she said was a copy of the Vauthier dress, but did not attach an image. Ms Henderson gave express instructions to add a belt. Ms Connell found C102 almost immediately - it had been first offered for sale the day before. She created three designs using software, and sent them to Ms Henderson. Ms Henderson chose one of those designs. Ms Johnston also produced sketches, but Ms Henderson favoured Ms Connell’s sketches. The design changed a lot during the development process, with Ms Johnston, Ms Connell, Ms Henderson and Dr Branney all contributing.
386. In my judgment, Ms Connell did not copy C102 - she referred to a number of dresses, but created her own drawings using her own imagination. Although pointed towards C102 by Ms Henderson, Ms Connell did not copy it.
387. Further, the documentary evidence before me was that D102 changed significantly through the development process. The Defendants pointed to a number of differences, including: (i) D102’s belt; (ii) the vertical ribs/boning in D102 that is not present in C102; (iii) the way the panels of fabric that have been used to create D102 differ from those for C102; and (iv) D102 has no zip or seam at the back. Whilst not obvious from the photographs, it is obvious from the samples I was shown that the construction and appearance of the bust also differs markedly. Taking all of these features into consideration, in my judgment, these differences taken collectively are sufficient to render D102 other than an article made substantially to the design of C102. Similarly, the differences are sufficient to strike the informed user so as to produce a different overall impression.
388. D102 does not infringe UKUDR or CUDR in C102.
389. In his closing skeleton argument, counsel for the Defendants wrote:
“The evidence from the designers at trial established that there is no difference in kind between the design processes that each side follows. Each keep image libraries of others’ garments to which they are constantly adding and referring back. Each creates mood boards. Each follows trends. Each uses images of third party garments as a useful reference for the factory when giving instructions on how to make or incorporate specific features. Each has a sampling process which involves experiments and modifying the original design idea and which can result in a significantly different garment at the end of the process.”
390. As will be apparent from the above, those are not my findings of fact on the evidence before me. Whilst I have found that Ms Johnston and Ms Connell designed three of the 20 garments in issue before me (following clear instructions from Ms Henderson in relation to what she wanted), the difference between the process of the Claimants’ designers and Ms Henderson’s 17 designs is significant. I have found that none of the Claimants’ 20 designs was slavishly copied from the third party designs which were before me. Of Ms Henderson’s 17 “designs”, I have found that 7 of the 9 in which the Claimants’ designs were “referenced” were, at law, copied. In every case in which the Claimants’ designs were not referenced, third party designs were. There is no reliable evidence before me that Ms Henderson ever created mood boards, digital or otherwise. What she did was identify third party designs that she liked, and send images of them to the factory to be copied.
391. By the end of the trial, the Claimants’ passing off case could be simply put: by copying many of the aspects of the Claimants’ business, the Defendants have misrepresented to the public that Oh Polly is a sister brand of House of CB. That misrepresentation has caused and is causing the Claimants damage.
392. Counsel for the Defendants accepted that, if such a misrepresentation were made out on the facts, it would constitute passing off. He also appeared to accept that what had in fact happened may amount to unfair competition under the laws of some other jurisdictions. But he denied, vehemently, that the Defendants had engaged in unlawful passing off under English law.
393. The law on passing off is well known and was not in dispute before me. It is also well known that passing off cases are very fact-dependent. Each side cited passages from prior cases which suggest different emphases on the basis of the facts of those cases. I set some of them out here.
394. In Reckitt & Colman Products Limited v Borden Inc [1990] RPC 341 (Jif Lemon) at 406, Lord Oliver of Aylmerton summarised the law of passing off as follows:
“The law of passing off can be summarised in one short general proposition - no man may pass off his goods as those of another. More specifically, it may be expressed in terms of the elements which the plaintiff in such an action has to prove in order to succeed. These are three in number. First he must establish a goodwill or reputation attached to the goods or services which he supplies in the mind of the purchasing public by association with the identifying ‘get-up’ (whether it consists simply of a brand name or trade description, or the individual features of labelling or packaging) under which his particular goods or services are offered to the public, such that the get-up is recognised by the public as distinctive specifically of the plaintiff's goods or services. Secondly, he must demonstrate a misrepresentation by the defendant to the public (whether or not intentional) leading or likely to lead the public to believe that goods or services offered by him are the goods or services of the plaintiff. Whether the public is aware of the plaintiff's identity as the manufacturer or supplier of the goods or services is immaterial, as long as they are identified with a particular source which is in fact the plaintiff. Thirdly he must demonstrate that he suffers, or in a quia timet action that he is likely to suffer, damage by reason of the erroneous belief engendered by the defendant's misrepresentation that the source of the defendant’s goods or services is the same as the source of those offered by the plaintiff.”
395. Counsel for the Defendants asked me to take into consideration case law relating to four aspects (in particular) of the law of passing off.
396. As will be apparent from what I have already set out, there was no contention that the parties trade under different names - HOUSE OF CB versus OH POLLY. There is therefore no misrepresentation in relation to the names used by the two businesses, and none was pleaded. Counsel for the Defendants therefore referred me to a number cases which suggest that cases of passing off by get-up will be more difficult for claimants to establish.
397. In Glaxo Wellcome v Sandoz [2019] EWHC 2545 (Ch) (Glaxo Inhaler) Arnold LJ said this in relation to misrepresentation caused by get-up (in the context of a purple, particularly-shaped inhaler):
“164. It is common ground that, in principle, it is possible for a claimant to acquire a goodwill in the shape and/or colour of a product and/or its packaging such that use of the same or a similar shape and/or colour by a defendant leads to a misrepresentation as to trade origin. The authorities show, however, that it is difficult, although not impossible, for claimants to establish that the shape and/or colour of a product and/or its packaging are distinctive of them.
165. As Jacob J explained in Hodgkinson & Corby [1994] 1 WLR 1564 at 1572-1573:
“The plaintiff’s problem of proof when there is no manifest badge of trade origin such as a trade mark becomes hard. This is so in the case of a descriptive or semi-descriptive word such as ‘camel hair’. It is perhaps even more so where one is concerned simply with the appearance of the article with no self-evident trade origin frill or embellishment. For people are likely to buy the article because of what it is, not in reliance on any belief of any particular trade origin. This is so whether they buy it for its eye-appeal (e.g. glass dogs) or for what it does (e.g. the copy Rubik cube…
The plaintiff’s problem of proof lies in relation to the first two items of the trinity, which are related. It is not good enough for him to show that his article is widely recognised - has a ‘reputation’ in that general sense. …
I believe that [Learned Hand J in Crescent Tool Co v Kilborn & Bishop Co (1917) 247 F 290 at 300-301] exactly encapsulates what must be shown when the plaintiff is complaining, in a passing off action, about a copy of his product as such. Is the public ‘moved to buy by source?’
It is, I think, because the difficulties of proof are so great that successful cases of passing off based on the shape of goods are so rare.”
166. Similarly, Floyd J (as he then was) stated in Numatic International Ltd v Qualtex Ltd [2010] EWHC 1237 (Ch), [2010] RPC 25 at [39]:
“It is recognised that it is more difficult to acquire a sufficient reputation and goodwill in the shape or get-up of a product. Whilst the principal function of a brand name is to denote origin, the shape and get-up of a product are not normally chosen for such a purpose. A member of the public seeing a product which looks identical to another (a red cricket ball is an example) does not necessarily, or even normally, conclude that they come from the same source. The claimant must prove that the shape of its goods has come to denote a particular source to the relevant public...””
398. I was also referred to the judgment of HHJ Hacon in Moroccanoil Israel v Aldi [2014] EWHC 1686 (IPEC) (Moroccanoil):
“30. MIL’s case is that a misrepresentation is generated by a combination of the name and get-up of Aldi’s product but the greater emphasis was placed on similarities in the get-up. Mr Edenborough submitted that passing off cases based on get-up that were successful were rare indeed, Jif Lemon being one of the few exceptions which had depended on unusual findings of fact at first instance. He referred to Professor Wadlow’s summary at ¶8-133 of The Law of Passing Off (4th ed):
“The difficulty confronting the claimant in all actions for passing-off based on get-up is that it is unusual for one trader’s goods to be distinguished from those of his competitors exclusively, or even primarily, by their get-up. Normally a brand name or other mark is chosen and given prominence and it is this on which consumers are expected and encouraged to rely. To make out a case based solely on similarities of get-up the claimant must show that deception is likely to [sic] notwithstanding the absence of his own brand name on the defendant’s goods and the likely presence there of the defendant’s brand name and perhaps other distinguishing matter. Not surprisingly, the cases in which passing-off has been found have predominantly been ones of deliberate deception.”
31. I agree with this summary. It is not a proposition of law, more a statement of the realities of the market place. Goods are almost always referred to by their trade names, not their get-up. Get-up may play a greater role in the recall and recognition of a particular product, but if the respective names of the claimant’s and defendant’s goods are distinctive, a misrepresentation by reason of similar get-up is likely to depend on the relevant public not noticing the name on the defendant’s product, or on a label attached to it, as was the case in Jif Lemon.”
399. Finally, I was referred to the comments of Miss Recorder Amanda Michaels (sitting as an Enterprise Judge) in Gama Healthcare Ltd v Pal International Ltd [2016] EWHC 75 (IPEC):
“29. It is not easy to establish passing off where the allegedly distinctive indicia consist only of get-up, and the parties’ products bear distinctive names. This was demonstrated in Schweppes Ltd v Gibbens (1905) 22 RPC 601. The parties’ goods were sold in similarly embossed bottles bearing labels of very similar shape, design and colour scheme, and wording in a similar layout and font. However, they respectively bore the distinctively different brand names “SCHWEPPES” and “GIBBENS” prominently on the label. Lord Halsbury LC held at pp.606-7:
“The question that we have to determine is whether in selling the bottle a person is likely to be deceived by the resemblance of the one thing to the other; and if a person is so careless that he does not look, and does not, … “treat the label fairly,” but takes the bottle without sufficient consideration and without reading what is written very plainly indeed upon the face of the label on which the trader has placed his own name, then you certainly cannot say he is deceived—in fact he does not care which it is. …
The whole question in these cases is whether the thing—taken in its entirety, looking at the whole thing—is such that in the ordinary course of things a person with a reasonable apprehension and with proper eyesight would be deceived.”
30. Similarly, Lord Lindley held at p. 607:
“It appears to me that the real answer to the Appellants’ case is this—that they invite your Lordships to look, not at the whole get-up, but at that part of the get-up which suits their case. The resemblances here are obvious enough, but, unfortunately for the Appellants, so are the differences. The differences are not concealed; they are quite as conspicuous as the resemblances. If you look at the whole get-up, and not only that part of it in which the resemblances are to be found, the whole get-up does not deceive.””
400. Counsel for the Claimants did not suggest that these passages do not accurately set out the law, and I accept them. However, none of these cases involved an allegation that the misrepresentation in issue was that of the defendant being a sister brand of the claimant. The very nature of that allegation is that the claimant and defendant are trading under different brands - but that the get-up used is so similar that consumers will consider there is a connection. That is unlike the circumstances in each of the above cases (and indeed in Jif Lemon) where the alleged misrepresentation was that the defendant’s product was the claimant’s product.
401. Counsel for the Defendants emphasised that it is not enough for the Claimants to prove that the Defendants have obtained a leg-up - the Claimants must prove that there has been deception. Again, he relied on the judgment of Arnold LJ in Glaxo Inhaler at paragraphs 158 to 162:
“158. As Jacob J forcefully stated in Hodgkinson & Corby Ltd v Wards Mobility Services Ltd [1994] 1 WLR 1564 at 1569-1570:
“I turn to consider the law and begin by identifying what is not the law. There is no tort of copying. There is no tort of taking a man's market or customers. Neither the market nor the customers are the plaintiff's to own. There is no tort of making use of another's goodwill as such. There is no tort of competition. …
At the heart of passing off lies deception or its likelihood, deception of the ultimate consumer in particular. Over the years passing off has developed from the classic case of the defendant selling his goods as and for those of the plaintiff to cover other kinds of deception, e.g. that the defendant's goods are the same as those of the plaintiff when they are not, e.g. Combe International Ltd v Scholl (UK) Ltd [1980] RPC 1; or that the defendant's goods are the same as goods sold by a class of persons of which the plaintiff is a member when they are not, e.g. Erven Warnink Besloten Vennootschap v J. Townend & Sons (Hull) Ltd [1979] AC 29 (the Advocaat case). Never has the tort shown even a slight tendency to stray beyond cases of deception. Were it to do so it would enter the field of honest competition, declared unlawful for some reason other than deceptiveness. Why there should be any such reason I cannot imagine. It would serve only to stifle competition.
The foundation of the plaintiff's case here must therefore lie in deception…”
159. It is not enough if members of the public are merely caused to wonder. As Jacob LJ explained in Phones 4U Ltd v Phone4U.co.uk Internet Ltd [2006] EWCA Civ 244, [2007] RPC 5:
“16. The next point of passing off law to consider is misrepresentation. Sometimes a distinction is drawn between ‘mere confusion’ which is not enough, and ‘deception’, which is. I described the difference as ‘elusive’ in Reed Executive Plc v Reed Business Information Ltd [2004] RPC 40. I said this, [111]:
‘Once the position strays into misleading a substantial number of people (going from ‘I wonder if there is a connection’ to ‘I assume there is a connection’) there will be passing off, whether the use is as a business name or a trade mark on goods.’
17. This of course is a question of degree—there will be some mere wonderers and some assumers—there will normally (see below) be passing off if there is a substantial number of the latter even if there is also a substantial number of the former.
18. The current (2005) edition of Kerly contains a discussion of the distinction at paras 15–043 to 15–045. It is suggested that:
‘The real distinction between mere confusion and deception lies in their causative effects. Mere confusion has no causative effect (other than to confuse lawyers and their clients) whereas, if in answer to the question: “what moves the public to buy?”, the insignia complained of is identified, then it is a case of deception.’
19. Although correct as far as it goes, I do not endorse that as a complete statement of the position. Clearly if the public are induced to buy by mistaking the insignia of B for that which they know to be that of A, there is deception. But there are other cases too—for instance those in the Buttercup case. A more complete test would be whether what is said to be deception rather than mere confusion is really likely to be damaging to the claimant's goodwill or divert trade from him. I emphasise the word ‘really’.”
160. In order for there to be passing off, a substantial number of members of the public must be misled: see Neutrogena Corp v Golden Ltd [1996] RPC 473 at 493-494 (Morritt LJ). Furthermore, it is not enough that careless or indifferent people may be led into error: see Norman Kark Publications Ltd v Odhams Press Ltd [1962] 1 WLR 380 at 383 (Wilberforce J).
161. The correct approach to this question was well described by Jacob J at first instance in Neutrogena at 482:
“The judge must consider the evidence adduced and use his own common sense and his own opinion as to the likelihood of deception. It is an overall ‘jury’ assessment involving a combination of all these factors, see ‘GE’ Trade Mark [1973] RPC 297 at page 321. Ultimately the question is one for the court, not for the witnesses. It follows that if the judge's own opinion is that the case is marginal, one where he cannot be sure whether there is a likelihood of sufficient deception, the case will fail in the absence of enough evidence of the likelihood of deception. But if that opinion of the judge is supplemented by such evidence then it will succeed. And even if one's own opinion is that deception is unlikely though possible, convincing evidence of deception will carry the day. The Jif lemon case (Reckitt & Colman Products Ltd v Borden Inc [1990] RPC 341) is a recent example where overwhelming evidence of deception had that effect. It was certainly my experience in practice that my own view as to the likelihood of deception was not always reliable. As I grew more experienced I said more and more ‘it depends on the evidence.’”
162. It is not fatal to a claim for passing off that there is no evidence of actual confusion, but where the rival goods have been sold side by side for a long period the absence of such evidence is very material unless satisfactorily explained: see Kerly’s Law of Trade Marks and Trade Names (16th ed) at 23-020.”
402. Counsel for the Defendants analysed the case before me as resembling “more closely one of unfair competition than of passing off”. He referred me to Jacob LJ’s rejection of passing off as a tort of unfair competition in L’Oréal SA and Ors v Bellure NV and Ors [2007] EWCA Civ 968 (L’Oréal v Bellure). Discussing the Hodgkinson case referred to by Arnold LJ above, Jacob LJ said:
“138. The facts of the case illustrate just what could be involved in some general law of unfair competition. The defendants had copied a special cushion which prevented sores for immobile users. There was no copyright, registered design or patent. So it was said the copying was “unfair”. The claim was rejected because there was no misrepresentation.
139. The rejected complaint shows just how anti-competitive a law of unfair competition would or might be. What one man calls “unfair”, another calls “fair”. The market involves the interests of traders, their competitors and consumers. They all have different perspectives. An established trader would like the law to hold off all his competitors—and as far as possible. He would want to prevent all copying of his products—and for as long as possible, preferably indefinitely. He would want as wide a gap as possible between his trade marks and those of others. He would oppose any form of comparative advertising. A newcomer will want to be able to copy—and to improve. He will want to be able fairly to advertise comparatively. And the consumer will want the best deal he can get. He would oppose anything deceptive, but probably nothing else.
140. So I think there are real difficulties in formulating a clear and rational line between that which is fair and that which is not, once one goes outside the requirement of no deception.
141. Moreover, the basic economic rule is that competition is not only lawful but a mainspring of the economy. The legislator has recognised that there should be exceptions. It has laid down the rules for these: the laws of patents, trade marks, copyrights and designs have all been fashioned for the purpose. Each of them have rules for their existence and (save for trade marks) set time periods for existence. Each has their own justification. It is not for the judges to step in and legislate into existence new categories of intellectual property rights. And if they were to do so they would be entering wholly uncertain territory.”
403. I was also referred to a case that pre-dates Jif Lemon, which was said to be analogous to the facts as pleaded by the Claimants in this case. In Wertheimer v Stewart, Cooper & Co (1906) 23 RPC 481, the defendant was said to have adopted the same advertising mode as its competitor, including the same “embellishments of his language, which are striking and have been copied to the very last particular”, but continued to use his own name (Stewart Cooper & Co) in the advertisement, which was different to the claimant’s (Gerome & Co). The claimant sued for passing off. Kekewich J said at 483:
“But unless that injures the Plaintiff in his property it seems to me to be wholly immaterial in a Court of law. Now, how does it injure his property? What is his property? He has no property in the advertisement. Of course he does not claim it as copyright. He has property in his goods; in any character by which his goods are known; in any distinguishing device or name; and he has the right also to protect his own name. None of these things, as it seems to me, does the Defendant at all offend against, so that the Court can, in any way, reach him.”
404. For my part, I do not consider that case to be analogous to this one - it is at the very centre of the Claimants’ argument before me that the names of the two businesses differ. Further, the allegations here go significantly beyond the wording of advertisements, and cover instead a whole array of aspects of the two businesses. Again, I did not hear counsel for the Claimant cavil with the propositions set out above - but they are all cases decided on their particular facts. As Jacob J (as he then was) said at first instance in Neutrogena, it all depends on the evidence.
405. It is well known that intention is not a requirement for a successful claim of passing off. It is not a requirement that the defendant intended to mislead. And a defendant who intends to mislead but fails is not thereby liable for “attempted passing off”. I was referred to the following passage from Lindley LJ’s judgment in Slazenger & Sons v Feltham & Co (1889) 6 RPC 531 at 537-538 (emphasis added):
“The real question is, whether the Defendants are endeavouring with any probability of success, to pass off their goods as those of the Plaintiffs. That depends upon the evidence, and the evidence is this that, whereas the Plaintiffs had got a trade mark, one part of which consists of the word “Demon”, and whereas the Plaintiffs had put that trade mark on their bats and put the word “Demon” at the top of the rim the Defendants look through the dictionary, see how close they can get to “Demon”, pick out “Demotic”, and put “Demotic” in exactly the same spot where the Plaintiffs put “Demon.” They put their own name on the bats, no doubt, and do not use the registered trade mark. Well, what is that for? One must exercise one’s common sense, and if you are driven to the conclusion that what is intended to be done is to deceive if possible, I do not think it is stretching the imagination very much to credit the man with occasional success or possible success. Why should we be astute to say that he cannot succeed in doing that which he is straining every nerve to do?”
406. I was also referred to the judgment of Hildyard J in Fine & Country v Okotoks Limited [2012] EWHC 2230 (Ch):
“179. Nevertheless, the fact that the First Defendant decided to live dangerously is, as it seems to me, still a factor which I am entitled to take into account as confirming my view that the average consumer might well be confused, and indeed to my mind should be taken as having been so: to quote again from the judgment of Robert Walker J (as he then was) in United Biscuits (UK) Ltd v Asda Stores Ltd [1997] RPC 513 at 531:
“I cannot escape the conclusion that, while aiming to avoid what the law would characterise as deception, they were taking a conscious decision to live dangerously. That is not in my judgment something that the court is bound to disregard.”
180. In my view, the recognition on the part of a defendant that his sign has the propensity to confuse and deceive, and that though he is himself satisfied that he is the right side of the line, there is an identified risk that he may not be, may well support a conclusion drawn from the court’s own perception and be evidence of (objective) deception. Further, it may colour the court’s approach in considering the evidence of the likelihood of confusion, to which I now return: I need to address the evidence in this context of the Defendants’ other ‘live’ witnesses, Mr Edward Mead and Mrs Green.”
407. The Defendants denied any intention to deceive or to “live dangerously” (see also HHJ Hacon’s comments in Moroccanoil at paragraph 35). But they did accept that the Defendants’ intention was something I could take into account. The Claimants went further, noting that where a defendant which knows its market well sets out to emulate aspects of a claimant’s get up, the Court can conclude that its purpose was to cause deception. I was referred to Sydneywide Distributors Pty Ltd v Red Bull Australia Pty Ltd [2002] FCAFC 157, a decision of the Full Court of the Federal Court of Australia (Branson, Weinberg and Dowsett JJ) at paragraph 117:
“Without wishing to labour the point unduly, we again point out that where a trader, having knowledge of a particular market, borrows aspects of a competitor’s get-up, it is a reasonable inference that he or she believes that there will be a market benefit in so doing. Often, the obvious benefit will be the attraction of custom which would otherwise have gone to the competitor. It is an available inference from those propositions that the trader, with knowledge of the market, considered that such borrowing was “fitted for the purpose and therefore likely to deceive or confuse ...”. Of course, the trader may explain his or her conduct in such a way as to undermine the availability of that inference. Obviously, this reasoning will only apply where there are similarities in get-up which suggest borrowing.”
408. Counsel for the Claimants suggested that this is “the logical consequence” of the principle set out by Lord Simonds in Office Cleaning Services Limited v Westminster Window and General Cleaners Limited (1946) 63 RPC 39 at 42:
“But if the intention to deceive is found, it will be readily inferred that deception will result. Who knows better than the trader the mysteries of his trade?”
409. Counsel for the Defendants did not suggest that the Federal Court of Australia’s reasoning did not reflect English law. The parties were therefore in agreement that I could take intention into account.
410. As set out above, the Defendants accepted that there was no legal bar to the case of passing off as alleged by the Claimants - they conceded that passing off is actionable for a misrepresentation as to trade connection (sister brand) as opposed to a misrepresentation that one undertaking is another. I was referred to a further warning, this time in Wadlow on Passing Off (Professor Christopher Wadlow, The Law of Passing-Off: Unfair Competition by Misrepresentation, Sweet & Maxwell, 5th Ed, 2016) at paragraph 7.14:
“In principle, any form of alleged or implied connection with the claimant may be actionable as passing-off, but always provided that it is a material one in the sense of being really likely to cause damage. This in turn normally requires that the representation should be one on which the relevant public are likely to rely. As Millett LJ explained in Harrods v Harrodian School: [at p.712]
“In a written summary of the plaintiffs’ points […] it was submitted:
‘In this case the belief engendered [in the minds of the public] is probably that Harrods sponsor or back the school. Obviously not every connection will found an action for passing off … but where the representation is to the effect that the plaintiff is behind the defendant in some way, that is a classic case.’
This is too widely stated. In my judgment the relevant connection must be one by which the plaintiffs would be taken by the public to have made themselves responsible for the quality of the defendant’s goods or services. …
It is not in my opinion sufficient to demonstrate that there must be a connection of some kind between the defendant and the plaintiff, if it is not a connection which would lead the public to suppose that the plaintiff has made himself responsible for the quality of the defendant’s goods or services.
411. The Claimants also relied on the two businesses being in the same field of activity, citing the Court of Appeal in Harrods v Harrodian School [1996] RPC 697 at 714 (Harrods):
“The absence of a common field of activity, therefore, is not fatal; but it is not irrelevant either. In deciding whether there is a likelihood of confusion, it is an important and highly relevant consideration
"...whether there is any kind of association, or could be in the minds of the public any kind of association, between the field of activities of the plaintiff and the field of activities of the defendant":
Annabel's (Berkeley Square) Ltd v. G Schock (trading as Annabel S Escort Agency) [l972] RPC 838 at page 844 per Russell LJ.
In the Lego case Falconer J likewise held that the proximity of the defendant's field of activity to that of the plaintiff was a factor to be taken into account when deciding whether the defendant's conduct would cause the necessary confusion.”
412. Again, counsel for the Defendants did not take issue with these statements of law, and I accept them.
413. The Claimants relied on evidence of actual confusion, and referred me to the pithy statement of Millet LJ in Harrods:
“Evidence of actual confusion is always relevant and may be decisive. Absence of evidence may often be readily explained and is rarely decisive. Its weight is a matter for the judge.”
414. Counsel for the Defendants did not suggest otherwise (although he did submit that none of the evidence relied on by the Claimants was actually evidence of confusion).
415. I have taken into account in my analysis all of the cases to which I was referred, including those I have set out above. But I also remind myself that what matters are the facts of the case in front of me. As counsel for the Defendants accepted, none of the cases cited to me had facts identical (or even similar) to the facts of this case, and that is what matters most. I have heeded the warnings that passing off is not a tort of unfair competition, and that deception by get-up will be difficult for a claimant to prove.
416. The Claimants rely on goodwill in what was defined in the Re-Amended Particulars of Claim as the Claimants’ Business and Marketing Style and Get-Up:
“…the concepts, style and get-up of its business, fashion, collections, garment design, packaging, marketing, publicity, website and social media, including the design and high quality of its collections and garments, the highly distinctive style of its packaging, marketing and websites, including the design and layout, look and feel, concept, style, presentation and content of its website, the use of social media and the customer experience and customer service on its website and stores…”
417. This definition is, of course, very broad. Particulars and sub-particulars were given at length, occupying 13 pages of the Re-Amended Particulars of Claim. I do not need to set them all out here, but I have taken them all into account. Some of the indicia were said to have more significance than others. By the end of the trial, it seemed to me that the following were the main indicia relied on:
i) Business model and focus;
ii) Garment design;
iii) Locations, themes and styling of photoshoots;
iv) Models;
v) Packaging;
vi) Logos; and
vii) Websites.
418. The Claimants emphasised that it is the combination of all of the pleaded elements that amounts to passing off. The misrepresentation was said to be:
“The Defendants have adopted the same or strikingly similar concepts, style and get-up for their business, fashion, collections, garment designs, packaging, marketing, publicity, website design and social media as [House of CB]. The advertisement, offer for sale and sale by the Defendants of “look-alike” fashion, collections, garments and accessories, in “look-alike” packaging on a “look-alike” website and the wholesale copying and/or imitation of [House of CB’s] Business and Marketing Style and Get-Up constitutes a misrepresentation by the Defendants…”
419. The Defendants submitted (and I accept) that I must also take into account the differences between the two businesses. I have taken them all into account, but note, for brevity, that the main differences were said to be:
i) Names;
ii) Whilst both business name their garments, they do so differently, with House of CB naming their garments with women’s names, and Oh Polly using puns or innuendo;
iii) Differences in the websites;
iv) House of CB does not sell jewellery whereas Oh Polly does;
v) Oh Polly introduced a photographic layer to its packaging in 2018 or 2019 which is not present in the House of CB Packaging; and
vi) Oh Polly has a different social media focus and strategy.
420. I return to each below.
421. The relevant dates were said by the Claimants to be:
i) The date on which the Defendants transferred their business from eBay to the initial website (being approximately August 2015); or, alternatively
ii) The date of the Oh Polly rebrand in May 2018.
422. I turn now to the evidence before me. There was a vast amount of it. I am not able to refer to all of it here, nor is it necessary for me to do so. Whilst many of the particularised elements of the Claimants’ Business and Marketing Style and Get-Up were discussed in some detail, what matters is how consumers react to the overall combination. So, having reviewed each of the similarities relied on by the Claimants, and differences relied on by the Defendants, I must then step back, and assess, overall, whether there has been an actionable misrepresentation. There is therefore significant interplay between the factors set out below.
423. As a starting point, I dismiss counsel for the Defendant’s submission that what really matters is the name of each business: because they differed, he said, there can be no passing off. There was plenty of evidence before me, including from the Defendants’ witnesses, that each business is far more than its name and the clothes it sells. For example, Dr Branney said:
“…you want the customer to feel as if they are buying into a certain kind of lifestyle and not just a product. This is highly common in the women’s online fashion industry.”
424. It was clear from both sides’ evidence that the “lifestyle” (I prefer the word “brand”) was created through the various indicia relied on by each side as similarities and differences. There was evidence before me that each of the indicia is important to consumers - to greater and lesser degrees, of course, but important none the less. For example, Dr Branney accepted that all the following were important to the Oh Polly brand: style of garments; models; packaging; locations; use of social media; aspects of marketing strategy; and geographic origin of the brand. Ms Henderson accepted that the following were important to the Oh Polly brand: garments; the look of photoshoots; packaging; website layout; style of e-commerce photography; and use of social media. I therefore reject the submission that all that matters is the name: that was clearly not the evidence of the Defendants’ witnesses.
425. I also do not accept the submission from counsel for the Defendants that where a diffusion or sister brand is created, it will be named similarly to the headline label. Whilst he gave the examples of DKNY for Donna Karan and CK Jeans for Calvin Klein, other examples include House of CB and Mistress Rocks in this case, as well as Mui for Prada, and T for Alexander Wang.
426. Both the Claimants and the Defendants operate an internet business selling celebrity-inspired fashion. Both use social media heavily. Both generate publicity from third parties wearing their garments (including influencers and celebrities). Neither does much or any traditional hard copy advertising (magazines, bill-boards etc), although others in the industry (such as BooHoo) do. Both the Claimants and the Defendants focus on figure-hugging bandage and bodycon dresses designed for a curvy figure. Both sell dresses for younger women to wear on a night out. These similarities were not contested: the parties are competitors, operating very similar businesses.
427. The Defendants suggested that their social media approach was “quite different”, with Ms Mitchell suggesting that Oh Polly focusses on getting lots of engagement, including lots of likes and comments, whereas, she suggested, House of CB focuses more on a strong aesthetic. I accept that Ms Mitchell considered that to be her focus, but it is inconsistent with the other evidence before me, particularly the care with which Ms Henderson approached the aesthetics of the Oh Polly social media presence, and the steps taken to make it look like House of CB. I do not consider the social media outcomes to be different, even if Ms Mitchell felt her rationale for the Oh Polly social media strategy was different.
428. I accept Counsel for the Claimants’ submission that the designs of the parties’ garments is an important aspect of the distinctiveness of their brands - it is something each relies on for the “lifestyle” (or brand) to which I have referred above. Dr Branney said:
“Our garments are also very tight-fitting and are designed to stand out and get our customers seen: typically they will show off either the wearer’s cleavage or their legs, and are recognisable as Oh Polly garments for the sexy design, “bodycon” fit and high-quality material.”
429. The Claimants relied on 123 pairs of garments, some of which were taken from the Claimants’ design infringement case, and some of which were not. In each case, the pair shows a House of CB garment side by side with an Oh Polly garment which is similar. Counsel for the Claimants submitted that, “taken collectively”, these show that the “house style” of Oh Polly is “objectively similar” to that of House of CB: closely fitted, glamorous, sexy, revealing, often incorporating stretchy fabrics, bold colours, with frequent use of motifs such as lacing, thigh slits, peplum hems, cut-out panels and deep plunge necklines.
430. The Claimants also relied on the Defendants’ admitted “references” of 27 of the designs in issue in the proceedings, as well as any findings I might make as to design infringement (which are set out above).
431. The Claimants also relied on public recognition of the frequent similarity of Oh Polly designs to House of CB designs, including the following social media posts:
“Girl I got a dress from Oh Polly for $63 that was $180 on House of CB…”
“Seeing as OhPolly copy all of HouseOfCb designs, are their clothes good quality?”
“Wish @ohpoly would put as much effort into makin their website work properly as it didn’t take people’s money & cancel orders, as it did into copying designs from House of CB”.
“Oh Polly … just COMPLETELY copy House of CB designs…”
“Oh Polly just rip off House of CB designs…”
“Oh Polly fully copying House of CB designs”
432. Whether or not the above examples indicate confusion, they do indicate the consumers’ views that the garments were similar, and that the House of CB garments were perceived as being first in time.
433. The Defendants submitted that the designs in issue in these proceedings were a small fraction of the many thousands of designs produced to date by the two business. However, as counsel for the Claimants submitted, and I accept, the Defendants did not seek to argue that the garments before me were not representative of the two businesses overall. Further, counsel for the Defendants submitted that the social media comments above do not evidence confusion - but they do demonstrate independent public perception that some Oh Polly garments are seen as similar to (indeed, copies of) House of CB garments.
434. In relation to the 20 garments in issue before me, I have found that 7 of the Defendants’ garments infringe the Claimants’ unregistered design rights. The corollary of that is that the other 13 garments do not infringe. Be that as it may, there is clear evidence of a public perception, at least among some social media posts, that Oh Polly has copied House of CB’s garments, in the context where the parties are agreed that the two businesses operate in the same market, in similar ways.
435. The Claimants relied on the use of curvy models, styled with a non-trend-based highly glamourous look, photographed in provocative poses in striking locations.
436. Ms Henderson accepted that the “look” of a brand’s photoshoots - including hair, make-up, posing and location - is key:
“…every brand has their own unique style based on their use of make-up and hair styles, shooting on location or in a studio and the sort of garment the model is wearing.”
“What makes a difference is what you do with the model, the location, the hair and make-up when you’re shooting.”
437. Dr Branney was clear about Oh Polly’s intention in two email exchanges, one with a modelling agency and one with a photographer (emphasis added):
“We are a very similar style to House of CB but our prices are high-street prices. We want our pictures to emulate the style that House of CB present”.
“If you are familiar with the likes of House of CB then that is the style of image we are looking for.”
438. I was taken to many examples of images from Oh Polly which were said to be both similar to and later in time than House of CB images. Whilst some were rather more “generic”, others were strikingly similar. Two of the more striking examples were a photoshoot with cars at a racetrack and greenish smoke, and a beach shoot where the models were posed against the same collection of driftwood. Images are shown here (the comments between the images are those of the Claimant):
439. The documents before me show that shoot locations were often discussed by reference to House of CB, including these two emails received by Ms Henderson (emphasis added):
“Found 2 locations for the Valentines campaign … I think these locations are very textured (quite like House of CB latest campaign).”
“I’m proposing we fly to Lanzarote to shoot at the black sand/volcanic rocks … I’ve drawn most of the inspo from House of CB campaigns. Their imagery is always really clean, crisp and eye-catching which is what I aim to achieve with this campaign”.
440. In evidence before me were strikingly similar campaigns shot against volcanic rocks.
441. The Defendants submitted that there were only so many locations available in Los Angeles, and it was coincidental that the Defendants had used the same locations as the Claimants.
442. I do not accept that submission. As the Lanzarote shoot demonstrates, there are suitable locations in many part of the world outside Los Angeles. The documentary evidence is clear that the Lanzarote shoot was chosen to mimic House of CB. The examples before the court show a clear course of conduct by the Defendants, often within weeks, to use identical locations for shoots as those used by the Claimants.
443. The Defendants also submitted in relation to the posing of models at shoots that models have a number of learned poses which they tend to use, and that the Defendants did not mandate how models should pose for their photoshoots. Again, I reject this submission. In evidence before me was a very detailed memo from Ms Henderson setting out in great detail how she wished models for a shoot to be posed. Ms Henderson wrote:
“Second problem is poses, apart from 1 or 2 items, the majority of poses are the same. … I’ve even sent you a print out of different poses and asked that PLT/Missguided/House of CB be monitored so you can see exactly what I want (from the latter, house of CB/Oh Polly) and what I don’t want from the former (PLT/Missguided).”
444. This email demonstrates that Ms Henderson was monitoring House of CB’s model’s poses, and suggesting that they be emulated for Oh Polly shoots.
445. As late as January 2019, after these proceedings were underway, Ms Henderson wrote:
“The Stassie video should be fast, sexy and fierce. Lots of movements - lots of sharp catwalk style walk ups, dances, cute movements. The House of CB I passed over does this really well and is unique and more “out there” than the standard boring posts/angles.”
446. These documents demonstrate a keen and detailed interest from Ms Henderson in how House of CB poses its models, and also demonstrates that she directed Oh Polly models how to pose. I therefore do not accept the submission that models were left to their own devices on shoots. To the contrary, I find that the Defendants directed the models in how to pose, and then chose the photographs they wished to use, rejecting any poses with which they were not happy.
447. In my judgment, the similarity of poses in identical locations, combined with the use of the same models (see below) was a course of conduct adopted by the Defendants in order, as Dr Branney put it, to emulate the style of House of CB’s shoots.
448. The Claimants’ case is that the similarity of look and aesthetic of Oh Polly’s images is enhanced by the Defendants’ use of the same models as used by House of CB and its sister brand Mistress Rocks. Ms Richards’ evidence was that all except one of the models used by House of CB for photographic work had been used by the Defendants. The uses can be summarised in the following table:
|
Model |
Date of first job with the Claimants |
Date first seen in Oh Polly |
|
Audreyana Michelle |
25 January 2017 |
10 July 2017 |
|
Sierra Skye |
23 August 2017 |
11 May 2018 |
|
Tiana Parker |
27 September 2017 |
4 July 2018 |
|
Brookell B |
20 February 2017 |
20 September 2018 |
|
Rona Mahal |
21 August 2017 |
18 October 2018 |
|
Ravel Tracey |
11 October 2018 |
8 November 2018 |
|
Mariama Diallo |
19 April 2017 |
21 March 2019 |
|
Carolina V Marie |
28 May 2019 |
14 August 2019 |
|
Priscilla Ricart |
February 2019 |
16 November 2019 |
|
Maggie Rawlins |
1 July 2018 |
29 January 2020 |
|
Arbenita Ismajli |
24 October 2018 |
9 February 2020 |
|
Muriel Villera |
24 August 2015 |
October 2019 |
449. I should add that no criticism was made of the models themselves - both sides conceded that the models were quite entitled to work for both businesses, as none was under an exclusive contract with House of CB or Oh Polly.
450. Ms Henderson emailed a modelling agency in early 2016:
“The style we are going for is very much like Hamissa with House of CB”.
451. Again, I was provided with social media commentary from consumers who it was said had noticed the similarities in the models used by the two businesses:
“oh Polly wanna be house of cb so bad lol down to the same damn models”
“I find it weird how oh polly use the same models as house of CB and create very similar social media content.”
“Oh and forgot to mention [@ohpolly copies] the shoot locations, models, and the overall content … it’s so sickening.”
452. Whether or not these consumers were confused, these comments indicate that the authors had noticed the similarities.
453. The Claimants submitted that the use of the same models photographed in strikingly similar poses in the same or similar locations and similarly styled suggests that “it is effectively the same lifestyle and brand that is being portrayed and sold to the consumer.” I accept that submission.
454. The Defendants submitted that there are only so many curvy models in Los Angeles and hence it was not surprising that some of the same models were used by both entities. I do not accept that submission. Whilst there was no evidence before me as to the number of curvy models in Los Angeles, it does seem to me, first, that as a Scottish brand, the Defendants are not limited to models in Los Angeles and, second, there must be, in any event, more than 12 curvy models in Los Angeles. Again, I conclude that there was a course of conduct adopted by the Defendants to use the same models who had already been used by House of CB in order, again paraphrasing Dr Branney, to emulate House of CB. Consumers noticed the similarities, and commented on them in social media.
455. The Claimants’ packaging changed when it moved from Celeb Boutique to House of CB. The Celeb Boutique packaging was a bright pink colour with slogans printed on it to give the appearance of handwriting. The website address was printed in white capital letters. The Celeb Boutique packaging is shown here:
456. On moving to House of CB, the Claimants’ packaging became more sophisticated and minimal in design, with a lighter shade of pink, a logo front and centre, and the website and social media icons on the side of the box. The House of CB packaging is shown below:
457. In May 2018, the Defendants updated the Oh Polly packaging. Previously, Oh Polly garments had been delivered in a bright pink bag with slogans printed on it to look hand-written. The packaging had OHPOLLY.COM on the side in white capital letters, along with social media icons. An example of the pre-2018 packaging is shown here:
458. The new Oh Polly packaging was a pale pink box with a central logo, and website and social media icons on the side. An example is shown here:
459. The evidence before me was that the House of CB and Oh Polly boxes are very different from those of competitors including Abyss by Abby, Zara, I Am Gia, Meshki and Tiger Mist.
460. Ms Henderson accepted that the new Oh Polly boxes were similar to House of CB’s: she sent an email on 16 June 2018 stating (emphasis added):
“Absolute main priority here is to include the upper layer photograph in the boxes, as right now these will be looking identical with House of CB”.
461. Both House of CB and Oh Polly use a pink “anti-wardrobing” ribbon. “Wardrobing” is a term used to describe the activity of customers who order garments on-line, wear them, and then return them to the retailer for a refund. The anti-wardrobing ribbon is a thick ribbon which passes through a sleeve or other part of the garment and is secured by a rivet. The ribbon allows the garment to be tried on for fit, but prevents the wearer from wearing the garment publicly. Returns are not accepted if the ribbon has been severed. It was common ground that the Defendants adopted an anti-wardrobing ribbon having seen it used with House of CB garments.
462. The similarity of packaging was also noted in social media posts:
“How have you not sued yet? @HouseofCB. Girls did you also realise even the logo and colours are similar. Even the packaging. @ohpolly @ohpollyhelp you should be embarrassed.”
463. The Defendants claimed that their packaging was different, because it included a photographic sheet. However, there were no examples before me of what that photographic sheet looked like. Further, it was not introduced until after the Relevant Dates. Indeed, it appears to have been introduced because of concerns by Ms Henderson that the Oh Polly packaging was too close to that of House of CB: see her email I have excerpted above.
464. It is obvious to me from the photographs set out above that Oh Polly initially adopted packaging similar to that of Celeb Boutique - bright pink, printed with “handwritten” slogans. There was no evidence that any other trader was doing that at the time. Following the Claimants’ rebrand to House of CB, Oh Polly followed, adopting a lighter shade of pink, abandoning the “handwritten” slogans, and opting for a more sophisticated and minimal design. That look, again, was different from what other competitors were doing. In my judgment, this was in order to look similar to House of CB, and dissimilar from any other competitors. It was a similarity that consumers noticed.
465. I do not consider that the anti-wardrobing ribbon was something consumers noticed as particular to House of CB, and hence the fact that Oh Polly also used such a ribbon will not have contributed to the pleaded misrepresentation.
466. The Claimants relied on similarities between the designs of the House of CB logo and the Oh Polly logo. The logos on the packaging are shown above.
467. Since 2014, the Claimants have used a logo with the word HOUSE in large serif capital letters; a very small OF; the letters CB in a curvy script as if handwritten; all against a pale pink background.
468. Prior to May 2018, the Oh Polly logo was the words OH POLLY in san serif capital letters, in pink against a grey background. An example is shown here:
469. From May 2018, the Oh Polly logo moved closer to the House of CB logo, such that it comprised the word POLLY in large serif capital letters; the word OH in a curvy script as if handwritten; all against a pale pink background.
470. The Defendants submitted that whist there were some similarities between the logos, they were not similar overall, and that any similarity was counteracted by the very different names for the two businesses.
471. I do not find the similarities between the logos to be particularly striking - but I am satisfied that the rebrand in 2018 moved the Oh Polly logo closer to the House of CB logo. The Defendants were well aware of the Claimants and their logo, and so must have been aware that they were moving closer to it.
472. The Claimants relied on multiple similarities between the House of CB website and the Oh Polly website, set out in the Re-Amended Particulars of Claim. The two websites are shown here, in both desktop and mobile formats:
House of CB
Oh Polly
House of CB (mobile)
Oh Polly (mobile)
473. It would elongate this judgment still further to go into detail. It suffices for present purposes to note that there are many obvious similarities between the two websites, including:
i) Banners;
ii) Footers;
iii) Layout of the images of the garments; and
iv) Use of mottled pink as the background for studio photographs.
474. Two similarities struck me as particularly noteworthy. The first was the arrangement of garments by colour - the evidence before me was that this is not a usual way of arranging garments on an e-retail site, and so that similarity stands out. Second, the House of CB website directs garment returns to a returns “depot”. Oh Polly also adopted this expression. In this context, “depot” is an American English expression that struck me as an odd term to adopt for a Scottish website.
475. It is also clear from the documents before me that Ms Henderson paid close attention to the House of CB website. The following are excerpts from emails written by Ms Henderson in early 2015 in relation to the development of the Oh Polly website:
“I think the banner on the main page at the top also be a bit longer. Do you agree? Ours looks small and unnoticeable compared to that on (
www.houseofcb.com) say for example.”
“Thanks for the banners. Can we have another go at these … A really good example of banners is House of CB.
www.houseofcb.com - they look clean and crisp - the text is incorporated into the image which I think makes it a lot clearer. Their font is also strong and clear.”
“I’ll collect some Instagram/model style pics that I like the look of. I really like the celeb boutique outside shoot for Mariana Hewitt”.
476. I was also taken to five email exchanges in a similar vein between Dr Branney and those developing the Oh Polly website, each of which refers to House of CB. In one of them, Dr Branney writes “Have a look at House of Cb on mobile and see how they approach this and show how it would look on ours”.
477. The Defendants pointed to the differences between the various websites, including: the “My Account” feature; wish-list functions; cookie policy placement; search bar functionality; and the provision of size guide information alongside more recent differences such as the Claimant’s messenger function; multi-language function and the way that the Oh Polly website can select multiple colours whilst the House of CB website can filter by one colour only.
478. In my judgment, there are aspects of the Defendants’ internet presence that clearly emulate that of the Claimants, and intentionally so. I do not consider this to be a further coincidence. Whilst counsel for the Defendants valiantly pointed out how each individual feature appears in at least one other website (although some were outside the clothing sector), that does not appear to me to go far enough. In circumstances where the evidence showed the Defendants monitoring closely House of CB’s activities, including its online activities, and pointing out that it wished to emulate the House of CB website in certain ways, I cannot conclude that the various similarities are coincidental. Rather, they are the result of a course of conduct to create a look and feel similar to House of CB.
479. The Claimants relied on the following aspects of branding, marketing and strategy which they said were, in conjunction with other pleaded indicia, recognisable as the Claimants’:
i) House of CB used a single model for its whole collection (other than swimwear) until April 2016, using multiple models thereafter. The Defendants also used a single model until 2016, and multiple models thereafter. This was not an aspect of the Claimants’ case that was noted in social media: I do not consider it likely that it is something that consumers would notice sufficiently to contribute to the pleaded misrepresentation.
ii) The Claimants said that “unusually” they released new garments weekly, with special collections for events such as Christmas and St Valentine’s Day. The Defendants did the same. Again, whilst I accept that this was a similarity between the ways in which House of CB and Oh Polly garments were marketed, there was no evidence to support the notion that consumers had noticed, or identified that factor, even along with the other pleaded indicia, as indicative of House of CB.
iii) House of CB garments are designed in London. This was considered by Connor Walker to be important, and was highlighted to customers following the move from Celeb Boutique. The Defendants also adopted the phrase “Designed in London” in social media, which Dr Branney accepted as untrue, as Oh Polly garments were never designed in London. He had no sensible explanation for this widely publicised untruth. Whilst there were no social media posts specifically mentioning this similarity, I accept that it is something that consumers would be aware of (given the prominence both House of CB and Oh Polly gave it) and that it would be therefore something that consumers would take into consideration in their understanding of the two brands. The fact that consumers consider both brands to be designed in London would be seen as a similarity between them, to support the Claimants’ case that consumers consider Oh Polly to be a sister brand of House of CB.
iv) Both House of CB and Oh Polly used “flatlays” - presenting images of garments laid flat. The concept is widely used in the industry, but the Claimants’ case was that the Oh Polly flatlays were very close to the House of CB flatlays, and unlike those of other competitors. Examples are shown here:
House of CB Oh Polly
Again, I consider that Oh Polly adopted flatlays similar to those of House of CB in order to emulate the look of House of CB. Again, whilst no social media comment on the similarity was brought to my attention, given the importance of social media to both brands, I consider that this is something that consumers would take into account in considering whether Oh Polly is a sister brand to House of CB.
v) Both sides stage dinners and other events for influencers and bloggers, photographs of which are then used extensively on social media. This practice is not so unusual that the notion of holding dinners would be sufficient, in my judgment, even combined with the other indicia, to lead to the pleaded misrepresentation.
vi) Both sides emphasise the quality of their garments. Again, this is not unusual in garment vendors. I do not consider that this could contribute to the pleaded misrepresentation.
480. The Claimants submitted that the Defendants’ giving of names to their garments was another similarity. The evidence before the Court was clear that the Claimants’ garments have women’s names, whereas the Defendants’ garment are named with puns or innuendo. For example, C2 is an off the shoulder dress called Tiggy whereas D2 was called “Off the Richter”. It was accepted by the Claimants that none of the Defendants’ names evoke those of the Claimants.
481. It does not seem to me that the mere fact of naming garments assists the Claimants. The evidence of whether that was unusual or not at the Relevant Date was not clear, but I do not think it matters. Even if it was unusual to name garments at the time (and the evidence on this was unclear), it was not something that consumers referred to in social media posts, and I do not consider it would be something that would mislead consumers into thinking that the two brands are related, even in combination with the other indicia, when the names themselves are so different. I might have concluded otherwise had the names themselves been similar - but I consider the different names to be a differentiating factor, rather than a similarity.
482. House of CB does not sell jewellery. Oh Polly does. Counsel for the Defendants pointed to this as a difference. Whilst it is a difference between the two brands, it is not one that, in my judgment, would occur to a consumer as something that would mean two brands were unlikely to be sister brands.
483. There was evidence before me of Oh Polly garments being wrongly returned to House of CB, and vice versa. I can deal with this briefly, albeit that it occupied several witness statements and some time in cross-examination. The numbers of incorrect returns were tiny given the number of garments sold by each side. In some cases, the mistake was made by the courier company (which is not evidence of confusion/deception). In other cases, it was clear that the purchaser had purchased both House of CB and Oh Polly garments at the same time, and got in a muddle. The instances were so few in the scheme of things as not to assist me.
484. The Claimants also rely on the timings of the various steps they say the Defendants took to emulate House of CB. For example:
i) In almost all cases, Oh Polly used the same models after House of CB, in some cases only shortly after;
ii) The speed with which the Defendants copied the Claimants’ garments; and
iii) The use of the same or very similar locations, often shortly after they had been used by the Claimants.
485. I take this into account.
486. There was in the evidence and in some of the earlier submissions and questioning a suggestion that House of CB garments were targeted at a different market to Oh Polly garments. In his written closing submissions, counsel for the Defendant acknowledged that Oh Polly considers House of CB to be its biggest competitor, and that both brands sell garments to “fashion-conscious young women in the UK who are looking to buy sexy clothing online and, in particular, dresses for a night out”. He also acknowledged that their customer bases overlap.
487. This position is consistent with the evidence that consumers buy from both brands.
488. It was clear from the evidence that the Defendants closely monitored the House of CB website and social media, amongst those of other competitors. There is nothing wrong with that, of course. But the Claimants submitted that the Defendants went beyond monitoring, and set out to emulate aspects of the House of CB business, setting out to trade off the Claimants’ brand and goodwill.
489. Ms Henderson and Dr Branney both accepted under cross-examination that they viewed and admired the House of CB style of imagery. But both said, emphatically, that they did not set out to emulate House of CB.
490. I have some difficulty with this denial. The documents before me demonstrate both a clear course of conduct of reviewing many aspects of House of CB’s activities (amongst other competitors) as well as objective similarities between the activities of House of CB and Oh Polly that Dr Branney and Ms Henderson were unable to explain. I cannot accept, for example, that Oh Polly’s use of identical models, in similar poses, with similar hair and make-up, in identical locations is mere coincidence, particularly when no acceptable explanation was given.
491. I accept the evidence of other Oh Polly witnesses that they did not intend to copy House of CB garments or indicia. But they were not directing the business - they were all told what to do by Ms Henderson, sometimes through Dr Branney, but often in great detail. Ms Henderson was responsible for the creative side of the business - Dr Branney looked after the administrative side, and the abundant correspondence before me showed that he usually deferred to her on any creative issues. It is therefore Ms Henderson’s state of mind that matters for the purposes of the Claimants’ passing off case. I have found that her evidence is not to be accepted unless corroborated by other evidence, and the other evidence, some of which I have set out above, clearly shows the similarities of the Oh Polly business, adopted after spotting, and liking, them in House of CB. Therefore, in my judgment, the only conclusion open to me is that Ms Henderson set out to emulate House of CB, and directed Oh Polly staff to hire the same models, rent the same locations, adopt similar hair and makeup, adopt similar flatlays, follow the House of CB packaging from bright pink to softer pink, adopt a similar website etc. She herself copied some House of CB garments, as I have found above, albeit that she also copied garments of other designers as well.
492. On that basis, the authorities tell me that I should not strain every sinew to find that there has been no passing off: Ms Henderson knows her market, and if she set out to copy House of CB, it is likely that she achieved as much. But I accept counsel for the Defendants’ submission that what matters is the objective effect of the relevant activities: a finding of intention to emulate cannot rescue an objectively unmeritorious claim for passing off.
493. The Claimants submitted that the effect of the similarities I have set out above was obvious: deception. They did not argue for direct confusion, because they accepted that the names of the two brands are different. Rather, they argued that the deception will give rise to a mistaken belief that House of CB and Oh Polly share a trade connection, with the same entity behind each.
494. The Claimants further submitted that it is common in the fashion industry for different brands to share a trade connection - with so-called diffusion lines or sister brands being common. House of CB and Mistress Rocks are sister brands. So are Oh Polly and Bo+Tee, which sells sportswear.
495. If it were not deception prior to the rebrand, counsel for the Claimants argued that the Oh Polly brand certainly was deceptive following the May 2018 rebrand. She pointed to Dr Branney’s acceptance under cross-examination that the rebranded Oh Polly was “much more similar to the House of CB brand … than it was before” and his acceptance that the Oh Polly business was more similar to House of CB than to any other business selling women’s clothing at that time. Dr Branney also accepted that the similarities would be recognised by at least some House of CB customers who came across the rebranded Oh Polly business. The Defendants, through their counsel, also accepted that some of these similarities have in fact been pointed out by apparently unconnected individuals in social media posts.
496. The Defendants pointed out that House of CB and Oh Polly have been trading in the same market for five years, and so criticised the Claimants for not adducing evidence from any witness who has been deceived into thinking that there is a trade connection between Oh Polly and House of CB. Rather, the evidence relied on by the Claimants was classified by the Defendants into four categories:
i) Social media posts commenting generally on the two businesses and their garments;
ii) Direct customer inquiries to one or other business regarding a specific purchase or more generally;
iii) Miscellaneous other direct communications; and
iv) Wrongly returned items or items returned in the wrong packaging.
497. Counsel for the Defendants provided a detailed 14 page table of written submissions on each instance, which I do not need to repeat here, but which I have read. He also made the following general submissions:
i) The Claimants had done a thorough job, so these instances could not be said to be the tip of the iceberg;
ii) Even if the instances showed genuine confusion on the part of the relevant public, there was nothing to tie that confusion to the Business and Marketing Style and Get-Up;
iii) The fact that some people consider that Oh Polly has strayed too close to House of CB does not establish an actionable misrepresentation;
iv) There is a handful of cases where a social media user has casually wondered whether there is a connection between Oh Polly and House of CB: but this, he said, was merely wondering, not deception;
v) There is also a handful of instances where a customer, influencer or other social media user has made a mistake - but there is no indication that the mistaken person considers that Oh Polly and House of CB are sister brands; and
vi) There are no cases where it is clear that a connection between the two companies has been assumed - which is what, he submitted, the Claimants need to show.
498. I accept that many of the comments put forward are either not relevant to these proceedings (because for example, they were posted by some-one obviously not in the UK) or do not evidence actual confusion. I consider that the following posts come closest to what the Claimants need in order to be to demonstrate passing off:
“Out of curiosity, are you guys affiliated/connected with Oh Poly in any way? It baffles me that they steal every single design after you guys release collections. 
how do they get away with it!?”
“oh polly is owned by house of cb 
”
“Is Oh Polly and House of CB the same thing?”
“Are “ohpolly and “HouseOfCB the same company?”
“Is oh Polly the same as house of CB?!”
“Hey quick question…Is “Oh Polly” one of your companies? The reason I’m asking is because I’m a bit confused…I stumbled across their Instagram page and they seem to have similar models, similar sort of styles etc…”
“Do you also own “OhPolly”? they have a very similar style to House of CB”
“see a lot of similarities between you & “ohpolly” (models, style, insta) are you sister cos?”
“Hi. Is Oh Polly part of House of CB company? I’ve seen many of their design are very similar to yours”
“I even thought maybe House of CB bought them, that’s how similar they are lool, I’m not complaining though”
499. Of these, only the second one is a clear statement that House of CB and Oh Polly are related. The remainder, as counsel for the Defendants pointed out, evidence only that the writer has been “caused to wonder”. I therefore accept counsel for the Defendants’ submission that there is no evidence of a substantial number of consumers being deceived or misled into considering that Oh Polly is a sister brand of House of CB. As set out above, I consider that the evidence establishes that consumers noticed similarities between some of the indicia relied on by the Claimants - and hence I am satisfied that goodwill subsists in each relevant indicium including the designs, use of models, locations, GIFs and Instagram presentation as set out above. But I do not consider that the social media posts indicate that there is a level of deception in the market or demonstrate that Ms Henderson has achieved that which she set out to: the emulation of the House of CB brand. There is, simply put, no evidence, even after several years of the similarities being apparent to consumers, that any of those consumers considers (other than the one reference to which I referred) that House of CB and Oh Polly are sister brands. There is no substantial evidence that consumers consider there to be a trade connection.
500. Stepping back, on the basis of all the evidence before me, and having read and listened very carefully to counsel’s submissions, it does seem to me that the Claimants have proved their case in relation to goodwill in some of the indicia as set out above, but not that sufficient members of the relevant public have been deceived. I therefore find that the first element of Jif Lemon has been made out (goodwill), but not the second (misrepresentation). The passing off case therefore fails.
501. I reach this conclusion with some reluctance. It does seem to me that the Defendants have been able to ride on the coat-tails of the Claimants’ successful business model, and rather than investing in their own development, or spent money and time trialling different models, styles, locations, packaging etc, they have obtained an advantage by copying a successful competitor. But, as Jacob LJ has set out so clearly in L’Oréal v Bellure, passing off is not a tort of unfair competition. Passing off requires a misrepresentation, and there is insufficient evidence before me to find that consumers do or will consider House of CB and Oh Polly to be sister brands. That was the Claimants’ case, and it has not been made out.
502. As Mr Recorder Iain Purvis QC (sitting as a Deputy Judge of the Patents County Court) said in WS Foster & Sons Limited v Brooks Brothers UK Limited [2013] EWPCC 18 at paragraph 55:
“Although proof of damage is an essential requirement of passing off cases, it will generally be presumed where a misrepresentation leading to a likelihood of deception has been established, since such deception will be likely to lead to loss of sales and/or more general damage to the exclusivity of the Claimant's unregistered mark. Mr Aikens accepted that if there was a misrepresentation in the present case, then he had no separate case on damage. I hold that damage is inevitable, at least in the sense recognised in Sir Robert McAlpine v Alfred McAlpine [2004] RPC 36 at 49 (the 'blurring, diminishing or erosion' of the distinctiveness of the mark).”
503. The Defendants initially accepted that a finding of damage would follow from a finding of goodwill and misrepresentation, but resiled from that position at the start of the trial. Rather, counsel for the Defendants sought to rely on the absence of any reference to the Mistress Rocks sister brand on the House of CB website, submitting that if an association with House of CB were likely to encourage customers to buy garments from Mistress Rocks, the Claimants would have promoted Mistress Rocks on the House of CB website. I do not accept this submission. Passing off is concerned with protecting goodwill. I have found that the Claimants have goodwill in some of the pleaded indicia. Had I found that the Defendants’ activities amount to deception, it seems to me that it would have followed that that deception would be damaging to the Claimants’ goodwill, given that the Claimants and Defendants are competitors. In such circumstances, it would be easy to imagine consumers thus deceived who purchase garments from Oh Polly instead of House of CB, thinking they are buying from a sister brand. Similarly, it would be easy to imagine consumers dissatisfied with Oh Polly products or services who decide not to shop again with Oh Polly or any perceived sister brand, including House of CB.
504. The Claimants seek additional damages pursuant to Section 229(3) CDPA and/or Article 89(1)(d) of the Design Regulation in relation to UKUDR and CUDR infringement. I was not asked to assess the quantum of such damages - only to determine whether or not on the basis of the facts before me the Defendants should be liable to pay them. At the opening of the trial, the Defendants submitted that as additional damages are a question of quantum, not liability, they did not fall within the order of Deputy Master Nurse for what should be determined at this trial. I rejected that submission at the start of this trial for the reasons I gave then.
505. Section 229(3) of the CDPA provides:
“(3) The court may in an action for infringement of design right, having regard to all the circumstances and in particular to-
(a) the flagrancy of the infringement; and
(b) any benefit accruing to the defendant by reason of the infringement,
award such additional damages as the justice of the case may require.”
506. The law on additional damages has developed largely in the context of copyright infringement, but I see no reason not to apply that law to UKUDR infringement. In Nottingham Healthcare NHS Trust v New Group Newspapers Limited [2002] RPC 49, Pumfrey J (as he then was) noted at paragraphs 41 to 52 that “flagrancy” implies “scandalous conduct, deceit and suchlike; it includes deliberate and calculated infringement”. He considered that an attitude of carelessness sufficiently serious to amount to an attitude of “couldn’t care less” would be sufficient to found an award for additional damages.
507. Article 89(1)(d) of the Regulation provides:
“Where an action for infringement or for threatened infringement of a Community design court finds that the defendant has infringed or threatened to infringe a Community design, it shall, unless there are special reasons for not doing so, order the following measures:
…
(d) any order imposing other sanctions appropriate under the circumstance which are provided by the law of the Member state in which the acts of infringement or threatened infringement are committed …”
508. Counsel for the Defendants submitted that additional damages for CUDR infringement were not available on the terms of Article 89(1)(d) on the following basis. Article 89(1)(d) provides for “other sanctions … which are provided by the law of the member state”. The sanction relied on by the Claimants is section 229(3) CPDA, which only applies to UKUDR infringement, and is therefore not applicable to UCDR. Neat though that pleading point is, in my judgment, it is wrong. English law provides for additional damages in unregistered design rights cases, and so they are available here.
509. I have found that the Defendants have infringed the Claimants’ UKUDR and CUDR in relation to 7 garments. Whilst Ms Henderson, the named designer of the relevant garments, has denied copying, I have held her not to be an honest witness, and I have discounted her evidence unless it is corroborated by other evidence. What is clear from the written record is that for the Defendants’ garments D2, D4, D12, D13, D35, D61 and D91, Ms Henderson took an image of the Claimants’ garments, and sent it to the factory to be reproduced. In the circumstances of this case, that is sufficient to warrant an award of additional damages in relation to those designs. It was at least an attitude of “couldn’t care less” about the rights of others: a designer, as Ms Henderson claimed to be, of her years of experience ought to have known (if she did not in fact know) that such activity was not lawful.
510. In reaching that conclusion, I do not need to rely on the Claimants’ submissions in relation to the alleged scale of infringement, the alleged concerted efforts to emulate the House of CB business or the allegedly misleading positions taken by the Defendants in this litigation. I do, however, take into account the Claimants’ final submission on this point, which was that the Defendants have been on notice of the Claimants’ concerns since 2016. Rather than choose to stop “referencing” the Claimants’ designs, the Defendants continued, even after these proceedings were launched.
511. In summary:
i) None of the Claimants’ designs C2, C3, C4, C7, C9, C12, C13, C17, C21, C35, C47, C49, C61, C63, C66, C77, C81, C91, C93 and C102 lacks originality in the copyright sense or is commonplace over the prior designs which were before me: the UKUDR claimed in these designs are therefore not invalid;
ii) None of the Claimants’ designs C2, C3, C4, C7, C9, C12, C13, C17, C21, C35, C47, C49, C61, C63, C66, C77, C81, C91, C93 and C102 produces the same overall impression on the informed user as the prior designs which were before me: the CUDR claimed in these designs are therefore not invalid;
iii) In creating designs D2, D4, D12, D13, D35, D61 and D91, Ms Henderson copied the Claimants’ designs C2, C4, C12, C13, C35, C61 and C91;
iv) Designs D2, D4, D12, D13, D35, D61 and D91 are made exactly or substantially to the designs of C2, C4, C12, C13, C35, C61 and C91 and therefore infringe the Claimants’ UKUDR;
v) Designs D2, D4 (in cream), D12, D13, D35, D61 (in brown but not in white), D91 (in black but not in other colourways) produce on the informed user the same overall impression as C2, C4, C12, C13, C35, C61 and C91 and therefore infringe the Claimants’ CUDR;
vi) The Claimants are entitled to additional damages as a result of the flagrancy of the Defendants’ infringement; and
vii) The Defendants have not passed off their Oh Polly business as a sister brand of the House of CB business.
512. I will hear the parties’ representatives on the form of order to be made following this judgment.
513. As noted above, the Claimants’ claim of UKUDR and CUDR infringement in relation to a further 71 garments was stayed as part of the streamlining ordered by Deputy Master Nurse. It is my fervent hope that there are sufficient findings in this judgment to enable the parties to agree how the claims in relation to those 71 garments should be concluded. As became apparent during the course of the trial, the 20 garments in issue before me fell into three “buckets”: (1) where referencing was admitted, and the Claimants’ garment was uploaded to Trello at the beginning of the production process; (2) where referencing was admitted, but the Claimants’ garment was not uploaded to Trello until part way into the design process; and (3) where no referencing was admitted. I add to that garment D35, where copying was admitted. Whilst there were some minor variations within those buckets leading to slightly different outcomes (usually to do with colourways), I have found that, on the whole, the garments in bucket (1) infringe, but that those in buckets (2) and (3) do not. It seems to me that that ought to provide good guidance for dealing with the remaining 71 designs without the need for further intervention by the Court. Whilst not deciding the point, it seems to me at this stage that a further trial on liability in relation to those designs would likely be disproportionate. If those 71 garments cannot be resolved between the parties, the matter should be listed before me for directions.
