England and Wales Court of Appeal (Civil Division) Decisions

IN THE COURT OF APPEAL (CIVIL DIVISION)
ON APPEAL FROM THE HIGH COURT OF JUSTICE
BUSINESS AND PROPERTY COURTS OF ENGLAND AND WALES
INTELLECTUAL PROPERTY ENTERPRISE COURT
HIS HONOUR JUDGE HACON

[2017] EWHC 2927 (IPEC)

		Royal Courts of Justice Strand, London, WC2A 2LL
		09/10/2019

Lord Justice Floyd:

1. This is the judgment of the court to which all its members have contributed.

Introduction

2. This appeal concerns a dispute over the authorship of the screenplay of a film about the latter part of the life of Florence Foster Jenkins ("FFJ"). FFJ was a wealthy American socialite who loved music. She considered herself to be a talented operatic singer, but she was an awful one. The first and second claimants and respondents, Nicholas Martin and his company Big Hat Productions Limited, brought these proceedings against the defendant and appellant, Julia Kogan, for a declaration that Mr Martin was the sole author of, and sole owner of the copyright in, the screenplay. Ms Kogan, in response, sought a declaration that she was a joint author with Mr Martin, and thus a joint owner of the copyright. She claimed that Mr Martin and the second respondent had infringed the copyright in the screenplay by exploiting it without her participation, and claimed a share of the royalties received by them. By Part 20 proceedings, Ms Kogan also sought relief for infringement of copyright against the third to fifth respondents who are various companies involved in the production and financing of the film. I will refer to the third to fifth respondents as "the film companies".
3. After a two-day trial in the Intellectual Property Enterprise Court ("IPEC"), during which each of the rival protagonists was cross-examined for half of the first day, HHJ Hacon ("the judge") declared Mr Martin to be the sole author and owner of the copyright in the screenplay. The judge also held that Ms Kogan was estopped from asserting any right she might have had in the copyright against the film companies so as to interfere with the public performance of the film, because she had stood by in the knowledge that the film was being produced and released, and thus represented that she would not seek to prevent these activities.
4. Ms Kogan appeals, with permission granted by Henderson LJ. On the appeal, Ms Kogan was represented by Simon Malynicz QC, Lionel Bently and Ashton Chantrielle. Mr Martin and his company were represented by Tom Weisselberg QC, and the film companies were represented by Jonathan Hill. We are grateful to all counsel for the skilful and economical presentation of their cases in the written materials and in their oral submissions.

The facts in outline

5. Mr Martin was and is a professional writer of film and television scripts. The judge's findings do not provide a picture of his career up to the point at which he wrote the screenplay for the film, but his first witness statement provided an outline of it. He began his professional career in 1991 but, in his words, by 2000 he had not had a big television hit and none of the feature film scripts he had written had been produced. Between 2000 and 2006 no show which he had written was produced, a period of his career which he described as "testing". In that period he had written, amongst other things, a series of scripts entitled The Music Room, which concerned a corrupt policeman who fell in love with a music teacher, and also worked on a script for a feature film, The Angel of Ferrara, a historical thriller set during the Italian renaissance. In 2007 he wrote episodes of the television police drama The Bill and from 2008 to 2011 episodes of the television detective series Midsomer Murders. By 2012 he had written 55 television scripts of which 32 had been produced. He had also written 12 short films, three plays and nine feature films. As to his enjoyment of music, he described himself as a Classic FM rather than a Radio 3 man, knowing what he liked. The film Florence Foster Jenkins was his first feature film script to be produced. It was a success.
6. The judge described Ms Kogan as "principally a professional opera singer", but again did not refer in any detail to other aspects of her career which were covered in her evidence. She had attended Ohio State University where she was an English Literature and Vocal Performance major. She had authored three children's books while in her twenties. From 2006 she began creating her own theatre/music projects, the first of which was Troika (2006-2011), an album on which 8 composers set poetry of her choosing to music. The Lad (2008-2013) was an opera ballet which she claimed to have conceptualised. She says that she adapted and translated the libretto, as well as structured scenes. In 2010 she began researching lost and forgotten personal songs written by exiled composers in Hollywood for a recording project.
7. Mr Martin and Ms Kogan met in late September or early October 2011, and began a romantic relationship. According to Mr Martin, he was intrigued by Ms Kogan's life as an opera singer and Ms Kogan was fascinated by his work as a writer. It appears that they lived together from February 2012 to March 2014 when Ms Kogan moved out of Mr Martin's flat. On Ms Kogan's account, the relationship continued until October 2014, although the couple remained on reasonably good terms until March 2015.
8. It was common ground before the judge that it was through Ms Kogan that Mr Martin was introduced to the story of FFJ, although the judge made no findings as to the circumstances in which this occurred. Ms Kogan claimed in her witness statement to have had a long-standing interest in FFJ, having first heard about her in 1991 whilst at university. She said that shortly before she met Mr Martin she had been asked by an actor/pianist about the possibility of performing in a Parisian production of "Souvenir", a fantasy play based on the life of FFJ. Ms Kogan's account was that she had shown Mr Martin some YouTube clips of FFJ's singing. She said that she told Mr Martin that a film script about her should be written as soon as possible, as she was amazed that no one had made a film about FFJ. Her evidence was that she "talked Nick into working on FFJ despite his misgivings". Mr Martin denied this account and said he had come home to find Ms Kogan in the spare bedroom in his flat listening to a YouTube clip of FFJ, and that Ms Kogan had explained who FFJ was, i.e. that she was a wealthy woman who had booked Carnegie Hall and sung to a sell-out crowd. He said that over the next few days he found himself returning to FFJ's recordings and had looked at a Wikipedia article about her. It slowly began to occur to him, he said, that FFJ might make a good subject for a film. He told Ms Kogan about this and she agreed. It was his case, therefore, that Ms Kogan had no input into the initial idea of writing a screenplay for a film about FFJ. The judge did not resolve this conflict of evidence.
9. According to Mr Martin's Particulars of Claim, shortly after his initial encounter with FFJ, he and Ms Kogan together watched a documentary by Donald Collup about FFJ and her life. It is common ground that the Collup documentary proved an important source of factual information for the screenplay.
10. At all events, the project of writing a screenplay commenced ("the Florence project"). A draft outline treatment was produced, followed by a series of draft screenplays. The first draft was written between February and April 2013; the second draft was completed on 19 June 2013. In July 2013 Mr Martin travelled to Los Angeles where he met various people in the film industry with whom he reviewed the Florence project. These included Ben Lewin, a director, producer and writer of screenplays, and his wife Judi Levine. On his return to London work was done on the third draft. Mr Martin was concerned that Mr Lewin and Ms Levine would claim a share of the authorship, making it difficult for him to sell the film to finance companies, so he decided to offer them a share of his writing income. A figure of 15% was later agreed.
11. Ms Kogan alleged in her defence that on a train journey on 17 February 2014 Mr Martin asked her what proportion of the screenplay she thought she owned, and that Mr Martin had volunteered that it was 15%. The judge did not make any finding about this, however.
12. On 5 March 2014 the second claimant signed an agreement with the third Part 20 defendant for the financing of the film. The agreement contained a warranty that Mr Martin was the sole author of the script.
13. On 8 April 2014, following a therapy session which Ms Kogan and Mr Martin had attended in an attempt to save their personal relationship, Ms Kogan raised with Mr Martin the question of whether she should be given a share of the income.
14. There were many drafts of the screenplay. Ms Kogan accepted that her involvement in draft four and subsequent drafts was not significant as she had by then moved out of Mr Martin's flat.
15. The film "Florence Foster Jenkins" was released in 2016. The film starred Meryl Streep and Hugh Grant and was directed by Stephen Frears. All members of the court were invited to watch the film prior to the hearing, and we did so.

IPEC and the case management of the issues

16. Because the argument on one of the grounds of appeal concerns the procedural history of the case and the way in which the issues were case managed, we need to set those matters out in more detail than would otherwise be necessary. The IPEC is a court with a special jurisdiction for intellectual property cases, the successor to the Patents County Court. The rationale for the creation of the IPEC and its predecessor was to provide access to justice for individuals, and small and medium sized enterprises, who would not be able to pay the costs normally associated with intellectual property litigation in the High Court under the CPR. The special rules which govern the IPEC seek to achieve this objective by providing for a highly focused, intensively case-managed procedure. Two significant differences from the general procedure under the CPR should be noted. First, the rule about statements of case in the IPEC requires a party to plead "all the facts and arguments" on which he relies, in contrast to pleading only material facts: see CPR 63.20. Secondly, the default position is that, unless the judge orders them at the case management conference ("CMC"), there are to be no witness statements, experts' reports or cross-examination at trial. Only in exceptional circumstances will the court allow additional matters to be relied on which are not provided for by the order made at the CMC.
17. At the CMC in the present case the judge had pleadings from both Mr Martin and Ms Kogan to consider. Mr Martin's Particulars of Claim, dated 6 April 2016, identified ways in which he accepted that Ms Kogan had contributed to the Florence project, but contended that these were not sufficient to make her a joint author. Thus, he accepted that Ms Kogan provided factual information and observations about music, and especially opera, as well as about the lifestyle, vocabulary, idioms and idiosyncrasies of professional musicians. He also accepted that Ms Kogan had acted as a critic on drafts of the screenplay, and provided some editorial comments, suggestions and corrections on particular scenes. She had contributed the observation that FFJ was interested in "loveliness" in music and that idea had informed Mr Martin's writing of the script. He accepted that Ms Kogan's comments and suggestions occasionally extended to ideas for short lines of dialogue, which were on a very few occasions adapted by him in the script, although never accepted verbatim. Mr Martin accepted that Ms Kogan suggested some ideas for minor characters and minor dramatic events. He said in his pleading that he could not recall whether Ms Kogan suggested the idea of using Lily Pons, a famous contemporary opera singer, to appear in the screenplay, but explained that such a suggestion, if made, was of no relevance.
18. In paragraph 17 of his Particulars of Claim Mr Martin said:

"There were approximately 23 revisions to the shooting script and approximately five drafts of the pick up scenes (a pick up scene being a relatively minor recording that is shot after principal filming to augment existing footage). The final version of the Screenplay is annexed hereto."

19. Ms Kogan's Defence and Counterclaim was dated 9 June 2016. She alleged that she had introduced the concept of the screenplay to Mr Martin, and that thereafter she had persuaded Mr Martin to embark on the Florence project. She asserted that at all times when she and Mr Martin were living together, that is to say during the drafting of the first, second and third drafts, the parties were working closely together. In her pleading she defined "the Screenplay" as including references to the first, second and third drafts. She admitted and averred that she had made the contributions identified by Mr Martin in his pleading, but asserted that her contribution went further. She sought a declaration that she was a joint author of "the Screenplay". Clearly, in view of the definition of Screenplay, this declaration would include the first, second and third drafts. She sought an injunction against the claimants and the film companies from infringing her right to be identified as author, but not a general injunction against infringement of copyright. In addition, she sought an enquiry as to damages or an account of profits against the claimants and the film companies.
20. At the CMC, held on 14 December 2016, the judge identified the issues in a Schedule to his order as being:

"1. Whether the Defendant/Part 20 Claimant is a joint author of the final version of the Screenplay, as referred to in paragraph 17 of the Particulars of Claim.

2, Whether the Defendant/Part 20 Claimant is estopped or prevented by acquiescence, agreement or waiver from asserting or relying on any rights of sole or joint authorship."

21. There was discussion before the judge at the CMC as to whether Ms Kogan was asserting joint authorship of each individual draft, or only the final version. Counsel for the film companies submitted that the only thing which mattered was the final script from which the film was made, but he accepted that the earlier scripts might be relevant "as evidence of the contribution to the final screenplay". Counsel for Ms Kogan queried this approach. She pointed out that it had been pleaded that Ms Kogan had been involved up to a certain stage. If there were changes subsequent to that, "it would be a question of whether it is still a substantial part, whether they use a substantial part of that final work that she was involved in." The judge then said:

"What matters … is that there is a final document which effectively is the script that was used, and therefore what matters, ultimately, is whether Ms Kogan contributed sufficiently to that final script. Now, if she only went up to Version 6, then it will need to be shown that, for example, her contribution to Version 6 found itself into the final document such as to be sufficient to contribute. We are not talking about direct contribution to the final script. … It could be indirect. It is copyright".

22. It was thus clear to all involved that this was not a case where it was necessary to analyse the copyright work on a draft-by-draft basis. Ms Kogan's case would not fail if she had made no contribution to the specific work which went into converting draft six into the final version of the screenplay. If Ms Kogan had made sufficient relevant contributions up to the third, or the sixth draft, and these carried through into the film, she would both be in a position to establish her interest in the copyright and her allegations of infringement. As will be seen, the judge did not adopt this approach in his final judgment, acceding to an argument advanced on behalf of Mr Martin, without prior warning, that all that mattered was the contribution which went into the final draft, treating the final draft as a separate copyright work from its predecessors.
23. With permission from the judge granted at the CMC, each side served a lengthy witness statement explaining their respective versions of how the screenplay came to be written. Mr Martin's statement explained his approach to writing a script. His case was that he wrote the screenplay on his own. Ms Kogan was not a screenwriter, and he had taken objection to a suggestion that she had made at one point that they co-write another script together. At [99]-[100] he set out the ways in which he accepted that Ms Kogan had contributed to the first treatment phase:

"Julia's involvement in the first treatment phase was in four main ways:

a) Encouragement and support;

b) Sounding board/criticism;

c) Editing – spelling corrections and typos; and

d) Information on the New York music scene of the time in particular that Caruzo [sic] had died; that Homophone was not a major record label; that the Met would not have held a luncheon in Florence's honour; and that (in relation to the tag line) Florence had not actually sold out at Carnegie Hall.

Julia also observed that Florence's musical taste was for "loveliness" and this led me to the idea, in scene six, that McMoon gets the job as her pianist because he plays gentle music. She also noted Florence's slight pause before hitting the high notes. This suggested she had some doubts about her ability, but it didn't stop her from having a go. I thought this amusing and helpful."

24. Mr Martin went on to assert at [101]:

"At no time did Julia suggest scenes, characters or sequences nor did she contribute in any way to the construction of the film's structure… I wrote every word and made every decision about what should and should not be included, discarded or changed."

25. Then, at [132], in relation to the subsequent draft scripts, Mr Martin says again that Ms Kogan's role was to proof read, to provide historical background about the musical world in which Florence moved, about opera and about New York society at the time, to provide many suggestions for and about the music, to provide a very limited number of observations about scenes, the vast majority of which were rejected by him or later cut, to act as a sounding board and provide an invaluable source of encouragement and support.
26. Ms Kogan's account could not have been more different. She said that she and Mr Martin had several categories of working method on their various collaborative projects. Sometimes they would literally write together, with Ms Kogan doing the typing. At other times Mr Martin would write a scene or two first, and then show this to Ms Kogan for discussion and revision. This was how they worked when they were apart but also when they were at home. At [42] – [43] she said:

"Our work and our relationship were inextricably linked, and we explored our own relationship through characters, plots and storylines we were creating and discussing. Nick and I bounced ideas off each other with such speed, that it was impossible to know who came up with what specifically after the fact. Our ideas came thick and fast, snowballed and became inseparable. This process was at the heart of the FFJ screenplay.

It is largely impossible to separate out our distinct contributions to the FFJ screenplay, except in scenes with musical content, which are mostly my words, and the scenes that were entirely written without me, which are Nick's with input from others. I address this in as much detail as possible in Annex A to this witness statement."

27. Annex A was a scene-by-scene description of the film in which Ms Kogan explained the way in which she and Mr Martin had, she said, collaborated. In it, she frankly accepted that there were scenes developed by Mr Martin with no input from her, but made clear where scenes, or parts of them, were derived from their joint discussions.
28. Ms Kogan's witness statement ends at [177]:

"In conclusion, I wish to make it clear that I made a significant contribution to many aspects of the FFJ screenplay, including the original idea, the characters, the story and the dialogue. Though Nick was the main writer and contributed considerably more than I did, the process of writing the FFJ screenplay was a creative collaboration in which I participated as a partner. As a result of the way in which our ideas are intertwined and were built upon by both of us, our respective contributions in most of the script are inseparable."

29. That is how matters stood when the case came to trial. There was a clear dispute on the evidence as to whether the Florence project was a product of the collaboration of the two parties, and, if it was, as to the nature of that collaboration. If it was a collaborative project, then, although there was some common ground as to what Ms Kogan had contributed, there was a further large area of controversy as to what that contribution was.

The law on joint authorship of copyright works

30. Section 10 (1) of the Copyright, Designs and Patents Act 1988 ("the Act") governs works of joint authorship. It provides:

"10(1) In this Part a "work of joint authorship" means a work produced by the collaboration of two or more authors in which the contribution of each author is not distinct from that of the other author or authors."

31. The four elements of joint authorship are thus (a) collaboration, (b) authorship, (c) contribution and (d) non-distinctness of contribution. Although joint authorship is ultimately a unitary concept, it is helpful for the purposes of analysis to consider each element in turn.
32. The first element is collaboration. In Beckingham v Hodgens [2003] EWCA Civ 143, this court approved the statement of Keating J (with whom Byles and Montague Smith JJ agreed) in Levy v Rutley (1871) L.R. 6 C.P. 523 as a correct statement of the modern law:

"So, if two persons undertake jointly to write a play, agreeing in the general outline and design, and sharing the labour of working it out, each would be contributing to the whole production, and they might be said to be joint authors of it. But, to constitute joint authorship, there must be a common design".

33. A collaborative work is different from a derivative work. If an author takes as his or her starting point the work of another and decides to adapt, embellish or improve on it, the result may attract copyright, but it will not be a work of joint authorship. The second author may be entitled to copyright in the resultant work, but it will be a work of sole authorship, the originality of which could only lie in the contribution made by the second author. More pertinently, there will be cases where a person makes a contribution prior to the completion of the work being undertaken by an author, but the work will remain a work of sole authorship because the contribution was not made in the course of a collaboration pursuant to a common design. The work must be created by collaboration. Friends of the author who make suggestions otherwise than in the course of a collaboration will not become joint authors. A striking illustration is the contribution made by Ezra Pound to The Waste Land. When T.S. Eliot showed him the original draft, Pound proposed very extensive deletions and revisions, with considerable consequences for the poem as published (indeed he became its dedicatee). But neither poet ever considered it a work of joint authorship, and it has not been regarded as such in the century since it was published, even though Pound's contribution is now widely known. This is because he was acting as a friend and critic and not a collaborator in a common design. By contrast, a collaborator may become a joint author after an apparently lesser contribution than Pound's. Take, for example, Beckingham v Hodgens (cited above), where a successful claim for joint authorship was made by a session musician who contributed a distinctive four-bar riff to a pop song that had been substantially conceived before he was hired. These examples show the importance of identifying the true nature of the interaction between the parties in relation to the work, an assessment that provides the essential context for consideration of questions of authorship and contribution.
34. The bald statement that the joint authors must collaborate in the creation of the work requires some further explanation of the nature of a copyright work. Copyright, of course, does not subsist in mere ideas, but in their expression. That proposition is a well-established one in English law, and is now also to be found in a number of international treaties and EU Directives: see Article 9(2) of the Agreement on Trade-Related Aspects of Intellectual Property Rights ("TRIPs") and Article 2 of the WIPO Copyright Treaty. The ideas/expression dichotomy has, however, been described as "notoriously slippery": see Simone: Copyright and Collective Authorship Cambridge University Press 2019, at page 41. A mere idea, stripped of any context, is of course not the subject of copyright. Jacob J put it in this way in Ibcos Computers Limited v Barclays Finance Limited [1994] FSR 269 at 291:

"The true position is that where an "idea" is sufficiently general, then even if an original work embodies it, the mere taking of that idea will not infringe. But if the "idea" is detailed, then there may be infringement. It is a question of degree. The same applies whether the work is functional or not, and whether visual or literary. In the latter field the taking of a plot (i.e. the "idea") of a novel or play can certainly infringe – if that plot is a substantial part of the copyright work. As Judge Learned Hand said (speaking of the distinction between "idea" and "expression"): "Nobody has been able to fix that boundary and nobody ever can"."

35. If copyright protection can extend to the plot of a literary work, even where the precise words of the work are not taken, then it seems to us to be logical to suppose that the skill which goes into devising the plot is properly to be regarded as part of creating the work. A collaborative work may thus come into existence if, in the context of a particular joint project, one person decides on the plot and the other writes the words to give effect to the plot. There will be a panoply of other ways of working as well, for example the labour of writing may be shared, or the labour of working out the plot, scenes and characters may be shared. In this connection, in Brighton v Jones, [2005] FSR 16, Park J, after having rejected the allegation that Miss Brighton had played a part in the actual writing of the dialogue of a play, continued at [56(ii)]:

"The point made … above concerns the actual words used, and it is not in itself decisive. Copyright can subsist in a story or the plot, so that if what happened in rehearsals was that Miss Brighton determined what the plot of the play was to be (or Miss Brighton and Miss Jones determined in collaboration what it was to be), and then Miss Jones actually wrote the words to give effect to the plot, I can see that Miss Brighton might have been a joint author". (emphasis supplied).

It follows that, in deciding whether there is a collaboration, it can never be enough simply to ask who did the writing.

36. The second element to consider is that of authorship. Section 9(1) of the Act defines "the author" in relation to a work, as "the person who creates it". Section 10(3) provides that "references … to the author of a work shall, except as otherwise provided, be construed in relation to a work of joint authorship as references to all the authors of the work." Putting the definitions together, a work of joint authorship is a work produced by the collaboration of all the persons who created that work.
37. Here again it is wrong to focus exclusively on the person who does the writing. There is a further important distinction to bear in mind, which is that between the making of the work and its reduction to material form (or fixation). The work may exist before its reduction to material form. The distinction is reflected in section 3(2) of the Act, which provides:

"Copyright does not subsist in a literary, dramatic or musical work unless and until it is recorded, in writing or otherwise; and references in this Part to the time at which such a work is made are to the time at which it is so recorded."

38. Thus, it has been held that the person who wields the pen is not necessarily the sole author of the work. In Cala Homes (South) Limited v Alfred McAlpine Homes East Limited [1995] FSR 818, a director gave detailed instructions to an architect as to the design of a house which the architect then drew. Laddie J, a judge with prolific expertise in the law of copyright, held that the director was a joint owner of the copyright with the architect. He said at page 835:

"In my view, to have regard merely to who pushed the pen is too narrow a view of authorship. What is protected by copyright in a drawing or a literary work is more than just the skill of making marks on paper or some other medium. It is both the words or lines and the skill and effort involved in creating, selecting or gathering together the detailed concepts, data or emotions which those words or lines have fixed in some tangible form which is protected. It is wrong to think that only the person who carries out the mechanical act of fixation is an author. There may well be skill and expertise in drawing clearly and well but that does not mean that it is only that skill and expertise which is relevant. As Mr Howe has said, where two or more people collaborate in the creation of a work and each contributes a significant part of the skill and labour protected by the copyright, then they are joint authors." (emphasis supplied)

39. In an earlier passage in the judgment Laddie J had cited a passage from Laddie, Prescott & Vitoria, The Modern Law of Copyright (of which he was an author) which suggested that a joint author must have "contributed a significant amount of the skill and labour which entitled the work to copyright protection and must further be an "author" in the sense of accepting responsibility for the contents." Laddie J affirmed the first part of that statement but held that it had been correctly conceded that the second part - that the joint author must take responsibility for the contents - went too far.
40. In Robin Ray v Classic FM plc [1998] FSR 622, Lightman J took a limited view of the effect of this part of Laddie J's judgment in Cala Homes by suggesting that, on the facts of Cala Homes, the architect was acting as the scribe for the director. He suggested that such a situation would be exceptional. A joint author must "participate in the writing and share responsibility for the form of expression in the literary work".
41. We think that the approach of Laddie J is to be preferred. It is the skill and effort involved in creating, selecting or gathering together the detailed concepts or emotions which the words have fixed in writing which is protected in the case of a literary or dramatic work, whether the work is one of sole or joint authorship. Too much focus on who pushed the pen is likely to detract attention from what it is that is protected, and thus from who the authors are. Likewise, we do not think that an enquiry into who accepts responsibility for the form of expression in the work is a helpful concept, particularly in a collaborative work. The statutory test does not go any further than asking who contributed to the creation of the work.
42. The third element relates to the contribution. The contribution made by a joint author must be an "authorial" one. As Laddie J put it succinctly in Fylde Microsystems Ltd v Key Radio Systems Ltd [1998] FSR 449 at 455, the section "does not turn someone who is not an author into an author". In that case the person who laboured extensively in setting the specification for computer software, and in fixing errors and bugs in it, was not a joint author with the person who actually wrote the software. What counts as an authorial contribution may vary depending on the type of work, however. A screenplay is concerned with providing a plot and dialogue which enable the reader to visualise what is going to happen on the screen, and get a feeling for what the audience will experience. The author is directly concerned with such matters as the choice of characters and the incidents in the film which will bring out those characters, and display their emotions. The storyline and plot are also of direct concern to the author and are a part of what he or she creates. These aspects of a screenplay can, in principle, amount to a contribution of an authorial kind.
43. A further question is how much of a contribution to the creation of a work must be made before the contributor to a collaborative work becomes a joint author. At trial there appeared to be consensus that the tests for (a) what was a sufficient exercise of skill to attract copyright in an original work, (b) what was a sufficient part to be taken from the work to amount to infringement, and (c) what was a sufficient contribution to justify joint authorship were all aligned. By analogy with the decision of the Court of Justice of the European Union (CJEU) in Case C-5/08 Infopaq International A/S v Danske Dagblades Forening [2009] ECR I-6569; [2009] ECDR 16; [2010] FSR 20 ("Infopaq"), all three questions were tested by asking whether there were elements which expressed the author's own intellectual creation. Although there was some debate about how demanding the Infopaq test was, there was no attempt by either side on this appeal to upset the measure of agreement which existed before the judge.
44. In SAS Institute Inc v World Programming Ltd [2013] EWCA Civ 1482; [2015] E.C.D.R. 17 at [29] to [37] this court reviewed the jurisprudence of the CJEU on "intellectual creation". As Lewison LJ (with whom Tomlinson and Vos LJJ agreed) explained:

"The essence of the term is that the person in question has exercised expressive and creative choices in producing the work. The more restricted the choices, the less likely it is that the product will be the intellectual creation (or the expression of the intellectual creation) of the person who produced it."

45. Case C-145/10 Painer v Standard Verlags GmbH [2012] ECDR 6 concerned copyright in photographs. The CJEU gave some helpful guidance on the sort of creative choices with which copyright is concerned:

"88. As stated in recital 17 in the preamble to Directive 93/98, an intellectual creation is an author's own if it reflects the author's personality.

89 That is the case if the author was able to express his creative abilities in the production of the work by making free and creative choices (see, a contrario, Joined Cases C-403/08 and C-429/08 Football Association Premier League and Others [2011] ECR I-0000, paragraph 98).

90 As regards a portrait photograph, the photographer can make free and creative choices in several ways and at various points in its production.

91 In the preparation phase, the photographer can choose the background, the subject's pose and the lighting. When taking a portrait photograph, he can choose the framing, the angle of view and the atmosphere created. Finally, when selecting the snapshot, the photographer may choose from a variety of developing techniques the one he wishes to adopt or, where appropriate, use computer software.

92 By making those various choices, the author of a portrait photograph can stamp the work created with his 'personal touch'.

93 Consequently, as regards a portrait photograph, the freedom available to the author to exercise his creative abilities will not necessarily be minor or even non-existent.

94 In view of the foregoing, a portrait photograph can, under Article 6 of Directive 93/98, be protected by copyright if, which it is for the national court to determine in each case, such photograph is an intellectual creation of the author reflecting his personality and expressing his free and creative choices in the production of that photograph."

46. These principles are also illustrated by a recent decision of the CJEU. In Case C-479-17 Funke Medien NRW GmbH v Bundesrepublik Deutschland an issue was whether military status reports, called UdPs, could qualify as copyright works. Funke Medien contended that the structure of the UdPs, which consisted of a standard form filled in by various authors with purely factual material, could not so qualify, whereas the German government argued the contrary. The court held that it was for the national court to determine:

"…whether, in drawing up those reports, the author was able to make free and expressive choices capable of conveying to the reader the originality of the subject matter at issue, the originality of which arises from the choice, sequence and combination of the words by which the author expressed his or her creativity in an original manner and achieved a result which is an intellectual creation…. whereas the mere intellectual effort and skill of creating those reports are not relevant in that regard…"

47. The fourth element, namely that "the contribution of each author is not distinct from that of the other author", did not give rise to a separate debate in this case. For some reason, the word "distinct" in section 10(1) was used instead of the word "separate" which appeared in section 11(3) of the Copyright Act 1956. The point of the qualification may be that if the contribution of each author is distinct from the other, there is no need for joint authorship. Each author can rely on such copyright as may subsist in his or her distinct part. Where their respective contributions are fused into a whole, this is no longer possible.
48. In Beckingham v Hodgens (cited above), this court also rejected a contention that in addition to establishing (1) that the joint authors collaborated in the creation of the work, (2) that the party contending that he was a joint author had made a significant contribution to it, and (3) that his contribution was not "separate" within the meaning of the subsection in the 1956 Act, there was a fourth requirement, namely a joint intention (i.e. a subjective joint intention) to create a joint work. The basis for this contention was the decision of Mr Justice Cohen in the Supreme Court of British Columbia in Darryl Neudorf v Network Expressions [1999] RPC 935. After an extensive review of authority in Canada, the United States and England, Cohen J said at pp 962-3:

"In the result I find that the test for joint authorship that should be applied to the facts in the instant case is as follows:

i) Did the plaintiff contribute significant original expression to the songs? If yes,

ii) Did each of the plaintiff and Ms McLachlan intend that their contributions be merged into a unitary whole? If yes,

(iii) Did each of the [joint authors] intend the other to be a joint author of the song?"

49. As the Court of Appeal pointed out in Beckingham, the last of the three elements was policy driven, and was arrived at despite and not because of the language of the US statute. Jonathan Parker LJ (with whom Ward and Laws LJJ agreed) said at [53] that it was right to confine consideration to the language of the Copyright Act and not to look beyond the section into the uncertain realms of policy.
50. A further consideration, which does not fit neatly into any of the four elements considered thus far, is whether, in a case where two people contribute to the creation of a work, but one has the final say on what goes in and what stays out, the other can be a joint author. In Brighton v Jones (cited above) the question was whether Miss Brighton's suggestions for changes to a script of a play could make her a joint author, when Miss Jones had the final say. Park J regarded the fact that the main author could decide whether to accept or reject the changes as one of several reasons for rejecting the claim to joint authorship: see [56(iv)]. On the other hand, in Bamgboye v Reed [2002] EWHC 2922 (QB); [2004] EMLR 5, Hazel Williamson QC, sitting as a Deputy High Court Judge, held that the fact that one joint author ultimately selected what went in was relevant only to the quantum of the contribution (because the selection itself involved creative choice), and did not foreclose a claim to joint authorship.
51. We think that the fact that one contributor has the final say does not prevent the existence of joint authorship. That is not to say that it is irrelevant. As in Bamgboye, it may mean that one author is doing more of the work than the other, and so affect the share of the copyright to which each author is entitled. Likewise, it may have an evidential bearing on whether there is a true collaboration. We see no reason, however, to say that where a team collaborates on the production of a work, the fact that the team leader can exercise ultimate control means that there can be no joint authorship.
52. Finally, it is common ground that the shares of two joint authors are not required to be equal. If Ms Kogan and Mr Martin are joint authors, the court will have to go on and apportion the ownership according to their respective overall contributions.
53. Drawing all of this together:

1. A work of joint authorship is a work produced by the collaboration of all the people who created it.

2. There will be a collaboration where those people undertake jointly to create the work with a common design as to its general outline, and where they share the labour of working it out. The first task for the court in such a case is to determine the nature of the co-operation between the putative joint authors which resulted in the creation of the work.

3. Derivative works do not qualify. Works where one of the putative authors only provides editorial corrections or critique, but where there is no wider collaboration, do not qualify. Ad hoc suggestions of phrases or ideas where there is no wider collaboration do not qualify.

4. In determining whether there is a collaboration to create a literary or artistic work it is never enough to ask "who did the writing?". Authors can collaborate to create a work in many different ways. For example there may be joint authorship if one person creates the plot and the other writes the words, or if either or both of these types of labour is shared.

5. Joint authors must be authors, in the sense that they must have contributed a significant amount of the skill which went into the creation of the work. Again, it is not correct to focus exclusively on who fixed the work in writing. The statutory concept of an author includes all those who created, selected or gathered together the detailed concepts or emotions which the words have fixed in writing.

6. Contributions which are not "authorial" in the above sense do not count. What counts as an authorial contribution is acutely sensitive to the nature of the copyright work in question.

7. The question of what is enough of a contribution is to be judged by the Infopaq test, i.e. whether the putative joint author has contributed elements which expressed that person's own intellectual creation. The essence of that term is that the person in question must have exercised free and expressive choices. The more restrictive the choices the less likely it will be that they satisfy the test.

8. The contribution of a putative joint author must not be distinct.

9. There is no further requirement that the authors must have subjectively intended to create a work of joint authorship.

10. The fact that one of the authors has the final say on what goes into the work may have some relevance to whether there is a collaboration, but is not conclusive. The author with the final say must be given credit in deciding on the relative proportions of ownership, for the extra work involved in making those choices.

11. It follows that the respective shares of joint authors are not required to be equal, but can reflect, pro rata, the relative amounts of their contributions.

The judgment of HHJ Hacon

54. The judge identified the issues which he had to decide at [8] as follows:

"(1) What was the nature and extent of Ms Kogan's contribution to the writing of the Screenplay?

(2) Did that contribution make the Screenplay a 'work of joint authorship' within the meaning of s.10(1) of the Copyright, Designs and Patents Act 1988 ("the Act") and qualify Ms Kogan to be joint author with Mr Martin?

(3) Did any act or statement by Ms Kogan, or lack of the same, provide the Part 20 Defendants with a defence of acquiescence or estoppel?"

55. This summary is somewhat compressed. Wrapped up in issue 2 are the questions of whether the screenplay was produced in the course of a collaboration, the nature of that collaboration, and whether Ms Kogan's contributions met the quantitative and qualitative criteria to justify a finding of joint authorship. Whilst the summary is not inaccurate, the judge's failure to separate out the issues of collaboration and contribution makes some of his later reasoning difficult to follow.
56. Having identified the issues in this way, the judge made some observations about Ms Kogan's case at [9] to [12]. He recalled an exchange between himself and counsel for Ms Kogan at the CMC in which he was dealing with an application by the claimants that she should state her case as to the nature of and extent of her contribution with greater precision. Counsel's response was that their respective contributions were so interlinked that it would be impossible to spell out who did what. In the light of this response, further information was not ordered. The judge continued:

"10. When Ms Kogan's evidence was served it must have been apparent that she had changed her case. She now said that she and Mr Martin had each made separate and distinct contributions of various kinds, the remainder of the Screenplay having been created jointly. Annexed to Ms Kogan's witness statement was a breakdown of what had happened, running to 42 pages [ i.e. Annex A]. I did not see it until shortly before the trial when I found it vague and rambling. At the trial Mr Malynicz conceded that it did not explain Ms Kogan's case on her contribution in the most helpful way.

11. The upshot was that the trial started with no focussed idea of what Ms Kogan's case was. On the first day she stated in cross-examination that she alone had been responsible for writing significant parts of the Screenplay with the remainder mostly having been written jointly, although some of it written by Mr Martin alone. I asked Mr Malynicz to provide a marked-up copy of the Screenplay showing Mr Kogan's case as to who had done what. A copy was provided the next day, highlighted in three colours.

12. This was an advance in clarifying Ms Kogan's case but it did not shine light on her argument regarding non-textual contributions. Mr Malynicz said in his closing submissions that he would put Ms Kogan's case by reference to her six best contributions, consisting of both textual and non-textual input, which he would explain. He acknowledged that if Ms Kogan did not get home on these, her counterclaim was liable to fail. Mr Malynicz's selection of the six best contributions was very helpful, bringing some order to the arguments."

57. We do not think this criticism of the way in which Ms Kogan presented her case is really justified. It was and remained Ms Kogan's case that she could not spell out all the contributions she had made because, as is inherent in a case of joint authorship, the contributions of individual authors are not separate or distinct. If two people try to solve a crossword puzzle together, it may be easy to say who first uttered the correct answer to a clue, but more difficult to answer who contributed to it when the solution emerged from a prior discussion. Ms Kogan was nevertheless able, in addition to this basic case, to spell out some instances where she had made identifiable contributions. This did not put the other parties in difficulties, as her evidence was served a long time in advance of the trial.
58. The judge reviewed the law on joint authorship in a long passage between [13] and [56] of his judgment. In relation to the requirements of section 10(1) of the Act he correctly recognised the need for a collaboration, and for a contribution which was not distinct from that of the other author. At [15] he said:

"It was also common ground that s.10(1) contains an implied third requirement. Where a person has contributed to the creation of a work in collaboration with another, he or she must have made a sufficient contribution to qualify as a joint author. This has been explained and rationalised on the basis that a joint author is ipso facto an author within the meaning of s.9(1) of the Act and therefore must have contributed a significant part of the skill and labour protected by the copyright (see Fylde Microsystems Ltd v Key Radio Systems Ltd [1998] FSR 449, at 455)."

59. The judge then tackled, at [16] to [27], the issue of collaboration, but only in the context of two "short cuts" to the resolution of the whole case which had been offered to him by Mr Weisselberg for the claimants. The first short cut, which we have already alluded to, was based on the fact that Ms Kogan "admitted in cross-examination that her input had not been of any significance after the third draft" of the screenplay, completed in October 2013. Between the third and fourth drafts Ms Kogan had moved out of the flat which she shared with Mr Martin. The relationship broke down in October 2014, and the final draft was written thereafter, and the screenplay was probably not finally completed until shooting of the film began in May 2015. As each draft of the screenplay was a separate copyright work, it followed that the final draft was produced by Mr Martin alone "and that was the end of it". The judge accepted this argument. He held at [26] that "There was no collaboration between Mr Martin and Mr Kogan in creating the Screenplay itself." As the Screenplay was defined as the separate work created from the preceding draft, Ms Kogan was not a joint author of the Screenplay. This short cut conclusion is the subject of ground 1 of the appeal.
60. Although, in the judge's view, that was the end of Ms Kogan's claim, he went on to consider her case "on the assumption that her contributions to earlier drafts must be taken into account". Although he did not say so, this approach reflected that taken at the CMC, when Ms Kogan was persuaded to drop separate reliance on the earlier drafts.
61. The judge then turned to a second short cut, which he termed "the ultimate arbiter test". Ms Kogan had accepted that Mr Martin had the final word as to what would and what would not go into the screenplay. It was argued that it necessarily followed that he was the sole author. The judge rejected this submission. In his view the presence of an ultimate arbiter was a "relevant factor", but not determinative. He did not explain whether he considered it to be a factor which affected whether there was a collaboration, or one which went to the relative amounts of creative input by the putative joint authors.
62. Next, at [30] to [37], the judge dealt with the case law on what was a sufficient contribution. He referred, without comment at this point, to the observations of Laddie J and Lightman J in Cala Homes and Robin Ray (which we have commented on at [38] to [41] above), and to Brighton v Jones, but not to Bamgboye v Reed (which we have commented on at [49] to [50] above).
63. At [38] to [41] the judge dealt with what he called the multi-purpose test on what is sufficient to count in copyright (on which we have commented at [43] to [45] above).
64. At [43] to [52] the judge developed a novel theory of his own that it could be inferred from the decided cases that the significance of a contribution depended on the type of skill employed in making that contribution. He drew a distinction between the primary skill required to create a copyright work and other secondary skills. He continued:

"45. In the case of an artistic work for instance, the primary skill lies in the use of a pencil, brush, computer program or other means to create an image. In the case of a literary work such as a novel or screenplay, the primary skill is in the selection and arrangement of words in the course of setting them down.

46. Examples of secondary skills for, say, a painter are composition and selection of colour. For an author of a novel or screenplay, secondary skills include inventing plot and character.

47. I should emphasise that I do not here imply that secondary skills are necessarily less important in the creation of a work, particularly when viewed in other contexts such as commercial success. I draw the distinction only to explain my view on the law of joint authorship.

48. It seems clear that if the putative joint author has contributed by way of using the primary skill, the test of joint authorship is whether the contribution constitutes a substantial part of the whole. This will be akin to the test of substantiality in the context of infringement, a qualitative as well as quantitative assessment.

49. Where the putative joint author has only used secondary skills to make his or her contribution, the case law suggests that it is more difficult to establish joint authorship. Cala Homes provides an exceptional example of joint authorship being established in that way. If Lightman J's observation about Cala Homes [in Robin Ray] quoted above were to be taken too literally, then to succeed in a claim of joint authorship based on a contribution of just secondary skills it is necessary to show that the contribution is sufficiently overwhelming to render the person who provided the primary skills – the draftsman or writer, etc – nothing more than a scribe. I do not believe that this is what Lightman J meant.

50. If, for instance, an individual were to create the entirety of the plot of a novel or play and all the characters featured in it, and a collaborator were left to do the writing with discretion as to wording, such that the collaborator could not be dismissed as a scribe, I doubt that the possibility of joint authorship could be ruled out.

51. Although the case law sets the bar high when it comes to secondary skills, I take the view that there was no intention on the part of the courts in question to create a distinction in law between primary and secondary skills in relation to their intrinsic capacity to give rise to joint authorship. It may often be harder to establish joint authorship solely by reference to secondary skills. But that will be for evidential reasons and/or because it can be difficult to establish that the contribution relied on is the product of relevant secondary skills and thus protected by the law of copyright, as opposed to being what are sometimes called 'mere ideas' which are not protected. The latter distinction turns on a related but separate principle of law."

65. The judge's distinction between primary and secondary skills in assessing contribution has no basis either in the statute or in decided cases. To draw such a distinction tends to imply that less weight is to be given to ideas than to written words, when both are essential components of the work. In our view the distinction is positively unhelpful. Depending on the facts, the person who contributed the ideas may be the "major" author: see per Laddie J in Cala Homes (cited above) at 836.
66. We think that a number of other specific errors are revealed in this passage. At [45] and [46] the judge describes a screenplay as a literary work, like a novel, and describes the primary skill as the selection and arrangement of words in the course of setting them down. This passage fails to make what in our view is an important distinction between a novel and a screenplay. We think a screenplay is more accurately described as a dramatic work, as its primary purpose lies in being performed, as opposed to being read, like a novel. The importance of the distinction is put in this way by the authors of Copinger and Skone James on Copyright, 17^th Edition at para 7.93:

"… a basic distinction between literary works and dramatic works is that the choice of dramatic incident and the arrangement of situation and plot may constitute, to a much greater extent, the real value of a dramatic work. … It should be remembered that dramatic works include not only plays and screenplays …".

67. Irrespective of this distinction, it is wrong to describe inventing plot and character as a secondary skill: see in this connection, the critique of this part of the judge's reasoning in Simone (cited above), at page 39. The arrangement of situation and plot, as well as character development, can be of the essence of a literary or dramatic work. We accept that the judge then went on to disclaim reliance on "a distinction in law between primary and secondary skills in relation to their intrinsic capacity to give rise to joint authorship", but he nevertheless gave (at [50]) a fairly extreme example of where joint authorship "could not be ruled out": namely where a co-author contributed the whole of the plot and all the characters. It is fair to assume that this and his other remarks (at [51]) ("sets the bar high"; "harder to establish joint authorship solely by reference to secondary skills") did carry through into his analysis of Mr Kogan's contributions in the present case.
68. The judge then summarised the law as follows at [54]:

"(1) A party will be joint owner of the copyright in a work only if he or she (or in the case of a company, its employees) collaborated in the creation of the work. The collaboration must be by way of a common design, i.e. co-operative acts by the authors, at the time the copyright work in issue was created, which led to its creation.

(2) The contribution of each author must not be distinct from that of the other author or authors.

(3) Contributions by a putative joint author (including those done by way of collaboration) which formed no part of the creation of the work are to be disregarded in the assessment of joint authorship.

(4) No distinction is to be drawn between types of contribution that did form part of the creation of the work. In particular, there is no distinction which depends on the kind of skill involved in making the contribution.

(5) The contribution, assuming it is relevant to the assessment of joint authorship, must be sufficient. This depends on whether the contribution constitutes a substantial part of the whole of the work in issue.

(6) That will be the case if the contribution would be protected by copyright in the work. Thus, if the contribution alone were copied by an unlicensed third party and such copying would result in an infringement of the copyright, the contribution constitutes a substantial part of the whole.

(7) The test of substantiality in the context of joint authorship of copyright, as in the context of infringement, involves a qualitative as well as quantitative assessment.

(8) Suggestions from a putative joint author as to how the main author should exercise his or her skill – for instance by way of criticism or editing of a literary work – will not lead to joint authorship where the main author has the final decision as to the form and content of the work.

(9) It is thus relevant, but not decisive, whether an author is the ultimate arbiter as to the content of the work.

(10) If joint authorship is established, the court may apportion ownership of the copyright."

69. We read the judge's fourth proposition as stating, correctly, that contributions which formed part of the creation of the work are not to be ruled out of account because they are, for example, contributions of a non-textual nature. As we have said, however, the judge had earlier concluded that there was a "high bar" for contributions of plot and character, a proposition with which we disagree, particularly when applied to dramatic works. We would also sound a note of caution in relation to the judge's sixth proposition. It is in the nature of a work of joint authorship that the contributions of the authors are not distinct. It follows that there are logical difficulties in assuming that one can separate out the contribution of one author and ask if it is protected by the copyright in the work. The sixth proposition is better understood as an example rather than as an all-purpose test. It is preferable simply to ask whether the joint author contributed a substantial part of the relevant creative skill which is protected by the copyright in the work.
70. The judge then embarked, between [57] and [61], on a lengthy consideration of the reliability, as a general matter, of the recollection of witnesses. He started by pointing out that, unsurprisingly, the main evidence of who did what came from Mr Martin and Ms Kogan. The evidence of one contradicted that of the other. His overall impression was that both were stating their honestly held views. He accepted that this "superficially makes no sense", but proceeded to explain his approach by citing at length the observations of Leggatt J (as he then was) as to the fallibility of human memory in Gestmin SGPS S.A. v Credit Suisse (UK) Ltd [2013] EWHC 3560 (Comm) at [15] to [22] and in Blue v Ashley [2017] EWHC 1928 (Comm) at [68] to [69]. The judge understood these passages as "an admonition that the best approach for a judge is to place little if any reliance at all on witnesses' recollections of what was said in meetings and conversations and instead base factual findings on inferences drawn from documentary evidence and known or probable facts". He considered that approach to be appropriate in the present case.
71. The judge then summarised, at [62] to [74], Ms Kogan's six best contributions which she had put forward for testing her case before, at [75] to [85], going on to deal with the facts. He structured his decision by dividing it into Ms Kogan's contributions to the text, and her non-textual contributions. At [78] he found that Ms Kogan's textual contributions as a sole writer were limited to contributions of technical musical language, incorporated into drafts one to three. The judge went on to reject Ms Kogan's claim that most of the remainder of the text had been co-written by her with Mr Martin. He accepted that there was some limited evidence in support of this, consisting of an email in which Mr Martin referred to Ms Kogan as "his special collaborator" and an email from Ms Kogan in which she said that it was "such fun for me to work on these things with you". These were, however, "very early days and the affectionate language of emails gives little idea of how Mr Martin and Ms Kogan planned to work together on the subsequent writing of the draft screenplays for the Florence project". The judge continued:

"81. By contrast there was more telling agreed and documentary evidence which suggested that the first three drafts of the screenplay were written by Mr Martin alone, with only very limited help from Ms Kogan:

(1) By the time of the Florence project Mr Martin was a screenwriter of some standing, having written several screenplays for TV drama shows. Ms Kogan was an operatic singer by training and profession. She had written magazine articles and children's books, but claimed no experience of screenwriting before she met Mr Martin.

(2) In July 2012 Ms Kogan had suggested that she and Mr Martin should jointly write a script for a different project. Mr Martin reacted badly to the suggestion and told Ms Kogan that he always worked alone. Despite his being in a romantic relationship with Ms Kogan, there was no evidence that he changed his mind about this, aside from Ms Kogan's assertions that he had.

(3) Mr Martin was the ultimate arbiter of what went into the screenplay drafts.

(4) Emails at the time the story outlines were being written (i.e. before the first draft of the screenplay) indicate that Ms Kogan was making typographical corrections and other comments of a proof-reading kind, but nothing more substantial. An email dated 31 March 2012 from Ms Kogan is typical in this regard. It also includes this:

"Of course, please ignore whatever you please of what I say, this is totally your baby."

(5) Mr Martin started work on the first draft of the screenplay in February 2013 while Ms Kogan was in France. It was completed in early April. Ms Kogan wrote an entry in her diary on 3 April 2013:

"N finishes Flo."

(6) The second draft was completed on 19 June 2013. Mr Martin emailed a copy to Ms Kogan on 21 June 2013. Ms Kogan replied on 23 June 2013:

"...you have done it – you have accomplished the great task of creating something truly important. ... All the more so because you have written Florence under so much pressure, in such difficult circumstances. ... Making something from nothing but your own talent is a value in itself. To create in a near vacuum is the real thing."

(7) Mr Martin went to Los Angeles in July 2013 where he met people in the Hollywood film industry with whom he reviewed the Florence project, in particular Ben Lewin, a director, producer and writer of screenplays and Mr Lewin's wife, Judi Levine. Their son carried out some research for the project. In an exchange of texts on 27 July 2013 Ms Kogan advised:

"You must protect your position as the writer."

(8) After Mr Martin had returned to London, work was done on the third draft. Mr Martin was concerned that Mr Lewin and Ms Levine would claim a share of the authorship, making it difficult for him to claim sole authorship when selling the project to finance companies. To neutralise the difficulty, he decided to offer them part of his writing income. 15% was later agreed. In an email to Ms Kogan dated 23 January 2014 Mr Martin discussed this – he called it "the Ben and Judi problem" – and asked her what she thought about offering part of his income. On the same day Ms Kogan replied "It makes sense...". On 5 March 2014 Mr Martin's company, the Second Claimant, signed an agreement with Qwerty, the third Part 20 Defendant, for the financing of the Film. It included a warranty that Mr Martin was sole writer of the script. Throughout this Ms Kogan did not suggest that she should be given a share of the income. It was not raised until 8 April 2014, following a therapy session attended by Ms Kogan and Mr Martin in an attempt to save what had by then become a difficult relationship.

(9) Ms Kogan knew that Mr Martin had registered the first three drafts of the screenplay with the Writers' Guild of America with Mr Martin listed as sole author. She raised no objection."

72. The judge made these findings, true to his belief that he was taking the course advocated by Leggatt J in Gestmin and Blue, exclusively by reference to inferences drawn from agreed and documentary evidence and disregarding assertions made by Ms Kogan. He concluded at [82]:

"In my view, the documentary and undisputed evidence supports Mr Martin's claim to have been the sole writer of the text of drafts one to three of the screenplay with only limited input from Ms Kogan. This consisted of musical expressions of a technical nature, already mentioned, together with some minor editing changes."

73. The judge then turned to Ms Kogan's non-textual contributions, which he gave very short shrift, confined to the following two paragraphs:

"83. I think it is likely that Ms Kogan was also responsible for non-textual contributions to the first three drafts, made in the course of her discussions with Mr Martin. Looking at the six best examples advanced on her behalf, the non-textual input probably consisted of the following ideas:

(1) using Lil[l]y Pons as a character in the Film and using the Bell Song;

(2) using The Swan in McMoon's audition scene;

(3) having Florence visit Melotone Records.

84. As with the textual input, it was Mr Martin who decided which of Ms Kogan's ideas were to be used in the Screenplay and which not."

74. The judge's overall conclusion at [85] was that:

"… the textual and non-textual contributions made by Ms Kogan never rose above the level of providing useful jargon, along with helpful criticism and some minor plot suggestions. Taken together they were not sufficient to qualify Ms Kogan as a joint author of the Screenplay, even had those contributions all been made in the course of a collaboration to create the Screenplay. Mr Martin was the sole author."

75. The judge went on to deal with the defence raised by the film companies based on acquiescence or estoppel, which did not arise in the light of his findings thus far. Ms Kogan had sought to argue that the defence was not open to the film companies, as it was precluded by EU law. As this argument had not been pleaded, the judge did not consider it further. The judge made limited findings about the acquiescence/estoppel defence as follows:

"91. Ms Kogan's statements and behaviour towards the Part 20 Defendants, and indeed everyone else, were always consistent with, and I think took the form of, a representation that she believed that the Film should be produced and released. In my view, had she been joint author of the copyright in the Screenplay it would have been unconscionable for her rely on joint ownership of the copyright to restrict the public performance of the Film in any way. To that extent, in my view the defence of estoppel would have succeeded.

92. I can take it no further because to do so would require hypothetical facts arising from the assumption that Ms Kogan had made significant textual and other contributions to the Screenplay but had not informed the Part 20 Defendants about them or made any claim to joint authorship until late in the day. Whether Ms Kogan acted unconscionably would depend on the precise nature and circumstances of those facts."

The appeal

76. Ms Kogan advances seven grounds of appeal. Ground 1 attacks the judge's conclusion on the first short cut, by asserting that changes made by one of the authors once the collaboration has ended do not prevent the second author from claiming to be a joint author of the work as a whole. Ground 2 asserts that the judge failed to apply EU law on "the author's own intellectual creation", by failing to consider whether the input concerned expressive and creative choices, and failing to consider that isolated sentences or even parts of sentences may convey originality. Grounds 3 and 4 focus on the judge's assessment of Ms Kogan's textual contributions to The Rehearsal Scene, The Scene in the Lift and the Further Rehearsal Scenes, asserting that he ignored the true nature and extent of Ms Kogan's contribution, ignored key documents and ignored key passages of cross-examination. Ground 5 asserts that the judge's assessment of Ms Kogan's non-textual contribution was flawed, in particular his assessment of the Lily Pons Scene and The Bell Song, The McMoon Audition Scene and the scene at Melotone (Homophone) Records where Florence takes charge of the action herself. Ground 6 asserts that the judge erred by failing to take account of Ms Kogan's largely unchallenged evidence detailing her full contribution to the writing of the screenplay. Finally, ground 7 attacks the judge's conclusion on the film companies' defence of acquiescence/estoppel.
77. It will be more convenient to consider the issues in a somewhat different order, but we start with ground 1.

Ground 1: The first short cut

78. Mr Malynicz submitted that a work could still be a work of joint authorship if two authors had collaborated up to a certain version of the work, but where the final version had been the work of one author alone. He submitted that this followed from the language of section 10(1) which defined a work of joint authorship as "a work produced by the collaboration of two or more authors". The final version could still be produced by collaboration where the contribution of one collaborating author was retained from the earlier version or versions. He also argued that the issues for decision at the trial had been defined on the assumption that this was a correct view of the law, and that the case had been managed on this basis throughout. He also submitted that the manner in which this point was taken by Mr Martin was unfair. Despite the strict rules of pleading in the IPEC, it had not been pleaded that it was a complete answer to the claim that the parties had not collaborated on the work involved in creating the final screenplay from the preceding drafts. The point had also not been advanced in Mr Martin's skeleton argument for trial.
79. Mr Weisselberg supported the judge on this point. The judge was technically correct, and any perceived procedural unfairness was mitigated by the fact that the judge had, in the alternative, rejected Ms Kogan's claim when considered across all the drafts.
80. Where a copyright work is created by working on a series of drafts, it is possible to analyse the artistic or literary input in two ways. One approach is to treat each draft as a separate, derivative copyright work. On this approach only the skill and labour which goes into producing the draft in question is assessed for originality. A second approach is to treat the work holistically as a single work which required the totality of the skill and labour involved in producing all the drafts to produce it. Often the choice between these two approaches will not matter, for example if it is clear that one author produced all the drafts, or all the authors made the work in the course of their employment for a single employer in whom the copyright will vest by operation of law. Where all the relevant copyright is vested in the same person, splitting up the skill and labour into historical chunks will only add complexity. The present case, by contrast, was one where one could see from the pleadings that it could well matter which approach was taken, as it was clear that Ms Kogan had been more involved in the early drafts than the later ones. Certainly, this was never a case which would turn on the circumstances relating to the converting of the penultimate draft into the final screenplay.
81. If the case is to be analysed on a draft-by-draft basis, then we would not accept Mr Malynicz's ingenious argument based on the word "produced" in section 10(1) of the Act. Whilst earlier works may have been produced by collaboration, the work in question was not. It was a derivative work of which Mr Martin was the only author.
82. The real question, as it seems to us, is whether it was open to Mr Martin to treat this case as one which was to be analysed on a draft-by-draft basis. In our judgment it was not. It is clear from the transcript of what was said at the CMC that the case proceeded on the agreed footing that Ms Kogan could succeed in her action for infringement of copyright if she proved contributions into earlier drafts. Restricting the case to the final draft only made sense, therefore, if Ms Kogan's earlier contributions which carried forward into the final draft had the legal consequence (if otherwise meeting the tests for joint authorship) that she would become entitled to a share of the copyright in the final draft. It was only on this basis that Ms Kogan's explicit reliance in her pleadings on the copyright in the earlier drafts could be treated as unnecessary.
83. If Mr Martin was going to advance the short cut point, then in our judgment it was incumbent on him to raise it much earlier. A case that Ms Kogan's claim was unmaintainable on the ground that any collaboration between the parties ceased before the creation of the final draft was exactly the sort of knockout point which should have been flagged up at the CMC or, if that was not done, pleaded. As Peter Jackson LJ pointed out in the course of argument, this point, if correct, rendered all the elaborate evidence about what occurred in relation to the earlier drafts completely unnecessary. Moreover, if the point had been raised earlier, there is no doubt that it would have provoked an application on the part of Ms Kogan to revert to reliance on the copyright in the earlier drafts.
84. It is true, as Mr Weisselberg submitted, that the judge did not rely solely on this point, but, in our judgment, he should not have relied on it at all. It is either a free-standing answer to Ms Kogan's claim or it is not.
85. The unfairness to Ms Kogan in allowing this unpleaded point to defeat her claim is in contrast to the judge's strict adherence to the IPEC rules of pleading on other issues. Thus, he did not allow Ms Kogan to advance her legal answer to the film companies' defence of estoppel/acquiescence precisely because it had not been pleaded, and even though it raised a pure point of law. In our judgment he should, in fairness, have dealt with the short cut point in exactly the same way.

Ground 6: The judge's treatment of the evidence

86. Mr Malynicz submitted that the judge had wrongly and unfairly ignored Ms Kogan's evidence, contained in her witness statement and its Annex A. Annex A was a detailed scene-by-scene analysis of Ms Kogan's contribution to the screenplay, which was largely unchallenged. It was not "vague and rambling" as the judge had unfairly characterised it, and the judge's statement that he had considered it "shortly before trial" was not a ground for dismissing it. The witness statement with Annex A had been served, in accordance with the judge's own directions, on 5 May 2017, five months before the trial.
87. Mr Weisselberg's response was that Annex A had not been ignored by the judge, who must have read it in order to describe it as he did. The judge simply did not find it helpful, and preferred to test Ms Kogan's case by reference to the six best contributions, assisted by a colour mark-up of the screenplay produced during the trial, and by reference to the documents passing between the parties.
88. We think that there is real substance in this ground of appeal. We start by recalling that the judge read Leggatt J's statements in Gestmin v Credit Suisse and Blue v Ashley as an "admonition" against placing any reliance at all on the recollections of witnesses. We consider that to have been a serious error in the present case for a number of reasons. First, as has very recently been noted by HHJ Gore QC in CBX v North West Anglia NHS Trust [2019] 7 WLUK 57, Gestmin is not to be taken as laying down any general principle for the assessment of evidence. It is one of a line of distinguished judicial observations that emphasise the fallibility of human memory and the need to assess witness evidence in its proper place alongside contemporaneous documentary evidence and evidence upon which undoubted or probable reliance can be placed. Earlier statements of this kind are discussed by Lord Bingham in his well-known essay The Judge as Juror: The Judicial Determination of Factual Issues (from The Business of Judging, Oxford 2000). But a proper awareness of the fallibility of memory does not relieve judges of the task of making findings of fact based upon all of the evidence. Heuristics or mental short cuts are no substitute for this essential judicial function. In particular, where a party's sworn evidence is disbelieved, the court must say why that is; it cannot simply ignore the evidence.
89. Secondly, the judge in the present case did not remark that the observations in Gestmin were expressly addressed to commercial cases. For a paradigm example of such a case, in which a careful examination of the abundant documentation ought to have been at the heart of an inquiry into commercial fraud, see Simetra Global Assets Ltd & Anor v Ikon Finance Ltd & Ors [2019] EWCA Civ 1413 and the apposite remarks of Males LJ at paras. 48-49. Here, by contrast, the two parties were private individuals living together for much of the relevant time. That fact made it inherently improbable that details of all their interactions over the creation of the screenplay would be fully recorded in documents. Ms Kogan's case was that they were bouncing ideas off each other at speed, whereas Mr Martin regarded their interactions as his use of Ms Kogan as a sounding board. Which of these was, objectively, a correct description of their interaction was not likely to be resolved by documents alone, but was a fundamental issue which required to be resolved.
90. Thirdly, having decided to follow the Gestmin/Blue approach, the judge did not apply it to documents which greatly assisted Ms Kogan's case. The two documents to which the judge referred at [79] strongly supported an inference that the parties were collaborating on the screenplay at the outset, but the judge declined to draw any inference from them, instead observing that this was early in the project and that the affectionate language used gave little idea of how the parties planned to work together on the subsequent drafts. It was, however, important for the judge to come to a conclusion as to the basis on which the original outline was created. He was tasked with deciding, in the words of Keating J, whether Ms Kogan and Mr Martin had "undertaken jointly to write a [screen]play, agreeing in the general outline and design, and sharing the labour of working it out". Their approach at the outset was highly relevant.
91. The judge's inconsistent approach to the documents is also apparent much later in the process. In a draft email, albeit in the context of seeking to find Ms Kogan a position as voice coach for Meryl Streep in the film, Mr Martin said that Ms Kogan was intimately involved in helping to develop the character of Florence. A colleague to whom Mr Martin showed the draft email described it as honest, generous and fair. Mr Martin did not suggest that he had not been telling the truth in the email. The judge does not refer to this email at all. Recourse to the Gestmin/Blue approach does not justify a selective treatment of this kind.
92. Yet further, when a friend sent a storyline to Mr Martin for him to comment on, he replied by suggesting that he send it to Ms Kogan as well because "she is good at story". The judge makes no reference to this exchange, although it supported an inference that Mr Martin relied on and valued Ms Kogan's abilities with "story".
93. Fourthly, the documents on which the judge relied in support of Mr Martin's case were themselves a shaky foundation for drawing the hard inferences which he drew from them, particularly with regard to where the non-textual creative input was coming from.
94. Thus, at [81(4)] the judge refers to emails from the time when the story outlines were being written which showed Ms Kogan making only typographical corrections and other comments of a proof-reading kind. One email of 31 March 2012, much relied on by Mr Martin, contained the line:

"Of course, please ignore whatever you please of what I say, this is totally your baby."

95. Ms Kogan's witness statement had provided the context for that email, however. She explained that she had been sent a revised version 2 of the FFJ outline on 29 March 2012 (incorporating details which they had discussed that day). She and Mr Martin were "bouncing the drafts back and forth between us". On 30 March 2012, Mr Martin sent her a revised version 3 of the outline asking what she thought of the revisions, and thanking her for helping him with the work on 29 March. She was concerned, however, that Mr Martin was tiring of what he saw as constant nit-picking with the story and characters, so she decided to tread carefully, diminishing the negative comments while hyping up the positives. She said that:

"For the most part, Nick followed my suggestions about both plot and character decisions and changes, as in our Skype conversation of 29 March 2012, but he also could become irritated on occasion. This is why I wrote that he should ignore whatever he pleases of what I say, as this is "totally your baby"."

96. In any event, this email might go no further than establishing that Mr Martin was what the judge called "the ultimate arbiter", which was common ground. It did not necessarily establish that Ms Kogan had not made a collaborative contribution of the relevant kind.
97. The judge also identified at [81(5)] an entry in Ms Kogan's diary which recorded "N finishes Flo". This justifiably supports the inference that Mr Martin was the one who was fixing the work in writing and wielding the pen, but does not come close to making it improbable that Ms Kogan was providing significant creative input. At [81(6)] the judge identified an email from Ms Kogan congratulating Mr Martin on finishing the second draft of the screenplay. She said that he had done so "from nothing but your own talent". But it is impossible to treat this document as literally true, given the admissions by Mr Martin as to what Ms Kogan had contributed, and, if so, it is difficult to understand why the judge did not regard this as mere affectionate encouragement. She was hardly likely to say "Look what you have achieved with all my help".
98. Fifthly, the judge quoted Ms Kogan's advice to Mr Martin that he must "protect your position as the writer", without referring to both parties' evidence that they were supporting each other in their careers. Ms Kogan's evidence was that she wanted Mr Martin to receive all the plaudits, but this does not diminish the significance of her contribution.
99. Sixthly, the judge also placed weight, whilst disregarding Ms Kogan's evidence about them, on the public statements made about Mr Martin being the sole screenwriter, to Ms Kogan's knowledge. It is, however, hardly surprising that, whilst the parties were in a relationship, and supporting each other in their careers, Ms Kogan did not agitate for a public acknowledgement of her contribution. That was her evidence, but, applying Gestmin, the judge ignored it.
100. To be fair to the judge, it may be that he did not rely on these documents when it came to considering non-textual input, as he does not cross-reference them in the short section of his judgment where he deals with that input. If so, his reasons for rejecting the case based on non-textual input are correspondingly harder to discern.
101. As a separate point we think that the judge should not have criticised Annex A in the way he did, far less give the impression that he had had insufficient time to absorb its contents. This was bound to leave Ms Kogan with the impression that her case had not been fully considered. Having read Annex A, we consider that it was a genuine attempt to give a fair picture of the way in which Ms Kogan said that she had contributed to the creation of the screenplay. It made frank admissions where she considered she had provided no input to a scene, but explained the nature of her contribution where she considered to the contrary. Whilst it may have been longer than required and contained some passages which did not assist, it was not right for the judge to dismiss all of its contents out of hand.
102. One of the items of agreed evidence relied on by the judge was the incident in July 2012 where Mr Martin had reacted badly to the suggestion that he and Ms Kogan should jointly write a script for a different project, and where he had told Ms Kogan that he always worked alone. The judge observed that, despite his being in a romantic relationship with Ms Kogan, there was no evidence that he changed his mind about this, aside from Ms Kogan's assertions to the contrary. We fail to see why Ms Kogan's assertions to the contrary should be ignored. Whilst the judge was plainly entitled to rely on this incident for what it was worth, it could not possibly amount to a complete answer to Ms Kogan's case on collaboration, given what had occurred both before and after July 2012 in relation to the Florence project and particularly when one considers whether there was collaboration in relation to non-textual matters, such as character, plot and dramatic incident.
103. The judge's approach of relying only on documentary and agreed evidence has led him to fail to make findings on key issues. As we have already pointed out, his summary of the issues concealed a number of sub-issues which he needed to decide. This included the question of whether there was a collaboration between the parties to create the work, and what was the precise nature of that collaboration. The judge has left unresolved the question of whose idea the Florence project was. None of his findings negates Ms Kogan's case that they would bounce ideas off each other in discussions so that it was impossible to tell who came up with what, or, more generally, that they collaborated at the non-textual level.
104. We are of course deeply conscious of the guidance given at the highest level to all appellate courts that findings of fact by lower courts must not be disturbed except in very limited circumstances. Our problem is a rather different one. Because the judge has expressly declined to have regard to the witness evidence which he heard, we are left without the benefit of what we consider to be essential findings of primary fact, which we are in no position to make ourselves.
105. This was not a case where the conflicts in the written and oral evidence of the parties could be described as honest differences of recollection. If the judge's findings are correct, Ms Kogan's written and oral accounts of the process of creating the screenplay in her witness statement, Annex A and her oral evidence were simply untrue. The upshot is that a reader of the judgment is left without any clear idea of the nature of the interaction between Mr Martin and Ms Kogan in the creation of the screenplay. We will have to return to the consequences of this conclusion when we have considered the other grounds of appeal.

Ground 2: Wrong legal standard?

106. Mr Malynicz submitted that the judge had wrongly imposed too high a threshold for determining what was a sufficient contribution to justify a finding of joint authorship. All that was required was that the contribution be of elements which were the expression of the author's own intellectual creation, by which was meant that the author was making free and expressive choices. This was a relatively undemanding test.
107. Subject to the points we have already made about paragraph 54, we do not think that the judge can be criticised for not setting out the law correctly in his list of propositions in that paragraph. Mr Malynicz rather tentatively questioned paragraph 54(7) where the judge said that the test for joint authorship involved a quantitative requirement. Given his acceptance of the multipurpose test (paragraphs 43 to 45), he correctly identified the quantitative threshold by reference to the test propounded by Mr Malynicz. Despite this, and as we have explained at [65] to [67] above, we think that that the judge had in mind a more demanding test for what he described as contributions of secondary skills, within which he included plot and character.
108. Mr Malynicz therefore focused on the judge's application of the test for joint authorship. Those submissions are best considered when dealing with the judge's rejection of Ms Kogan's contributions as insufficient, which we now address.

Ground 3 to 5: The judge's rejection of the sufficiency of Ms Kogan's contribution

109. Before turning to some of the specific examples of contribution on which Ms Kogan relied, it is necessary to understand what the judge decided on the logically prior issue of whether there was a collaboration. The judge made a clear finding at [26] that there was no collaboration between Mr Martin and Ms Kogan in creating the final draft of the screenplay. Mr Weisselberg submitted that the judge had made a finding that there had been no collaboration at any earlier time.
110. We do not consider that the judge did make such a finding except in relation to the final draft. His carefully phrased finding at [82] that Mr Martin was the sole writer of the text does not preclude the existence of an authorial collaboration. His brief treatment of the non-textual contributions at [83] to [84] does not explain whether he found there to be a collaboration or not. The only point made is that Mr Martin was the ultimate arbiter of what went in. That fact does not preclude, as the judge found elsewhere, a finding of joint authorship. At [85] he finds the contributions to be insufficient "even had those contributions all been made in the course of a collaboration to create the Screenplay".
111. We do not read those last words as a finding that the contributions were not made in the course of a collaboration to create the screenplay. First, some of the contributions were made at an early stage when there was clear documentary evidence that Mr Martin regarded Ms Kogan, at least in some sense, as his collaborator, and when this was Ms Kogan's pleaded and articulated case. Secondly, as we have noted, in framing the issues for his decision at [8] he did not include as a separate issue the question of whether there was a collaboration, and his subsequent specific consideration of the issue was only in relation to the short cut. Probably as a consequence, it does not appear that the judge anywhere asked himself the separate question whether the contributions that he accepted had been made were made in the course of a collaboration to create the screenplay. Thirdly, as we have explained at [102] above, the incident when Mr Martin protested that he always worked alone could not possibly extend to the whole of Ms Kogan's contributions both before and after the incident, and was consistent with working alone on committing the work to writing after discussions with Ms Kogan on the plot and characterisation.
112. We consider that the judge has decided the case on the nature and extent of Ms Kogan's contribution without making any specific finding as to whether there was a collaboration.

Ms Kogan's textual contributions

The Rehearsal Scene

113. The judge described the Rehearsal Scene in these terms at [67]:

"Florence's first rehearsal with McMoon also takes place in her apartment. She is coached by Carlo Edwards who is an assistant conductor at the Metropolitan Opera. Her partner, St Clair Bayfield, is present. She begins by singing the Bell Song. The aria features a short introduction, here by McMoon on the piano, followed by a long unaccompanied vocal flight from Florence. She is tutored and encouraged by Edwards who uses language such as "Raise the soft palette!" and "Think of the mask, Florence. The squillo! The voice is the mask!". (Mr Martin explained squillo, having been informed by Ms Kogan: "… it is about getting the cavities in the forehead to resonate so that the sound will reach the back of the room with a certain ping.")"

114. The judge accepted at [78] that Ms Kogan contributed the technical musical language used in this and other scenes as sole writer. Anyone reading the scene can see this jargon is a major part of the scene. In cross-examination Mr Martin characterised this jargon as "guff", suggesting that it did not matter what phrases were used. Be that as it may, it is clear that the choice of language has an important impact on the scene, and there can be no doubt that Ms Kogan had a wide range of possible expressions to call on, not all of which would have contributed to the comedic dialogue in the same way.
115. The judgment does not explain in any detail how, despite his acceptance that Ms Kogan was the sole writer of the musical language used in scenes such as this, the judge nevertheless felt able to reject Ms Kogan's case that she and Mr Martin jointly created the dialogue in this scene. It cannot depend on Mr Martin's insistence on working alone, because it is impossible to see how she could have written the musical language in the scene without some form of collaboration with Mr Martin. Mr Weisselberg submitted that it was because all Ms Kogan had done was to provide a list of musical expressions for Mr Martin to choose from, for example as a researcher might have done. He seeks support for this in the judge's conclusory paragraph [85] which says that Ms Kogan's contributions did not "rise above the level of providing useful jargon", but this does not explain how she could have "written" these phrases as sole writer and it could not be more flatly contrary to Ms Kogan's evidence. In Annex 1 Ms Kogan had said:

"For me, writing this scene was a delight. I relied on my experience and knowledge of the world of professional singing and delved into my headful of musical and singers' psychobabble to write many of the lines."

116. She went further and said:

"Some of the other funny dialogue was written by us both. We had a tremendous laugh choosing our double entendres for the lines, as in, "But you've never sounded better!", "It's hard to believe, isn't it," and "There is no one quite like you…".

117. The judge plainly needed to resolve the issue of how this contribution came to be made, but failed to do so.
118. If the contribution was made as a part of a collaboration to create the work, then there were two possible grounds on which the judge could reject it. The first would be that it was not a contribution of a character that could count. The second would be that, taken with all other relevant contributions, it was quantitatively insufficient.
119. In our judgment, if the judge was rejecting the contribution to this scene on the ground that it was not "authorial" he was wrong to do so. There can be no dispute that the contribution and setting of these musical expressions was of the right kind. It follows that, if Ms Kogan's contribution to this scene was made as part of a collaboration, it deserved to be put in the scales, together with the other evidence of a contribution, to determine whether the total contribution was a substantial part of the skill and labour which went into creating the work.

The Scene in the Lift and the Further Rehearsal Scenes

120. Mr Malynicz's submissions under this head are of a similar character. Ms Kogan claimed, and Mr Martin accepted, that the complex phrase "her vocal chords do not phonate freely … her phrasing is haphazard … and as for her sub glottal pressure …" used by McMoon to describe FFJ's voice came from her. Mr Weisselberg's response is that this again is the type of information one might request of a researcher. For example, someone writing a scene about a cricket coaching session might ask for phrases such as "move your feet", "keep your eye on the ball" and so on. That is a perfectly fair point if the relationship is one between author and researcher, but the analogy breaks down if the phrases are supplied as a contribution to a joint creative process. In the latter circumstance the contribution needs to be put in the scales, together with the other evidence of a contribution, to determine whether the total contribution was a substantial part of the skill and labour which went into creating the work. Again, we would reject any suggestion that the contribution could be rejected because it is not authorial.

Ms Kogan's non-textual input

Lily Pons and The Bell Song

121. The first item of non-textual input on which Ms Kogan relied was the introduction of the character Lily Pons and the choice of The Bell Song (from Delibes' opera Lakmé). The judge accepted that it was Ms Kogan's idea to introduce the character and song. He dismissed the contribution, however, as "a minor plot suggestion".
122. Our attention was drawn to the cross-examination of Mr Martin on the significance of the introduction of Lily Pons and The Bell Song into the script. Mr Martin accepted that the choice of the song, with its long unaccompanied section, worked particularly well for introducing and displaying FFJ's woeful singing to the audience. The choice of Lily Pons also worked well because one saw a good performance of the song at the beginning of the film by an accomplished opera singer, and then a contrast when the same song is attempted later by FFJ. Mr Martin said "…that is the drama of it…".
123. There was also evidence that Ms Kogan later helped persuade the film's director Stephen Frears that the choice of the song was the right one, in the face of the suggestion that it should be replaced by Mozart's Queen of the Night aria from The Magic Flute. She explained why she had chosen it, and how it was linked to the script and the character of Florence:

"The opening of the Bell Song from Lackme [sic] is done with no accompaniment and is absolutely breathtaking in its purity and virtuosity. It is the sort of moment where one could hear a pin drop as everyone listens with bated breath. Quite magical and precisely the touching moment of communion Florence refers to in scene 18".

124. Mr Malynicz therefore submitted that Ms Kogan's contribution was an important part of the development of the serious side of the character of FFJ, and an important and integral part of the screenplay.
125. Mr Weisselberg characterised Ms Kogan's contribution as falling short of introducing these elements into the plot. The mere suggestion that a historical character be included in a work or that she sing a particular song was not a free and creative choice by the person making the suggestion. The relevant act of free and creative choice was by the person deciding to include that person or song and deciding how that person or song will be used in the context of the work that the author is creating. Here it was Mr Martin, or later the director and producer of the film, who made these choices, and not Ms Kogan.
126. The main difficulty with assessing these submissions is, again, the lack of a finding by the judge as to the nature of the interaction between Ms Kogan and Mr Martin at the time these elements were suggested by her. There is of course substance in Mr Weisselberg's submissions if the nature of their interaction was simply that Mr Martin asked Ms Kogan to provide the name of a contemporary opera singer and a likely song for her to sing. But if, as Ms Kogan suggested, she was fully aware of the context in which her suggestions were being made, and she was matching her suggestions to the drama of the screenplay and the character of FFJ that they were seeking to portray, then the force of these submissions falls away. If that were so, she was contributing important and detailed elements of the dramatic work in such a way that the work could become a jointly created one.
127. The judge's rejection of Ms Kogan's non-textual contributions as "minor plot suggestions" is expressly on the basis that they were not sufficient "even had those contributions been made in the course of a collaboration to create the Screenplay". He was not, therefore, approaching it on the basis on which Mr Weisselberg was inviting us to, namely of an acontextual suggestion of a historical character and song. Rather he was saying that, even assuming that Mr Martin and Ms Kogan were collaborating on producing the screenplay, these plot elements did not satisfy the remaining requirements for a work of joint authorship.
128. If there was a collaboration, then these plot elements, intimately interwoven as they are with the drama of the screenplay and the character of FFJ, also needed to be placed in the scales to see whether the combined contribution passed the relatively undemanding threshold to render Ms Kogan a joint author.
129. The judge appears to have relied heavily on the fact that it was Mr Martin who decided on which of Ms Kogan's ideas were to be used in the screenplay and which not. This point is singled out at [84], as the only point made under the heading of non-textual contributions, indicating that the judge was applying this as a relevant factor ("sometimes highly relevant") as he had held at [29] and [54(9)]. As we have pointed out, however, the question of who is the ultimate arbiter might be relevant in two quite different ways: as a factor affecting relative contributions and thus shares in the copyright, or an evidential factor in deciding whether there was collaboration at all. It is not clear in what way the judge deployed it. As we have already said, he made no finding about collaboration on the first three drafts. If he was saying that the extra work undertaken by Mr Martin in deciding whether to adopt Ms Kogan's plot suggestions somehow rendered those suggestions less significant, he should, in our view, have explained why.

The McMoon Audition

130. In this part of the screenplay several musicians are called to an audition to accompany FFJ as a pianist as she practises her singing. One, Cosmé McMoon, is selected on the basis of his playing of The Swan by Saint-Saëns. In an earlier version Ms Kogan had suggested a similarly sentimental piece, Babes in Toyland, but the producer rejected this. The judge accepted that Ms Kogan had suggested the idea of The Swan in the McMoon Audition Scene.
131. Mr Malynicz relied on this as an example of Ms Kogan playing an important part in the development of FFJ's character. We have already noted that Mr Martin accepted that Ms Kogan contributed the notion that FFJ was "into loveliness" in music, which was an idea which informed his writing of the script. The judge does not mention this common ground in [83] of his judgment where he lists the "ideas" contributed by Ms Kogan. He also does not mention Mr Martin's concession in the following extract from his cross-examination:

"What she did was make the observation that Florence was into loveliness, useful. She said some observations about the faults – the problems with Florence's voice. And in some respect how they reflected her character. For instance, Florence she is never quite sure that she is going to be able to reach the big note at the end so she tends to pause a little bit. There is a little indication there that's not quite as confident as she makes out. So that's useful."

132. It is true that the judge does refer to the suggestion of using The Swan in the McMoon Audition Scene, but this is to strip the suggestion of its dramatic significance.
133. Ms Kogan's role in the development of FFJ's character, if it was in the course of a collaboration, needed to be placed in the scales together with the other evidence to assess whether the combined weight met the threshold for joint authorship.

Melotone (Homophone) Records

134. In this part of the screenplay FFJ, Bayfield and McMoon go to the premises of Melotone Records where Florence pays to have a record made of her singing.
135. The judge only accepted that Ms Kogan suggested "having Florence visit Melotone Records". Mr Malynicz submitted that this significantly undervalued Ms Kogan's contribution as demonstrated by documentary evidence.
136. The original version of this part of the screenplay had the record company offering FFJ a recording contract. In a Skype chat in March 2012, however, of which a record exists, Ms Kogan persuades Mr Martin to change this part of the story so that FFJ "pays to have these recordings made to please her false friends". This was more in keeping with what would have happened, and with FFJ's character, and showed FFJ taking charge, rather than being in receipt of an offer from a recording company. It also poignantly showed how driven she was, that she should take such an initiative in the absence of her partner, Bayfield, upon whom she had hitherto so much relied.
137. We think that the judge has undervalued Ms Kogan's contribution to this part of the screenplay, which consisted of a significant change to the storyline. He ignored the contemporary documentary evidence about Ms Kogan's contribution, as well as the extent to which it showed expressly that there was collaboration. Perhaps most importantly he failed to appreciate the significance of the plot change to the development of FFJ's character.
138. Again, therefore, if this plot development came forward in the course of a collaboration to create the work, it needed to be placed in the scales to see whether it was sufficient to qualify Ms Kogan as a joint author.

Summary on grounds 3-5

139. It must be recalled, firstly, that the specific instances dealt with above were put forward by Ms Kogan as examples, and not as the totality of her case. Secondly, the task for the court was to ascertain whether Ms Kogan's contribution was made in the course of a collaboration, and, if so, whether it crossed the relatively undemanding threshold for joint authorship set by Infopaq and subsequent authority.
140. Putting aside the judge's failure to make a finding on the issue of collaboration, we consider that the judge's summary of the extent of Ms Kogan's contribution failed to take account of a number of significant matters, perhaps most significantly Ms Kogan's contribution to the shaping of FFJ's character, and the plot of the screenplay. We do not think that it is possible to dismiss any of her contributions as not being authorial in nature.
141. It also seems likely that, in rejecting some of Ms Kogan's contributions as "minor plot suggestions" the judge had in mind his high bar for contributions of secondary skills to which he referred at [51]. We have already explained why, particularly in the case of a dramatic work, such an approach is wrong in principle. It may also be because he considered that they fell on the wrong side of the ideas/expression dichotomy to which he also referred at [51]. If the parties were collaborating, however, these were not suggestions of "mere ideas", but contributions to building the situation depicted in the scene, intimately tied into the development of the plot.
142. Despite these conclusions, it would be open to us to dismiss the appeal if we were persuaded that the failure to take proper account of the matters we have referred to would not affect the outcome of the action. We do not think that course is open to us, however. We consider that it is entirely realistic to suppose that a reconsideration of all the evidence would show that Ms Kogan's contribution was indeed made as part of a collaboration and passed the quantitative threshold for joint authorship. It would be wrong for us to go any further in the light of the disposition of the appeal which we will propose.

Ground 7: Acquiescence/estoppel defence of the film companies

143. The judge upheld the film companies' defence of acquiescence/estoppel on the limited basis that it would be unconscionable for Ms Kogan to rely on her joint authorship of copyright to restrict the public performance of the Film in any way. This was because her conduct amounted to a representation that the Film should be produced and released. The judge did not go further and hold that Ms Kogan was estopped from claiming a declaration as to her interest in the copyright, or from claiming remedies, such as financial remedies, which did not restrict the public performance of the film. The relief claimed by Ms Kogan was (i) a declaration that she was a joint author of the screenplay; (ii) a declaration that the claimants and the film companies have infringed her copyright in the screenplay; (iii) an injunction to restrain the claimants and the film companies from infringing her right to be identified as an author of the screenplay; (iv) an order that she be identified as a joint author of the screenplay; (v) an inquiry as to damages or an account of profits for infringement of copyright by the claimants and the film companies; (vi) interest; (vii) an order that that the judgment be publicised. Of these heads of relief, the only ones which could "restrict the public performance of the Film in any way" is (iii), as it could prevent the public performance of the film in a form which did not identify Ms Kogan as one of the authors of the screenplay. Copies of the film might have to be recalled. Before us, Ms Kogan confirmed, through Mr Malynicz, that she no longer claims that injunction, and no longer seeks to be identified as author except through an entry on the website IMDB, an order which would not restrict the public performance of the film. She therefore no longer seeks any relief on which the estoppel found by the judge could bite. There is no respondent's notice of the film companies asking this court to make the more extensive findings about acquiescence/estoppel which the judge declined to make.
144. In these circumstances we consider that this ground of appeal has become academic. It raises, if Ms Kogan is permitted to raise them, quite difficult arguments as to whether these home-grown defences are available in the light of Directive 2001/84/EC: see the analogous treatment in the case of trade marks in Marussia Communications (Ireland) Limited v Manor Grand Prix Racing Limited [2016] EWHC 809 (Ch); [2016] Bus LR 808. Those arguments are better resolved in a case in which they matter.

Consequences and conclusion

145. Mr Weisselberg reminded us of high authority enjoining appellate courts from becoming involved in detailed and complex re-assessments of factual findings made by a judge. That principle, he said, applies with particular strength to appeals from a court such as the IPEC, which aims to provide a less complex and less costly alternative to the High Court. He is obviously right to do so and we accept that submission in its entirety. However, on the basis of the grounds of appeal which we have found to be made out, the judgment cannot stand. The judge has adopted an erroneous approach to the evidence, failed to make important findings of primary fact, failed to take account of material matters and applied incorrect legal standards to the assessment of the sufficiency of Ms Kogan's contributions. In those circumstances, and with reluctance, we see no escape from the conclusion that there must be a retrial. The increase in costs to the parties is of course regrettable, but we can see no sensible alternative.
146. We will therefore allow the appeal and set aside the judge's declarations and other orders. We will order a new trial before a different judge in the IPEC who should be a full-time circuit or High Court judge. There should be an early case management conference before the new judge. We would not wish to bind the new judge as to the directions he or she will give, but we would express our views as follows:

i) We see no necessity for new pleadings or written evidence.

ii) The new trial should be on the basis that Ms Kogan (a) may rely on all of her contributions to the Screenplay, but (b) accepts that her contributions after the third draft were limited.

iii) The list of issues to be determined at the new trial should expressly include (a) whether there was a collaboration between the parties, and (b) the nature of that collaboration.

iv) The new trial should be on the basis that the defence of acquiescence/estoppel succeeded to prevent Ms Kogan interfering with the public performance of the film, but the film companies are not prevented from arguing, if so advised, that the defence should prevent other monetary and non-monetary relief. We say nothing about how promising that argument would be.

v) Ms Kogan should not be able to take advantage of the new trial to resurrect her unpleaded legal answer to that defence.