Spontaneous oral communications, impromptu speeches and fixation in copyright law: a comparative analysis

Spontaneous oral communications, impromptu speeches and fixation in copyright law: a comparative... The author Dr Eugene C. Lim is a Faculty Member at the School of Law, City University of Hong Kong. This article The ‘fixation doctrine’, which generally requires that a work be reduced to material form before copyright protection can arise, appears to have become firmly entrenched in the copyright laws of many common law jurisdictions. However, there remains some uncertainty as to whether fixation of spontaneous oral communications, such as speeches and interviews, can be performed by a person other than the ‘author’. By engaging in a comparative survey of the approaches to ‘fixation’ in the common law world, this article critically evaluates the extent to which Canada and Australia can learn from the experience of the UK and the USA in regulating authorship and entrepreneurial rights over initially unfixed spontaneous oral communications that are subsequently recorded by third parties. I. Introduction A young child narrates an experience in class; the oral narration is recorded by the teacher and published commercially many years later with the teacher’s permission.1 A famous classical musician gives a series of interviews to a magazine journalist; the interviews are recorded and reproduced in print many years later as part of a book.2 Although these real-life events involve disparate characters in different settings, they raise a common question: who owns the rights to spontaneous oral communications that have been recorded by a party other than the speaker? This is an important question that concerns the scope of the ‘fixation’ doctrine in copyright law in relation to ‘works’ that owe their existence to an individual who expresses them, and a different individual who reduces them to material form. Where intellectual property rights are concerned, copyright law in the ‘Anglo-American’ tradition generally prescribes that a work must satisfy both the ‘originality’ and ‘fixation’ requirements.3 In this regard, spoken words that are transcribed into material form can be protected by copyright law as ‘literary works’ in many common law jurisdictions if they constitute sufficiently original expression reflecting creative choices or skill and judgment from the author,4 and if they convey knowledge, information or pleasure.5 What is less clear, however, is the identity of the ‘author’ if the speaking (the ‘expression’) and the transcription (the ‘fixation’) are done by different parties. Should the author be considered the speaker, or the person who reduces or ‘fixes’ the spoken words into material form? Would it matter whether there is any agency relationship between the speaker and the transcriber? These questions have significant implications for the extent to which copyright can be used as a tool with which to restrain the publication of recorded oral communications. This article will focus on the challenge of identifying the ‘first author’ of recorded spontaneous expression, and not on (the somewhat broader) issue of whether fixation should be retained as a requirement for the subsistence of copyright. There is currently no general consensus at the international level6 or even within the ‘common law world’ on the issue of whether copyright subsists in spoken words that are not personally recorded by the ‘speaker’. Unlike the copyright legislation in the USA and the UK, the copyright statutes of Canada and Australia do not contain explicit rules on whether spontaneous oral communications have to be recorded by the speaker for copyright to arise. In addition, the Canadian and Australian statutes are silent on who owns the copyright to a transcribed oral work if the transcription is not done by the speaker. Interestingly, there is some jurisprudence in Canada and Australia which suggests that ‘speakers’ who do not transcribe or record their casual (unstructured) oral utterances will have no claim in copyright for their uttered words. A requirement that speakers transcribe or record their own words to assert authorship over them would stand in stark contrast to the statutory approaches adopted in the USA and the UK, both of which recognize, albeit to slightly different degrees, that the acts of composition and fixation of oral works can be ‘decoupled’, and that they need not be part of the same ‘creative act’, nor performed by the same individual, in order for copyright to arise. This article will use the controversial Canadian case of Gould Estate v Stoddard Publishing (‘Gould Estate’)7 as a springboard to discuss the extent to which spontaneous oral communications can be protected under Australian and Canadian copyright law. Although Gould Estate is not a recent decision, it continues to be cited as one of a select few Canadian authorities on fixation,8 and unless it is expressly overruled, will continue to serve as a source of guidance to both authors and publishers of recorded oral expressions. Interestingly, there has not been much published material on the fixation doctrine in Canada, nor has much attention been paid to the limitations of Gould Estate as a ‘putative’ case authority on the material form requirement. This article seeks to build upon the current sparse literature on fixation in Canada by making two contributions. First, it seeks to clarify the somewhat ambiguous judicial position in Canada regarding the fixation requirement and to highlight common challenges faced by both Canada and Australia in regulating authorship rights in spontaneous oral works. Secondly, it seeks to highlight the doctrinal and social benefits of introducing a provision to regulate authorship and ownership rights in works that are fixed by third parties, and the valuable insights that Canada and Australia can derive from the experiences of their ‘immediate peer group’9 in the common law world—in the context of this article the USA and the UK. In developing the above points, this article will argue that Gould Estate introduces significant doctrinal uncertainties into the concept of fixation in Canadian copyright law by implicitly supporting a rule of ‘self-fixation’, that is, a requirement that a speaker (in this case the deceased Canadian musician Gould) fix his own oral expressions to assert ‘first authorship’ over them. The article will observe that the presence of this implicit rule of ‘self-fixation’ might also be inferred from Australian case law on authorship. By highlighting common themes in their treatment of spontaneous oral works, this article will identify shared challenges that Australia and Canada face in the legal regulation of proprietary interests in initially unrecorded spoken words. It will demonstrate that a concept of ‘self-fixation’ for oral works is out of touch with the approach adopted in other parts of the common law world, and is arguably inconsistent with the fundamental principles of copyright law. The article will argue that the copyright statutes of Canada and Australia would benefit from clarificatory provisions, of the sort found in the US and UK legislation, prescribing the circumstances under which authorship can be asserted by individuals whose (original) oral works have been recorded by other parties. Finally, the article will conclude by highlighting some of the residual concerns and uncertainties relating to ‘third-party fixation’ that would benefit from closer attention by Canadian and Australian policymakers in future copyright reform efforts. II. Whose work is it? Authorship, expression and the fixation of spontaneous oral works in the UK, the USA, Canada and Australia It is a commonly accepted principle, particularly in jurisdictions following the Anglo-American legal tradition, that a literary work, which consists of words that make up original expression, must be recorded in some tangible medium in order for copyright protection to subsist in that work. This requirement is known by different names in various common law jurisdictions—it is predominantly referred to as ‘fixation’ in the USA and Canada, and ‘reduction to material form’ in many parts of the British Commonwealth.10 The fixation requirement has often been justified on the basis that copyright protects expression and not ideas, and expression in a tangible form is the primary means through which the structure, existence and boundaries of a work can be recognized and protected by the law.11 That being said, the physical act of fixation is not necessarily intertwined with original expression to the same degree in all classes of work. In the case of an artistic work, such as a portrait or painting, the expressive act (which triggers copyright protection) is inextricably connected with the act of fixation, since there can be no painting if the paint is not applied to the ‘surface’ in question.12 In other words, a picture that is imagined by a person is not an artistic work until it is given shape and form that can be perceived visually as an artistic composition in a relatively stable medium. However, in the case of a literary work, the composition of the work in the mind of its ‘author’ (which constitutes the original ‘expression’ attracting copyright protection) can be ‘fixed’ by another person (such as a scribe) who plays no role in the selection or arrangement of words in the preparation of the work but who merely gives those words material form by recording them on a transcribed surface or template. This is particularly the case if an ‘author’ delivers an extempore, impromptu speech, or utters words in response to interview questions that are taken down, verbatim, by other individuals who exercise little or no creative choice over their arrangement. When dealing with the scope of copyright protection for oral works, therefore, it is necessary for two important issues to be considered: first, whether transcription by a third party can satisfy the fixation requirement, and second, whether the speaker is deemed to be the first author (and copyright owner) of the transcribed words. The above two issues are addressed explicitly in the Copyright, Designs and Patents Act (CDPA) of the UK and the federal Copyright Act of the USA. Section 3(3) of the CDPA states that a work can be fixed by a person independent of the speaker, and the speaker will still be considered the author.13 This suggests that, for oral works in the UK, composition and fixation can be ‘decoupled’, and need not be part of the same act of creation.14 Under UK copyright law, it matters not by whom the oral work was fixed or whether the work was fixed with the consent or authority of the speaker, as long as fixation has occurred.15 On the other hand, Title 17 of the United States Code requires that the work must be fixed ‘by or under the authority’ of the author.16 Hence, although US federal law allows for a speaker to claim authorship over spoken words transcribed by other parties, the circumstances under which such authorship can be asserted are more limited. In particular, words that are recorded without permission in the USA would fall outside the ambit of the fixation provision and consequently would not attract copyright protection. The copyright statutes in Canada and Australia are notably ‘out of step’ with those of the UK and the USA, in that the Canadian and Australian Acts are both bereft of explicit clauses governing fixation of oral works by third parties. The Copyright Act of Canada does not even contain a general fixation provision, although fixation is referred to as a requirement for certain classes of work.17 In addition, references to unfixed works (such as performances) in the Canadian Act suggest that the fixation requirement is not uniformly applied to all categories of protectable subject-matter.18 However, in the Canadian case of Canadian Admiral Corp v Rediffusion, it was held that for copyright to subsist in a work, ‘it must be expressed to some extent at least in some material form, capable of identification and having a more or less permanent endurance’.19 Further, although the Canadian Act accommodates the possibility that certain forms of oral works (such as structured speeches) may be protected by copyright, it unfortunately does not provide much explicit guidance on whether oral works can be fixed by individuals other than the author-speaker in order for copyright protection to arise. There are interesting parallels between the Australian and Canadian copyright models. Although certain aspects of Australian copyright law have come under review in recent years by the Australian Law Reform Commission, the issue of ‘third party fixation’ appears to have been relegated to the sidelines by other concerns such as the adequacy of fair dealing and statutory exceptions to copyright in the digital economy.20 In Australia, there is currently no statutory provision expressly governing the fixation of oral works by third parties. However, section 22 of the Australian Copyright Act 1968 does state that, in the context of a literary work, the work is made when it is ‘first reduced to writing or to some other material form’. There is nevertheless some ambiguity in another section of the Act, ie section 32, which requires an author to be a ‘qualified person’ in Australia during at least a substantial period over which the work was made.21 This leaves open the possibility that the work could have been ‘made’ by a person other than the author. Since ‘making’ is defined in section 22 as the time at which the work was reduced to material form, the language of section 32 appears broad enough to entertain the possibility that the person who does the transcription (or reduction to material form) does not necessarily have to be the ‘author’ of that particular work in order for copyright protection to arise. However, Australian jurisprudence appears to disapprove of such a broad interpretation of authorship by seeming to require that composition and fixation be part of the same ‘creative act’. In Telstra Corporation Ltd v Phone Directories Co Pty Ltd, Gordon J, citing IceTV Pty Ltd v Nine Network Australia Pty Ltd (IceTV)22 defined ‘authors’ as individuals ‘who bring the work into existence in its material form’ or those who ‘gather or organize the collection of material and who select, order and arrange its fixation in material form’.23 It has been suggested that the High Court of Australia judgment in IceTV has arguably raised the standard of authorship necessary for copyright to subsist in original literary works.24 Further, subsequent cases in the ‘post-IceTV’ era have resulted in the ‘fragmentary development’ of the law, giving rise to uncertainty about the prerequisites for the subsistence of copyright in literary works.25 Despite the somewhat conservative stance adopted by the Australian courts in Telstra and in the post-IceTV cases with regard to authorship, it has been suggested that an interpretation which links authorship to fixation is not clearly supported by the legislative history behind the ‘making’ provision in Australia.26 Interestingly however, it may be observed from the current provisions of the Act that there are no explicit guidelines concerning the circumstances under which a person may assert authorship over an oral work that has been reduced to material form by another party. As is the case in Canada, the Australian copyright framework leaves many key questions unanswered about the legal relationship between the concepts of original composition, authorship and fixation in respect of spontaneous verbal and oral communications that have not been transcribed by the ‘author-speaker’ in question. III. The Gould Estate case and the oddity of a ‘self-fixation’ requirement This section will use the Canadian case of Gould Estate to illustrate the problems surrounding a ‘self-fixation’ requirement for oral works in copyright law. Although the case is Canadian in origin, it raises important themes that are equally pertinent to the Australian copyright model. By engaging in an analysis of the court’s holding in Gould Estate, the article will highlight shared doctrinal challenges faced by both Canada and Australia in regulating copyright over spoken words that are not fixed by the speaker. The decision in Gould Estate concerned whether an interviewed celebrity had a valid copyright claim over spontaneously uttered responses to interviews that had been conducted many years earlier, in the spring of 1956, for a magazine article, which was eventually published under the heading ‘I Don't Think I'm At All Eccentric’. The celebrity in question, noted Canadian classical musician Glenn Gould, had given a series of interviews to a freelance magazine journalist and photographer by the name of Jock Carroll while on a planned holiday in Nassau, without having prepared for or transcribed the interview answers in advance. Gould’s interview answers were therefore spontaneous and ‘unfixed’ at the precise moment they were uttered. However, they were transcribed and recorded almost instantaneously by Carroll, who took copious notes and made extensive tape recordings of the conversations. In addition, Carroll took numerous photographs of Gould for the magazine article. Many years later, after Gould’s death, Carroll sought to include the photographs and interview answers as part of a book on Glenn Gould, entitled ‘Glenn Gould: Some Portraits of the Artist as a Young Man’, which also contained original commentary by Carroll and captioned photographs depicting selected moments from the musician’s life. The estate of Glenn Gould asserted copyright27 over the interview answers, photographs and other elements in the book and sought to restrain its publication. The court had to decide whether Gould’s estate or Carroll (the transcriber of Gould’s spoken words) held the copyright in respect of Gould’s spontaneous oral communications during the interview. It was held, by both the Ontario Superior Court of Justice and on appeal to the Ontario Court of Appeal, that Gould’s estate had no copyright claim in respect of the interview answers as they were merely casual, unstructured utterances that were not intended to attract any proprietary interest.28 Carroll was considered the author of the book, including the interview excerpts, photographs, and other literary elements, which taken together as a whole, constituted an original work attracting copyright protection.29 Further, the court agreed that once Gould had consented, without limitation, to be the subject of a journalistic piece, he (or his estate) did not have any proprietary claim over the finished product, nor could he restrain any further reproduction of the photographs or prevent the journalist from producing additional material about him.30 Since Carroll had received an assignment (from the magazine publisher) of all rights in the magazine article published in 1956, he was the sole owner of copyright in the entire book (published years later).31 Although the issues in Gould Estate could, in the words of Finlayson JA, be decided on the basis of ‘conventional principles relating to copyright’,32 and that there was no need for recourse to the relatively new tort of ‘misappropriation of personality’, the ‘intellectual property’ analysis regarding oral communications provided in the judgment of the Ontario Court of Appeal is surprisingly brief.33 There is considerable emphasis given to the fact that there was no contract, express or implied, between Carroll and Gould and to the fact that Gould had ‘clearly consented’, without limitation, to being photographed and interviewed in Nassau.34 It was also highlighted that Gould could not have been said to have commissioned the photographs of himself, since Gould had not paid for the photo-taking sessions with Carroll, who was in any event not a photographer for hire in the ordinary commercial sense.35 It is submitted that the copyright issue for the photographs, captions and commentary in the book is relatively uncontroversial, since in the absence of any contract to the contrary, Carroll would be the first owner of copyright in any original material that he had personally written or produced. These would include the photographs (which Carroll had composed and taken), the descriptions thereof, and the written reflections, all of which constitute Carroll’s original expression. What is more problematic from the perspective of copyright law is the legal entitlement to use the selection of words spoken by Gould, which did not ‘originate’ from Carroll in the authorial sense. With the aid of recording technology to back up any ‘slips’ in stenographic skill, and even if he had been ‘slavishly’ transcribing words without the assistance of a machine, Carroll could not have been said to have been exercising ‘creativity’ or ‘judgment’ in merely transcribing words uttered by another individual. Although the judgment as a whole is illuminating, the two points raised by Finlayson JA in relation to the interview transcript regrettably do not provide much substantive guidance on the fixation debate in Canada. Interestingly, although Gould Estate has been cited as an authority on fixation in the secondary literature,36 the word ‘fixation’ is not once mentioned in the judgments of either the Ontario Court of Appeal or the Ontario Court of Justice. One point raised by Finlayson JA, albeit with some degree of caution and scepticism, relates to the relaxed, spontaneous nature of the setting in which Gould gave the interviews. In this regard, it was noted, from the transcript of the holding by the motions judge, that Gould was not delivering a ‘structured lecture’ or dictating to Carroll, but rather was engaging in ‘easygoing conversation’.37 The second point raised is less clear and perhaps somewhat troubling. Finlayson JA asserted, without providing much by way of analysis, that it was ‘evident from this record’ that Gould did not have a copyright with respect to the transcriptions of his oral utterances.38 Without clearly distinguishing between the portions derived from Gould’s oral ‘contributions’ and the elements composed by Carroll, the court simply ruled that Carroll was the author of the ‘text and captions’ in the biographical account of Gould’s life and therefore was the copyright owner of the written material in the book. In particular, the basis of Carroll’s copyright entitlement to use Gould’s words is not articulated or fleshed out in the judgment; it is regrettably an unresolved question whether Carroll had become the author of the transcribed words because he had ‘transformed’ them into original expression of his own, or whether Gould’s estate had an inferior claim to the words as copyrightable subject-matter because Gould had not personally written them down in material form at or prior to the interviews. If the latter interpretation is adopted, it would bring Canada dangerously close to a doctrine of ‘self-fixation’ in copyright law. The existence of a ‘self-fixation’ requirement for casual oral communications39 in Canada appears to find some indirect support in the secondary literature, with two leading commentators suggesting that Gould’s oral statements did not attract copyright protection because ‘there was no fixation’40 (presumably by Gould himself). In relation to the first point raised by Finlayson JA, it is highly debatable whether ‘structure’ should be the determining factor in determining whether a spoken work constitutes protectable literary expression. There is a very fine line between formal and casual expression, and the distinction between the two is often a question of degree. Section 2 of Canada’s Copyright Act specifically mentions ‘lectures’ as an example of subject-matter that can qualify as an original (literary) work, and the section goes on to stipulate that the term ‘lecture’ includes addresses, speeches and sermons (corresponding to the French term ‘conférence’). In this vein, Finlayson JA emphasized in his judgment that some forms of ‘structured’ oral expression, such as lectures and sermons, could be protected, but not others. Yet there are significant dangers in assuming the presence of a ‘bright line’ between structured and unstructured works that have been reduced to material form. Valuable works of literature may be excluded from the pantheon of copyright protection if the ‘structured/unstructured’ dichotomy is applied too strictly. Some of the most enduring works in English literature, such as James Joyce’s Ulysses, do not conform to a ‘formulaic’ narrative structure. To some critics, Joyce’s work may read more like a spontaneous ‘stream of consciousness’ than a work with a clearly delineated narrative.41 Although Joyce’s novel was not a ‘speech’, if Canada is willing to allow copyright protection for certain forms of speeches but not others, then ‘structure’ alone provides a highly unsatisfactory basis for determining what to exclude from protection. There is a concomitant danger that judges may be making assessments of literary merit when they exclude works based on how ‘structured’ they are. Regardless of whether one is considering a written work or a transcribed oral work, the test for whether copyright subsists in the subject-matter should be driven by established principles of originality, rather than by the judge’s perception of ‘structure’ per se. Although the eventual decision by the Ontario Court of Appeal was probably justified in the circumstances, the analysis is unsatisfactory. The judgment is laudable in that it highlighted the public interest of allowing Carroll’s work to be published and circulated, thereby providing valuable insight into the life of a celebrated musician. However, the reasoning vis-à-vis ‘ownership’ of Gould’s words is awkward from a doctrinal perspective, and the same public interest considerations could have been achieved through the use of copyright defences or other doctrinal mechanisms validating Carroll’s use. The primary concern with the judgment relates to the manner in which the ownership of copyright in relation to Gould’s spoken words was dispensed with. The relationship between originality and fixation should also have been ‘disentangled’ in the judgment. A plausible argument could be made that despite their lack of ‘structure’, Gould’s interview responses were original literary works that not only reflected his choice of words, but also bore his ‘personal stamp’ as an author, revealing dimensions of his personality that would convey knowledge, information and pleasure to their intended audience. There is some authority from outside Canada, such as the Dutch case of Endstra Tapes,42 which suggests that spontaneous oral conversations can constitute works of authorship if they are the result of human creative choices.43 As for the fixation issue, whether Gould’s words were ‘fixed’ should not be a subject of any dispute, bearing in mind Carroll’s extensive ‘recording activities’ on site. Since it might be argued that Gould’s oral statements were both original and recorded in material form, one would need a plausible basis upon which to deny Gould copyright protection over his expression. One way of doing this would be to adopt an explicit ‘self-fixation’ requirement, which is merely implied in the judgment. However, in the absence of such a requirement, a more doctrinally coherent way of justifying Carroll’s copyright ownership in the work would be to employ the principle of ‘transformative use’. By selecting portions of the quoted statements and then commenting on them, and embellishing them with insights of his own, Carroll had effectively transformed the text containing portions of the interview transcript into an original literary work of his own.44 However, even original ‘derivative’ works can infringe if they borrow a qualitatively significant portion of text from another source.45 Nevertheless, if Carroll had only taken reasonable portions of text from the interview transcript for his book, he might have been able to avail himself of a fair dealing defence under Canadian copyright law.46 ‘Fair dealing for the purpose of criticism or review’, which is enshrined in section 29.1 of the Copyright Act of Canada,47 would provide Carroll with some shelter against liability for infringement of any copyright that might have subsisted in Gould’s oral responses, provided of course that Carroll adds insight of his own to the quoted statements, and includes clear attribution to the source of the interview answers in the published version or content. A further way of justifying a decision in favour of Carroll might be to imply the presence of a licence to use the interview transcript in future publications. Although there was no contractual agreement between Gould and Carroll, Gould had specifically invited Carroll to join him at the location in Nassau and acquiesced to the recording of notes for the purpose of writing a story about him for promotional purposes. Since there were no limitations placed on the extent to which the interview could be used in future publications, the unqualified ‘consent’ implied in this case might have provided Carroll with a defence against a copyright infringement claim. It is not entirely clear that Gould’s estate would fare any better under the Australian copyright system, if the facts were to occur today on Australian soil. Assuming, hypothetically, that Gould was a ‘qualified person’ in Australia, the primary stumbling block standing in the way of Gould’s claim to authorship would be the Australian judiciary’s interpretation of the fixation requirement. Bearing in mind that authors are defined as individuals who ‘bring the work into existence in its material form’,48 it would be difficult to argue that Gould had taken steps to ‘gather or organize the collection of material’ and to ‘select, order and arrange its fixation in material form’.49 An interesting parallel between the Australian case of Telstra and the Canadian case of Gould Estate is the weight given to ‘structure’, or ‘organization of material’ in the Australian context. In this respect, Gould could hardly be said to have been ‘organizing’ the transcription of his words, since he had neither hired Carroll to do the recording of his interview responses nor supplied him with the means or equipment to do so. Gould’s participation in the recording process can be more accurately likened to ‘passive acquiescence’ than active gathering of data. Even if it could be argued that Gould was ‘ordering’ the sequence of thoughts in his head, he would not be the party performing the act of fixation in material form. In this regard, Carroll’s role appears to be more akin to that of a ‘gatherer’, if not for the original ‘slavish’ transcription, then in the eventual compilation and selection of quotations for his published book. The acts of transcription and expression appear to be treated as an indivisible whole by the Australian courts, which echo the submerged assumptions driving the decision by the Ontario Court of Appeal in the Gould Estate case on the rights over recorded spoken words. Although Gould Estate is a case that governed casual oral utterances, the issues raised in that dispute have larger ramifications for the regulation of rights in (initially) unfixed spontaneous expressions that are subsequently recorded by third parties. The copyright statutes of Canada and Australia share a common weakness in that they do not specify with sufficient clarity the identity of the author in such situations. As demonstrated in the preceding analysis of the Gould Estate case, the Canadian and Australian jurisprudence appear to endorse the notion that fixation and expression are indivisible parts of the same creative act for oral works of a literary nature. While such a principle would arguably be justified for artistic works, it would create difficulties for literary works that are generated ‘spontaneously’ by an orator. The next section of the article will canvass some of the challenges that a doctrine of self-fixation would pose if applied uncritically to all works of creative expression. IV. A return to first principles: protecting original expression rather than mere labour in copyright law One of the arguments raised in the previous section relates to how a self-fixation requirement provides a highly unsatisfactory basis for discriminating between protectable and unprotectable oral expression. Although Gould could easily have recorded his interview answers as he was uttering them, the argument against imposing a ‘personal fixation’ requirement by an author could also be made in respect of works produced by individuals who can only (or choose only to) generate creative literary or musical output through unfixed, spontaneous performance. A gifted writer who weaves together tales of great intrigue by reciting lines of prose that are meticulously composed in his head, or a musician who can only produce stirring melodies from unwritten scores performed from memory, are examples of creators who would fall within this category. Without a recording of these works, the creative input of these spontaneous ‘creators’ might be lost forever. Although Gould Estate provides only one instantiation of how oral works can lead to disputes over intellectual property rights, the way in which the fixation issue is dealt with by courts and policymakers has significant implications for the regulation of rights in works that are ‘fixed’ by individuals other than those responsible for their first expression. The concern here is not that third-party transcribers can acquire authorship status by engaging in transformative use of recorded works, thereby creating original derivative works. Transcribers, such as Jock Carroll in the Gould Estate case, who exercise skill and editorial selection in producing derivative works reflecting their own creative choices, can and should be recognized as authors. However, the question here is whether speakers whose expression is recorded should be granted any rights in the recorded expression as literary works. If the law does not confer copyright on speakers in respect of their original oral works, but allows transcribers to assert claims over carefully edited selections of those recorded oral works, it runs the risk of creating a situation where an amanuensis is treated as the ‘default first author’ of expressions originating from another party. The interpretive ambiguities in the Canadian and Australian copyright statutes vis-à-vis third-party fixation might give rise to just such a scenario. A clearly delineated legal position identifying the author in cases of spontaneous expression would therefore help to obviate some of the uncertainties raised in Gould Estate and in the examples described in the preceding paragraph. Under the current vagaries of the Canadian and Australian jurisprudence, spontaneous ‘creators’ may, in some cases, find their identity as authors ‘usurped’ by individuals who meticulously and conscientiously ‘arrange for’ and perform the act of recording the work in material form. In other cases, such creators may find that their authorship of the oral work is not recognized by the law, even if it has been recorded by a third party. Imposing an implied ‘self-fixation’ requirement (à la Gould Estate) on authors would effectively deny ‘spontaneous creators’ any copyright in their work unless they fix it first. Unless we are dealing with subject-matter where the fixation is intimately connected with the act of expression (as in photography or visual art), granting first authorship of recorded spontaneous creative work to third-party recorders and editors, while disregarding any claim of the spontaneous ‘composer’, would severely undermine the doctrine of originality in copyright law. The test for whether a person can be considered an ‘author’ is strongly connected with the test for originality.50 A key criterion for authorship lies therefore in whether the expression contributed is of the ‘right kind and quantity’.51 In view of the strong connection between authorship status and originality, it would arguably be odd to disregard a speaker’s expressive contribution to a recorded speech, while recognizing a transcriber’s editorial contribution to a revised version of the same. V. The social utility of a third-party fixation clause The uncomfortable doctrinal oddities outlined in the preceding section suggest a need for statutory clarification on the identity of the first author in the case of (initially unfixed) oral works and other forms of spontaneously-generated content in the Canadian and Australian contexts. For simplicity, the term ‘third-party fixation’ or ‘TPF’ clause will be used in this context to refer to a statutory provision that stipulates the circumstances under which a work can be fixed by a person other than the author, in order for first ownership of copyright to vest in the author. This section of the article will discuss a number of ways in which an explicit TPF clause would help to promote social utility, and contribute to the preservation and dissemination of culturally valuable works that might otherwise be lost or forgotten. An important social benefit of having a TPF clause in copyright legislation is that it would provide clearer guidance and coordination to authors and publishers on the extent to which creators of initially unfixed spontaneous expression can restrain its subsequent publication by third parties. By identifying the first author of TPF works, such a clause would facilitate the negotiation of licensing, recording and profit-sharing arrangements between authors, publishers and broadcasters. The legal recognition of their rights in spontaneously-generated creative expression would encourage prospective authors of oral and unfixed works to enter into recording and distribution contracts with third parties, thereby ensuring that their intellectual legacy survives in some recorded form. Having a legal framework in place to facilitate the licensing of rights in relation to TVP works would incentivize both the creation and the recording of spontaneous expression. First, authors in whom copyright status is vested would be in a stronger bargaining position to dictate the conditions under which their work can be shared and disseminated in return for financial remuneration or other consideration, thereby increasing the likelihood that they might authorize the recording of their work by third parties. Second, media enterprises, publishers and individuals who are commissioned by the author, either before or after the fact of recordation, to assist in the dissemination of the work would have an incentive to maintain and preserve the work in a form which is suitable for public access. In the latter scenario, the incentive to record, preserve and disseminate the work would arise principally from the financial benefit derived by a third party from a recording or licensing contract with the author. The twin incentives operating on both sides of the ‘author-transcriber’ equation for commissioned oral works would facilitate the creation, dissemination and transmission of the works to the general public, thereby contributing to the growth and advancement of knowledge and culture in the longer term future. It is of course important to note that not all authors are necessarily eager to ‘showcase’ and disseminate their spontaneously-generated works to the general public. It is equally important to emphasize that in the US copyright system, unfixed spontaneous works that are subsequently recorded by third parties without authorization would not generally attract copyright protection. Hence, in such a scenario, where copyright protection is not available, unauthorized recordings might be shared and disseminated against the author’s wishes. However, it is submitted that even in the US context, a TPF clause plays an important role in enhancing social utility by clarifying and coordinating the dynamic interplay of rights enjoyed by creators of spontaneous content, publishers and third-party users. By providing public notice of the distribution of rights and responsibilities concerning the ownership and use of recorded oral works, an explicit TPF clause can help to reduce the incidence of litigation and legal disputes arising from an uncertain legal climate. Understanding the conditions under which the material form requirement must be satisfied in order for copyright to arise would also facilitate risk management by would-be authors in tailoring, adapting and arranging for the mode of fixation to suit the particularities of the copyright regime in which they produce their oral works. For instance, in a US-style TPF regime, an author wishing to limit access to a work could benefit from the guidance that the US TPF clause provides by taking risk management strategies to ensure that the expression is only made accessible to commissioned persons for the purpose of recordation, and shielded as far as possible from unauthorized third parties. The benefits associated with the ‘coordinating’ function of a TPF clause would also apply to third-party use of recorded oral expression. By providing clearer guidelines on when recorded expression falls within the scope of copyright protection, and when such expression enters directly into the public domain, a TPF clause can perform an important public notice function in helping users of information distinguish between legally permissible and infringing use. Fostering a climate of greater legal certainty and transparency with respect to the allocation of rights in TPF oral works would also have the added advantage of facilitating and legitimizing the broader dissemination of unprotected works to members of the public. In this regard, by providing guidance on the extent to which initially unfixed oral works are protected by copyright law, and when such works enter into the public domain, an explicit TPF clause would play an important role in circumscribing the circumstances under which dissemination of the works to the public would be legally permissible. Making such recorded oral works available to the public in a manner allowed by copyright law would, in turn, stimulate innovation, cultural growth and development by creating a more permissive and conducive environment for the production of derivative works, parodies and user-generated content, and contribute to the flourishing of creative expression, thereby promoting discourse and communication in contemporary civil society. Critics who are opposed to the idea of ‘decoupling’ expression and fixation might argue that granting ‘speakers’ copyright protection over their uttered words could have a chilling effect on ordinary discourse and communication by restricting the republication of those words by third parties. However, it should be borne in mind that, as for other categories of protected expression, copyright does not confer an absolute monopoly over TPF works, and is subject to important limitations and defences, such as fair dealing or fair use.52 In particular, an appropriately construed ‘users’ rights regime’ would help to ensure that reasonable limited uses of oral works can be made even during the term of copyright protection. Quoting selected excerpts from an interview that are subsequently commented upon in a biographical work of a well-known pianist would, for instance, potentially qualify for the fair dealing provision for criticism or review. Likewise, a newspaper journalist who reproduces a reasonable amount of text from a spontaneous oral speech delivered by a public figure for the purpose of writing an article on current affairs would enjoy some degree of shelter from liability under the fair dealing provision for news reporting.53 VI. Charting the way forward in clarifying the scope of a provision on TPF: future challenges Despite the social and doctrinal arguments in favour of an express statutory provision recognizing the authorial status of a ‘speaker-creator’, the statutory insertion of a TPF clause into copyright legislation is a challenge that should be approached with care and caution, taking into account the insights that might be gleaned from the US and UK experiences. This section of the article will highlight a number of residual concerns and remaining uncertainties in the US and UK approaches to third-party fixation—issues that would deserve closer scrutiny if Canada and Australia were to consider amending their copyright legislation to incorporate an explicit TPF clause. In updating their copyright legislation to address issues of third-party fixation, one potential issue that will likely arise in the Canadian and Australian contexts is whether authorization of fixation by the ‘speaker-creators’ of spontaneous or oral works should be made a necessary criterion for authorial copyright to arise in TPF works. This is a matter on which the legal positions of the USA and the UK diverge sharply. If Canada and Australia were to follow in the footsteps of the USA, and make authorization of fixation a requirement for copyright to subsist in TPF works, then the range of spontaneous oral works that would qualify for legal protection would be much narrower. In particular, impromptu speeches that are not fixed with the authorization of the speaker would fall outside the scope of copyright protection. The US approach to fixation would require not only that the author give consent to the recording, but also assumes an intention to create a work, since authorization to record expression necessarily entails a conscious desire to have that expression reduced to material form by a third party identified by the author. Such an approach would suggest that an intention to create a work serves as a prerequisite for copyright protection in the USA. This position has, however, been criticized in the European context as being too strict. In the Dutch case of Endstra Tapes,54 the Supreme Court of the Netherlands held that transcribed conversations that were the result of human creative choices, and that were neither banal nor trivial, could be considered works of authorship. It has been observed that a similar result would likely be obtained by applying the European standard of ‘the author’s own intellectual creation’, as articulated in the Infopaq case,55 to the facts of Endstra Tapes.56 In the European context, requiring an author to demonstrate a conscious desire to create a work would likely be viewed as unfairly burdensome.57 From this perspective, imposing too onerous a standard of proof relating to the mental state of an author, or on the purpose of creating a work, might arguably cause many otherwise original works to fall outside the scope of copyright protection. In view of the consequences of setting the authorship threshold too high, it is possible that important oral works of aesthetic value that are recorded without the permission of the ‘author’ or created without ‘authorial intent’ may simply be denied protection in the USA. Although this means that copyright would not be available as a tool to restrain the circulation of oral works recorded without authorization in the USA, it also means that would-be ‘authors’ who generate intellectual content spontaneously may be more reluctant to showcase their creative talent in venues where their expressions may be recorded by third parties. Further, denying a work copyright protection simply because the recording was not authorized by the person ‘expressing’ it quite simply ignores the latter’s intellectual contribution to the existence of the work. Any intellectual judgment and creativity that went into the work’s composition, even if done spontaneously, would, under the US copyright model, not be viewed as giving rise to copyright protection if the transcription had been done furtively or illicitly by an unauthorized third party. While the US approach to the authorship of initially unfixed TPF works may raise concerns about pitching the standard of ‘conscious creation’ too high, it is important to note that attempts at transplanting the UK model into Canada and Australia should also be approached with caution and careful forethought. If Canada and Australia decide to follow in the footsteps of the UK and enact a TPF clause styled after section 3(3) of the CDPA, some consideration should be given to the interpretive uncertainties surrounding the extent to which section 3(3) protects the neighbouring rights of transcribers and reporters in servile copies of recorded spontaneous works. In incorporating a UK-style approach, one of the challenges facing the drafters of an explicit TPF clause in Canada and Australia relates to determining the appropriate scope and nature of the neighbouring or entrepreneurial right (if any) to be granted to a ‘third-party fixer’ in relation to a mechanical transcript or record of a TPF work (as opposed to the underlying literary, dramatic or musical work in which creative copyright subsists and which vests in the author-speaker). The key issue here lies in elucidating the interplay of overlapping authorial and entrepreneurial rights that might subsist in an original oral work which is recorded through a mechanical process by a third party. If the third-party fixer has exercised original thought and selection in preparing the record, then the record may be considered an original derivative work in its own right. However, if the record is merely a servile rendering in material form of an initially unfixed oral work, an important question that needs to be answered is whether the mechanical record attracts a separate neighbouring right that vests in the transcriber, particularly if the record is reduced to a material form or tangible medium that is not expressly classified in the statute as copyrightable subject-matter. Although section 3(3) of the UK’s CDPA addresses the issue of neighbouring rights obliquely, there are unanswered questions and remaining uncertainties that will be discussed in the ensuing paragraphs. The CDPA currently recognizes eight categories of copyrightable subject-matter, four of which relate to original authorial works, and four of which relate to recorded or transmitted media in which entrepreneurial rights commonly subsist. The eight categories of copyrightable subject-matter are set out in sections 3 through 8 of the CDPA and comprise the following: literary works (including databases), dramatic works, and musical works (sections 3 and 3A), artistic works (section 4), sound recordings (section 5A), films (section 5B), broadcasts (sections 6 and 6A) and published editions (section 8). The latter four categories of subject-matter, namely sound recordings, films, broadcasts and published editions, are strongly associated with derivative content or formats, based on underlying works or expressions, that are prepared by entrepreneurial entities, such as recording companies, production companies, television stations and publishing houses. The possibility that overlapping authorial and entrepreneurial rights may subsist contemporaneously in recordings of TPF works is alluded to in the final part of section 3(3), which provides that ‘nothing in [section 3(2)] affects the question whether copyright subsists in the record as distinct from the work recorded’. Hence, literary, dramatic and musical works as defined in section 3 may be fixed by third parties in formats or media that are recognized by the CDPA, and the Act leaves open the possibility that derivative entrepreneurial rights in such recordings may be enjoyed independently by third parties. Although section 3(3) itself does not provide a list of permissible media on which TPF recordings may be fixed, the sections following section 3 provide specific categories of media in which copyright may subsist. One question that may be raised in this context relates to whether the CDPA provides a comprehensive and inclusive framework for media on which third-party fixation may occur. None of the above categories, for instance, relate to servile handwritten transcripts. As an example, it is not entirely clear whether a reporter who makes a handwritten copy of an impromptu speech delivered by a public figure, in a way that might be reminiscent of the classic case of Walter v Lane,58 will be granted any entrepreneurial or neighbouring rights under the CDPA in respect of ‘mechanical’ handwritten copies of the speech so produced. This hypothetical scenario presupposes that the reporter does not qualify for original authorship of a literary nature, but merely produces a ‘servile’ handwritten copy that seeks to be as faithful as possible to the oral work produced by the author-speaker. Although section 3(3) of the UK’s CDPA seems to allow for the possibility that neighbouring rights of third parties can subsist in recorded copies of initially unfixed expression, its language has been couched in relatively ambiguous terms.59 One commentator has interpreted the section to mean that section 3(3) does no more than leave open the question whether a reporter enjoys a separate (copy)right in his or her record.60 Whether or not a separate right is enjoyed would ultimately depend on whether the record falls within one of the recognized categories of protected subject-matter in the CDPA. Hence, if the reporter has made a sound recording, or a film recording, or a published edition of the recorded oral work, then the reporter’s record would be protected under the corresponding sections of the CDPA respectively. In these situations, the reporter’s neighbouring right in the recording or edition would subsist alongside the speaker’s authorial copyright in the literary work embodied in the recorded speech. However, what is less clear is whether handwritten transcripts (and transcripts fixed in other unclassified types of media) that are principally servile records of the transcribed (oral) work will qualify for concurrent protection as entrepreneurial works, since there is no specific category in the CDPA which recognizes reporters’ transcripts as protectable subject-matter per se. The above question has provoked an animated debate among UK copyright scholars, reflecting a range of views on how to treat the servile handwritten transcript as a possible candidate for (derivative) copyright protection. It has been suggested, for instance, that a record of a speech by hand qualifies as a ‘sound recording’, just as an audio recording would, under section 5 A of the CDPA.61 Another commentator has rejected this view, arguing that a sound recording must enable the reproduction of recorded auditory sounds, rather than the mere words of a literary work;62 and hence a handwritten, verbatim transcript of spoken words which falls outside the eight categories of subject-matter in the CDPA might not be eligible for any protection.63 In view of the limited reach of the eight extant categories of protected subject-matter, it has been proposed that a new category relating to ‘mechanical copyright’ in literary transcripts be introduced into the CDPA, so that verbatim handwritten records of TPF works can be protected alongside other entrepreneurial subject-matter such as sound recordings and typographical arrangements of published editions.64 In this vein, it has been observed that the need for copyright law to accommodate the protection of written transcripts would be particularly acute ‘where the result is a product of genuine cultural interest that might otherwise remain inaccessible to the public’.65 The extent to which entrepreneurial rights in recorded spontaneous works should be expressly protected by legislation is an issue that should be taken into account by Canadian and Australian policymakers when drafting a TPF clause that would be suitable for their domestic copyright systems. Handwritten transcripts constitute, of course, only one example of an entrepreneurial ‘product’ that is not specifically categorized in the CDPA as protected subject-matter. As technology develops, new forms of media will likely be made available in the future to facilitate the recording of information. The larger challenge of enacting a TPF provision à la section 3(3) of the CDPA lies, therefore, in providing clearer guidance on the types and categories of recording media that can potentially trigger neighbouring rights in third-party recordings of spontaneous oral works. In considering this challenge, a possible way forward might be to enact a TPF provision that is technology-neutral, which expressly elucidates the interplay of authorial and entrepreneurial rights in TPF material, and that is broad enough to encompass emerging forms of transcription and recording media in the digital era. Interestingly, some guidance can be derived from §102 of the US Copyright Act on copyrightable subject-matter in drafting a technology-neutral clause on fixation, by making reference to works and recordings that are ‘fixed in any tangible medium of expression, now known or later developed, from which they can be perceived, reproduced, or otherwise communicated, either directly or with the aid of a machine or device’ (emphasis added). If applied to the TPF context, a technology-neutral provision that reflects the general ambit of the quoted segment from §102 of the US Act would arguably be broad enough to encompass derivative entrepreneurial works that are produced by hand, as well as those recorded with the aid of a machine or other technology, now known or later developed. In learning from the experiences of both the USA and the UK with their own TPF provisions, it is accordingly important to recognize the twin challenges of stipulating the circumstances under which authorial copyright can subsist in TPF works and clarifying the range of permissible media through which third-party fixation can be effected for neighbouring rights to arise. In revising their respective copyright statutes to meet these twin challenges, incorporating an explicit TPF clause to regulate the interplay between authorship and entrepreneurial rights for TPF works in Canada and Australia would provide authors, reporters and other ‘entrepreneurial’ transcribers with more concrete guidance on the circumstances under which their efforts in producing and recording TPF works will be legally protected. A clearly defined statutory framework for the legal protection of overlapping authorial and entrepreneurial rights in TPF works would also play an important role in demarcating respective rights and responsibilities, in facilitating the negotiation of licensing agreements, and in providing third-party transcribers with incentives to reduce spontaneously-generated works to material form, thereby helping to ensure, at least in some small measure, that (initially) unrecorded works of cultural significance can be preserved and shared with others for the benefit of future generations. VII. Conclusion The doctrine of fixation continues to generate a fair degree of controversy in copyright law, and scholars opposed to a fixation requirement in copyright law might even suggest that the doctrine is facing an ontological crisis.66 For the time being, however, fixation appears to be firmly rooted in the copyright legislation of many common law jurisdictions, and is not likely to be removed as a requirement for the copyright protection of literary, musical and dramatic works in those jurisdictions, at least in the immediate foreseeable future. What is less clear is whether expression and fixation have to be done by the same party in order for copyright to subsist in spontaneously expressed creative output of a literary nature. This article has sought to highlight uncertainties in the copyright frameworks of Canada and Australia, both of which do not define with sufficient clarity the identity of the author when expression and fixation of oral works are performed by different entities. This article suggests that the lacuna in the copyright frameworks of Canada and Australia vis-à-vis the identity of the author in cases of ‘unfixed expressions’ can be remedied by means of a statutory provision, which expressly accommodates the possibility of third-party fixation for spontaneous oral works. Such a provision would clarify some of the ambiguities arising in Gould Estate and similar cases, and codify the principle that ‘self-fixation’ should not be a requirement for authors to enjoy copyright protection in their spontaneous oral expressions. A clarificatory provision on third-party fixation would help to foster greater doctrinal clarity in the protection of spontaneous oral expressions that are subsequently recorded, facilitate the negotiation of licensing and distribution agreements, and help to mediate the balance of power between authors, entrepreneur-recorders and subsequent users of those recorded spontaneous expressions. The in-built statutory limitations to copyright protection also serve as an important safeguard against excessive monopoly control over the recorded content, and ensure that reasonable uses can be made of the oral work even during the term of copyright protection, thereby facilitating the preservation and dissemination of valuable expression that might otherwise be lost, and enriching the public domain in the long term. Jock Carroll’s biography of Glenn Gould provides us with a rare and fascinating glimpse into the life of one of Canada’s leading classical musicians, and Gould’s own recorded words have, through Carroll’s efforts, been preserved as a precious testament to the musician’s life, personality and creative genius. A clearer, more doctrinally coherent analysis of the third-party fixation issue in the case would, nevertheless, have helped to solidify the interpretive legacy of this interesting decision on the materialization requirement in copyright law. Yet, despite its limitations, this case may provide valuable learning opportunities for jurisdictions such as Canada and Australia, and perhaps a focal point for reform as they continue to refine and update their copyright legislation for the modern information society. Footnotes 1 Gormley v EMI Records (Ireland) Ltd [2000] 1 IR 74. 2 Gould Estate v Stoddart Publishing Co [1998] 39 O.R. (3d) 545 (Ontario Court of Appeal). 3 See, for instance, ss 1(1)(a) and 3(2) of the Copyright, Designs and Patents Act (CDPA) of the UK, as well as 17 USC §101 (on the definition of a ‘fixed’ work) and §102(a) of the Copyright Law of the USA, Tıtle 17 of the United States Code. 4 Routine phrases and expressions commonly used in daily life are, on the other hand, less likely to be viewed as original. See Hector MacQueen, ‘“My Tongue is Mine Ain”: Copyright, the Spoken Word and Privacy’ (2005) 68 Mod L Rev 349, 363, where it is noted that routine words and catch phrases ‘with which we navigate our way through daily life may well fall short of even low standards of originality and literariness’. 5 See, for instance, Apple Computer v Computer Edge [1984] FSR 481 (FCA), 521, where it was held that in order for a literary work to convey information or instruction, it must be able to convey an intelligible meaning. 6 Art 2(2) of the Berne Convention for the Protection of literary and Artistic Works (1886) 828 UN Treaty Ser 221, to which most nations are signatories, provides considerable flexibility for members by stating that it shall be ‘a matter for legislation in the countries of the Union to prescribe that works in general or any specified categories of works shall not be protected unless they have been fixed in some material form’. See also Elizabeth White, ‘The Berne Convention’s Flexible Fixation Requirement: A Problematic Provision for User-Generated Content’ (2013) 13 Chi J Int’l L 685, 687, who describes the Berne Convention’s reference to fixation as being expressed in ‘non-binding and arguably unhelpful’ terms. 7 Gould Estate v Stoddart Publishing Co Ltd 39 OR (3d) 545; 161 DLR (4th) 321; 80 CPR (3d) 161; [1998] OJ No 1894 (QL); 114 OAC 178. 8 See Daniel J Gervais and Elizabeth F Judge, Intellectual Property: The Law in Canada (2nd edn, Thomson Canada 2011) 51, noting that there was no copyright in Gould’s unstructured oral responses at the time of the interview because there was no fixation. 9 See Elizabeth Adeney, ‘Authorship and Fixation in Copyright Law: A Comparative Comment’ (2011) 35(2) Melbourne Univ L Rev 677, 696, who also uses the expression ‘immediate peer group’ when comparing Australia’s approach to the notions of ‘work’ and ‘authorship’ to those of other jurisdictions, particularly those with which Australia has strong legal, cultural and historical connections. 10 See, for instance, s 3 of the UK’s CDPA and s 22(1) of the Copyright Act of Australia. 11 It is interesting to note, however, that some critics of the fixation doctrine have suggested that copyright should subsist in any ‘perceptible expression that emanates from the intellect of a person and is intended to appeal to the aesthetic sense or intellect of others, regardless of materiality’. See for instance David Brennan and Andrew Christie, ‘Spoken Words and Copyright Subsistence in Anglo-American Law’ (2000) 4 IPQ 309, 349, who argue that there is nothing inherent in the objectives of copyright law that requires the imposition of a material form requirement. Imposing such a requirement may, in their view, lead to outcomes contrary to the objectives of copyright protection. 12 Paul Torremans, Holyoak & Torremans Intellectual Property Law (8th edn, OUP 2016) 190. 13 S 3(3) prescribes, inter alia, that it is immaterial whether the work is recorded by or with the permission of the author. See also Lionel Bently and Brad Sherman, Intellectual Property Law (4th edn, OUP 2014) 92 as well as Torremans (n 12) 239 (noting that speeches can constitute original literary works once they are recorded, and that it is not necessary for speakers to do their own recording). 14 See Adeney (n 9) 677, 681. 15 But it is noteworthy that s 3(3) also mentions that it does not affect the question of whether copyright subsists in the record as distinct from the work recorded. 16 See §101 of the Copyright Law of the USA, Tıtle 17 of the United States Code. It ought to be borne in mind, however, that federal law does not necessarily pre-empt all aspects of state law. The basic principle enshrined in art VI clause 2 of the US Constitution provides that federal law will prevail over state law in the event of a conflict between the two. 17 As an example, the definition of ‘dramatic work’ in s 2 of the Copyright Act of Canada includes ‘any piece for recitation, choreographic work or mime, the scenic arrangement or acting form of which is fixed in writing or otherwise’ (emphasis added). In the case of musical works, the 1993 amendments of the Copyright Act removed the express fixation requirement for such works. See Gervais and Judge (n 8) 49–50. 18 See for instance section 15(1)(a) of the Copyright Act of Canada, which makes reference to copyright protection of unfixed performances. See also Adeney (n 9) 677, 689, where it is noted that ‘unfixed works’ fall within the concept of authored works in Canada. 19 [1954] Ex. C.R. 382, 20 C.P.R. 75 (Can. Ex. Ct) para 28. 20 See, for instance, Australian Law Reform Commission, Copyright and the Digital Economy, Report No 122 (2013) (published on 13 February 2014), <http://www.alrc.gov.au/publications/copyright-report-122> (accessed 23 April 2018). One of the key areas of inquiry in this report relates to whether the exceptions and statutory licences in the Copyright Act 1968 (Cth) are adequate and appropriate in the digital environment. 21 See s 32(1)(b) of the Copyright Act 1968 of Australia. 22 IceTV Pty Ltd v Nine Network Australia Pty Ltd (2009) 239 CLR 458 (High Court of Australia) (IceTV). 23 See Telstra Corporation Ltd v Phone Directories Co Pty Ltd (2010) 264 ALR 617, 624, [20], referring to dicta expressed by French CJ, Crennan, Kiefel, Gummow, Hayne and Heydon JJ in IceTV (n 22). 24 See Alexandra George, ‘Reforming Australia's Copyright Law: An Opportunity to Address the Issues of Authorship and Originality’ (2014) 37(3) UNSW L J 939, 941. 25 ibid. 26 See Adeney (n 9) 677, 680, 689–91. 27 The estate also sued the photographer and book publisher for breach of contract and misappropriation of personality. However, this article will focus on the legal significance of the copyright action only. 28 See Gould Estate v Stoddart Publishing Co [1998] 39 O.R. (3d) 545 (Ontario Court of Appeal) at [22]. The Ontario Court of Appeal relied on a case from the US, Falwell v Penthouse International Ltd, 215 U.S.P.Q. 975 (1981, Vir.Dist. Ct.) 977. It was held in Falwell that no copyright subsisted in unstructured remarks made to members of the media by the Reverend Jerry Falwell. See also Gould Estate v Stoddart Publishing Co [1996] 30 O.R. (3d) 520 (Ontario Court of Justice (General Division)) at [31]–[32], noting that the interview was conducted in informal, relaxed and casual settings, and Gould’s responses were not part of a private conversation or a structured speech, but instead were ‘offhand comments’ which he knew could find their way into the public domain. 29 Gould Estate v Stoddart Publishing Co [1998] 39 O.R. (3d) 545 (Ontario Court of Appeal) [24]. 30 ibid [26]. 31 ibid [24]. 32 ibid [5] and [27]. 33 ibid [26]–[27]. 34 ibid [26]. 35 ibid [21]. 36 See Gervais and Judge (n 8) 51. 37 Gould Estate v Stoddart Publishing Co [1998] 39 O.R. (3d) 545 (Ontario Court of Appeal) [22]. 38 ibid [23]. 39 The irony of the situation would, of course, be that an oral utterance would no longer be ‘casual’ or ‘spontaneous’ if it had earlier been written down in material form by the particular ‘author-speaker’ delivering it. 40 See Gervais and Judge (n 8) 51, where they write: ‘The later fixation of those conversations in the interviewer’s book was copyrighted expression belonging to the author-interviewer, but Gould’s oral speech at the time of the interview was not protected by copyright because there was no fixation.’ Gervais and Judge do not expressly mention a ‘self-fixation’ requirement. However, they assert that copyright protection was denied because Gould’s expressions were unfixed. This reference to the lack of fixation should probably be interpreted as being unfixed by the speaker, since it would be untrue to say that Gould’s words had not been reduced to material form (in view of Carroll’s avid on-site ‘recording’ activities). 41 See JA Cuddon, A Dictionary of Literary Terms (Penguin Books 1984) 660–61, where stream of consciousness is defined in literary criticism as a narrative method that seeks to depict the multitudinous thoughts and feelings which pass through the mind. 42 Zonen Endstra v Nieuw Amsterdam [2008] EIPR N73 (Endstra Tapes). (Netherlands Supreme Court, 30 May 2008). It should be borne in mind that the Endstra Tapes case was decided under the ‘European’ standard of originality as articulated by the European Court of Justice in Infopaq Int v Danske Dagblades Forening, Case C-5/08 [2009] ECR I-6569 (ECJ, Fourth Chamber). 43 The case concerned whether the contents of an interview by the police with an informant could constitute protectable subject-matter. In overturning the decision of the Court of Appeal of Amsterdam, the Dutch Supreme Court held that the informant’s conversations met the standards for protection, as they reflected human creative choices that were neither banal nor trivial. See also Bently and Sherman (n 13) 110. 44 Such a proposition would also be supported in the case law of the UK. See for instance Express Newspapers v News (UK) [1990] F.S.R. 359, in which it was held that a journalist who had selected extracts from a recorded interview for a report had exercised the necessary skill and judgment for copyright to subsist in the report. See also Torremans (n 12) 239, who notes that reporters who edit a work by selecting parts of a speech and adding their own comments are applying skill, labour and judgment to their report, which constitutes an original literary work and attracts its own copyright. 45 See Redwood Music v Chappell [1982] RPC 109, which articulates the principle that a derivative work can be both original and infringing. 46 See CCH Canadian Ltd v Law Society of Upper Canada [2004] 1 S.C.R. 339, in which the Supreme Court of Canada held that fair dealing was a ‘users’ right’ to be interpreted broadly and liberally, and was not merely an ‘exception’ to copyright protection. 47 The analogous provisions in the UK on fair dealing for the purpose of criticism or review can be found in s 30 of the CDPA. 48 Telstra Corporation Ltd v Phone Directories Co Pty Ltd (2010) 264 ALR 617, 624, [20]. 49 ibid. 50 See Bently and Sherman (n 13) 126, who note that the test for whether a person who contributed to the production of a work can be considered an ‘author’ is similar to the criteria used to evaluate whether a contribution is ‘original’. 51 Bently and Sherman (n 13). See also Gordon J in Telstra Corporation Ltd v Phone Directories Company Pty Ltd (2010) 264 ALR 617 at 685, who notes that: ‘Authorship and originality are correlatives. The question of whether copyright subsists is concerned with the particular form of expression of the work.’ 52 See, for instance, ss 29 and 30 of the UK’s CDPA (which protects activities such as research, private study, criticism, review and news reporting) and §107 of the US Copyright Act. 53 See, for instance, s 30(2) of the UK’s CDPA. 54 See Zonen Endstra v Nieuw Amsterdam [2008] EIPR N73 (Netherlands Supreme Court, 30 May 2008). 55 Infopaq Int v Danske Dagblades Forening, Case C-5/08 [2009] ECR I-6569 (ECJ, Fourth Chamber). 56 See Bently and Sherman (n 13) 110, where it is argued that there is ‘no reason to think that any different result would have occurred from applying the European standard of “author’s own intellectual creation”’. 57 ibid. 58 [1900] AC 539. 59 The position of the reporter who writes down a spontaneously delivered speech but who does not have any sound or film recording was expressly considered by Lord Beaverbrook in the debates on s 3(3). See, for instance, Hansard, HL, Vol 495, cols 610–611 (29 March 1988). Interestingly, it was noted by Lord Beaverbrook, in reference to opinions expressed by Lord Merton and Lord Williams, that the door may have deliberately been left ajar so as to allow courts the flexibility to decide in a particular case whether a reporter had expended sufficient skill and judgment to attract original copyright in the transcript. 60 See Nigel P Gravells, ‘Reporter’s Copyright and Sound Recordings: A Reply to Professor Derclaye’ (2013) 2 IPQ 91, 92. 61 See Estelle Derclaye, ‘Debunking Some of UK Copyright Law’s Longstanding Myths and Misunderstandings’ (2013) 1 IPQ 1, 10: ‘Section 5A is broad and includes a record of a speech by hand as it is no less a sound recording than a record of the same speech in analogue or digital audio format.’ 62 See Gravells (n 60) 91, 95. 63 ibid. 64 See Justine Pila, ‘An Intentional View of the Copyright Work’ (2008) MLR 535, 549, 557. 65 ibid 549. 66 There is a strand in the copyright literature which calls for the removal of the fixation criterion as a requirement for copyright protection. See for example Brennan and Christie (n 11) 309. © The Author(s) 2018. Published by Oxford University Press. All rights reserved. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Journal of Intellectual Property Law & Practice Oxford University Press

Spontaneous oral communications, impromptu speeches and fixation in copyright law: a comparative analysis

Loading next page...
 
/lp/ou_press/spontaneous-oral-communications-impromptu-speeches-and-fixation-in-0Af6pg6Fmt
Publisher
Oxford University Press
Copyright
© The Author(s) 2018. Published by Oxford University Press. All rights reserved.
ISSN
1747-1532
eISSN
1747-1540
D.O.I.
10.1093/jiplp/jpy064
Publisher site
See Article on Publisher Site

Abstract

The author Dr Eugene C. Lim is a Faculty Member at the School of Law, City University of Hong Kong. This article The ‘fixation doctrine’, which generally requires that a work be reduced to material form before copyright protection can arise, appears to have become firmly entrenched in the copyright laws of many common law jurisdictions. However, there remains some uncertainty as to whether fixation of spontaneous oral communications, such as speeches and interviews, can be performed by a person other than the ‘author’. By engaging in a comparative survey of the approaches to ‘fixation’ in the common law world, this article critically evaluates the extent to which Canada and Australia can learn from the experience of the UK and the USA in regulating authorship and entrepreneurial rights over initially unfixed spontaneous oral communications that are subsequently recorded by third parties. I. Introduction A young child narrates an experience in class; the oral narration is recorded by the teacher and published commercially many years later with the teacher’s permission.1 A famous classical musician gives a series of interviews to a magazine journalist; the interviews are recorded and reproduced in print many years later as part of a book.2 Although these real-life events involve disparate characters in different settings, they raise a common question: who owns the rights to spontaneous oral communications that have been recorded by a party other than the speaker? This is an important question that concerns the scope of the ‘fixation’ doctrine in copyright law in relation to ‘works’ that owe their existence to an individual who expresses them, and a different individual who reduces them to material form. Where intellectual property rights are concerned, copyright law in the ‘Anglo-American’ tradition generally prescribes that a work must satisfy both the ‘originality’ and ‘fixation’ requirements.3 In this regard, spoken words that are transcribed into material form can be protected by copyright law as ‘literary works’ in many common law jurisdictions if they constitute sufficiently original expression reflecting creative choices or skill and judgment from the author,4 and if they convey knowledge, information or pleasure.5 What is less clear, however, is the identity of the ‘author’ if the speaking (the ‘expression’) and the transcription (the ‘fixation’) are done by different parties. Should the author be considered the speaker, or the person who reduces or ‘fixes’ the spoken words into material form? Would it matter whether there is any agency relationship between the speaker and the transcriber? These questions have significant implications for the extent to which copyright can be used as a tool with which to restrain the publication of recorded oral communications. This article will focus on the challenge of identifying the ‘first author’ of recorded spontaneous expression, and not on (the somewhat broader) issue of whether fixation should be retained as a requirement for the subsistence of copyright. There is currently no general consensus at the international level6 or even within the ‘common law world’ on the issue of whether copyright subsists in spoken words that are not personally recorded by the ‘speaker’. Unlike the copyright legislation in the USA and the UK, the copyright statutes of Canada and Australia do not contain explicit rules on whether spontaneous oral communications have to be recorded by the speaker for copyright to arise. In addition, the Canadian and Australian statutes are silent on who owns the copyright to a transcribed oral work if the transcription is not done by the speaker. Interestingly, there is some jurisprudence in Canada and Australia which suggests that ‘speakers’ who do not transcribe or record their casual (unstructured) oral utterances will have no claim in copyright for their uttered words. A requirement that speakers transcribe or record their own words to assert authorship over them would stand in stark contrast to the statutory approaches adopted in the USA and the UK, both of which recognize, albeit to slightly different degrees, that the acts of composition and fixation of oral works can be ‘decoupled’, and that they need not be part of the same ‘creative act’, nor performed by the same individual, in order for copyright to arise. This article will use the controversial Canadian case of Gould Estate v Stoddard Publishing (‘Gould Estate’)7 as a springboard to discuss the extent to which spontaneous oral communications can be protected under Australian and Canadian copyright law. Although Gould Estate is not a recent decision, it continues to be cited as one of a select few Canadian authorities on fixation,8 and unless it is expressly overruled, will continue to serve as a source of guidance to both authors and publishers of recorded oral expressions. Interestingly, there has not been much published material on the fixation doctrine in Canada, nor has much attention been paid to the limitations of Gould Estate as a ‘putative’ case authority on the material form requirement. This article seeks to build upon the current sparse literature on fixation in Canada by making two contributions. First, it seeks to clarify the somewhat ambiguous judicial position in Canada regarding the fixation requirement and to highlight common challenges faced by both Canada and Australia in regulating authorship rights in spontaneous oral works. Secondly, it seeks to highlight the doctrinal and social benefits of introducing a provision to regulate authorship and ownership rights in works that are fixed by third parties, and the valuable insights that Canada and Australia can derive from the experiences of their ‘immediate peer group’9 in the common law world—in the context of this article the USA and the UK. In developing the above points, this article will argue that Gould Estate introduces significant doctrinal uncertainties into the concept of fixation in Canadian copyright law by implicitly supporting a rule of ‘self-fixation’, that is, a requirement that a speaker (in this case the deceased Canadian musician Gould) fix his own oral expressions to assert ‘first authorship’ over them. The article will observe that the presence of this implicit rule of ‘self-fixation’ might also be inferred from Australian case law on authorship. By highlighting common themes in their treatment of spontaneous oral works, this article will identify shared challenges that Australia and Canada face in the legal regulation of proprietary interests in initially unrecorded spoken words. It will demonstrate that a concept of ‘self-fixation’ for oral works is out of touch with the approach adopted in other parts of the common law world, and is arguably inconsistent with the fundamental principles of copyright law. The article will argue that the copyright statutes of Canada and Australia would benefit from clarificatory provisions, of the sort found in the US and UK legislation, prescribing the circumstances under which authorship can be asserted by individuals whose (original) oral works have been recorded by other parties. Finally, the article will conclude by highlighting some of the residual concerns and uncertainties relating to ‘third-party fixation’ that would benefit from closer attention by Canadian and Australian policymakers in future copyright reform efforts. II. Whose work is it? Authorship, expression and the fixation of spontaneous oral works in the UK, the USA, Canada and Australia It is a commonly accepted principle, particularly in jurisdictions following the Anglo-American legal tradition, that a literary work, which consists of words that make up original expression, must be recorded in some tangible medium in order for copyright protection to subsist in that work. This requirement is known by different names in various common law jurisdictions—it is predominantly referred to as ‘fixation’ in the USA and Canada, and ‘reduction to material form’ in many parts of the British Commonwealth.10 The fixation requirement has often been justified on the basis that copyright protects expression and not ideas, and expression in a tangible form is the primary means through which the structure, existence and boundaries of a work can be recognized and protected by the law.11 That being said, the physical act of fixation is not necessarily intertwined with original expression to the same degree in all classes of work. In the case of an artistic work, such as a portrait or painting, the expressive act (which triggers copyright protection) is inextricably connected with the act of fixation, since there can be no painting if the paint is not applied to the ‘surface’ in question.12 In other words, a picture that is imagined by a person is not an artistic work until it is given shape and form that can be perceived visually as an artistic composition in a relatively stable medium. However, in the case of a literary work, the composition of the work in the mind of its ‘author’ (which constitutes the original ‘expression’ attracting copyright protection) can be ‘fixed’ by another person (such as a scribe) who plays no role in the selection or arrangement of words in the preparation of the work but who merely gives those words material form by recording them on a transcribed surface or template. This is particularly the case if an ‘author’ delivers an extempore, impromptu speech, or utters words in response to interview questions that are taken down, verbatim, by other individuals who exercise little or no creative choice over their arrangement. When dealing with the scope of copyright protection for oral works, therefore, it is necessary for two important issues to be considered: first, whether transcription by a third party can satisfy the fixation requirement, and second, whether the speaker is deemed to be the first author (and copyright owner) of the transcribed words. The above two issues are addressed explicitly in the Copyright, Designs and Patents Act (CDPA) of the UK and the federal Copyright Act of the USA. Section 3(3) of the CDPA states that a work can be fixed by a person independent of the speaker, and the speaker will still be considered the author.13 This suggests that, for oral works in the UK, composition and fixation can be ‘decoupled’, and need not be part of the same act of creation.14 Under UK copyright law, it matters not by whom the oral work was fixed or whether the work was fixed with the consent or authority of the speaker, as long as fixation has occurred.15 On the other hand, Title 17 of the United States Code requires that the work must be fixed ‘by or under the authority’ of the author.16 Hence, although US federal law allows for a speaker to claim authorship over spoken words transcribed by other parties, the circumstances under which such authorship can be asserted are more limited. In particular, words that are recorded without permission in the USA would fall outside the ambit of the fixation provision and consequently would not attract copyright protection. The copyright statutes in Canada and Australia are notably ‘out of step’ with those of the UK and the USA, in that the Canadian and Australian Acts are both bereft of explicit clauses governing fixation of oral works by third parties. The Copyright Act of Canada does not even contain a general fixation provision, although fixation is referred to as a requirement for certain classes of work.17 In addition, references to unfixed works (such as performances) in the Canadian Act suggest that the fixation requirement is not uniformly applied to all categories of protectable subject-matter.18 However, in the Canadian case of Canadian Admiral Corp v Rediffusion, it was held that for copyright to subsist in a work, ‘it must be expressed to some extent at least in some material form, capable of identification and having a more or less permanent endurance’.19 Further, although the Canadian Act accommodates the possibility that certain forms of oral works (such as structured speeches) may be protected by copyright, it unfortunately does not provide much explicit guidance on whether oral works can be fixed by individuals other than the author-speaker in order for copyright protection to arise. There are interesting parallels between the Australian and Canadian copyright models. Although certain aspects of Australian copyright law have come under review in recent years by the Australian Law Reform Commission, the issue of ‘third party fixation’ appears to have been relegated to the sidelines by other concerns such as the adequacy of fair dealing and statutory exceptions to copyright in the digital economy.20 In Australia, there is currently no statutory provision expressly governing the fixation of oral works by third parties. However, section 22 of the Australian Copyright Act 1968 does state that, in the context of a literary work, the work is made when it is ‘first reduced to writing or to some other material form’. There is nevertheless some ambiguity in another section of the Act, ie section 32, which requires an author to be a ‘qualified person’ in Australia during at least a substantial period over which the work was made.21 This leaves open the possibility that the work could have been ‘made’ by a person other than the author. Since ‘making’ is defined in section 22 as the time at which the work was reduced to material form, the language of section 32 appears broad enough to entertain the possibility that the person who does the transcription (or reduction to material form) does not necessarily have to be the ‘author’ of that particular work in order for copyright protection to arise. However, Australian jurisprudence appears to disapprove of such a broad interpretation of authorship by seeming to require that composition and fixation be part of the same ‘creative act’. In Telstra Corporation Ltd v Phone Directories Co Pty Ltd, Gordon J, citing IceTV Pty Ltd v Nine Network Australia Pty Ltd (IceTV)22 defined ‘authors’ as individuals ‘who bring the work into existence in its material form’ or those who ‘gather or organize the collection of material and who select, order and arrange its fixation in material form’.23 It has been suggested that the High Court of Australia judgment in IceTV has arguably raised the standard of authorship necessary for copyright to subsist in original literary works.24 Further, subsequent cases in the ‘post-IceTV’ era have resulted in the ‘fragmentary development’ of the law, giving rise to uncertainty about the prerequisites for the subsistence of copyright in literary works.25 Despite the somewhat conservative stance adopted by the Australian courts in Telstra and in the post-IceTV cases with regard to authorship, it has been suggested that an interpretation which links authorship to fixation is not clearly supported by the legislative history behind the ‘making’ provision in Australia.26 Interestingly however, it may be observed from the current provisions of the Act that there are no explicit guidelines concerning the circumstances under which a person may assert authorship over an oral work that has been reduced to material form by another party. As is the case in Canada, the Australian copyright framework leaves many key questions unanswered about the legal relationship between the concepts of original composition, authorship and fixation in respect of spontaneous verbal and oral communications that have not been transcribed by the ‘author-speaker’ in question. III. The Gould Estate case and the oddity of a ‘self-fixation’ requirement This section will use the Canadian case of Gould Estate to illustrate the problems surrounding a ‘self-fixation’ requirement for oral works in copyright law. Although the case is Canadian in origin, it raises important themes that are equally pertinent to the Australian copyright model. By engaging in an analysis of the court’s holding in Gould Estate, the article will highlight shared doctrinal challenges faced by both Canada and Australia in regulating copyright over spoken words that are not fixed by the speaker. The decision in Gould Estate concerned whether an interviewed celebrity had a valid copyright claim over spontaneously uttered responses to interviews that had been conducted many years earlier, in the spring of 1956, for a magazine article, which was eventually published under the heading ‘I Don't Think I'm At All Eccentric’. The celebrity in question, noted Canadian classical musician Glenn Gould, had given a series of interviews to a freelance magazine journalist and photographer by the name of Jock Carroll while on a planned holiday in Nassau, without having prepared for or transcribed the interview answers in advance. Gould’s interview answers were therefore spontaneous and ‘unfixed’ at the precise moment they were uttered. However, they were transcribed and recorded almost instantaneously by Carroll, who took copious notes and made extensive tape recordings of the conversations. In addition, Carroll took numerous photographs of Gould for the magazine article. Many years later, after Gould’s death, Carroll sought to include the photographs and interview answers as part of a book on Glenn Gould, entitled ‘Glenn Gould: Some Portraits of the Artist as a Young Man’, which also contained original commentary by Carroll and captioned photographs depicting selected moments from the musician’s life. The estate of Glenn Gould asserted copyright27 over the interview answers, photographs and other elements in the book and sought to restrain its publication. The court had to decide whether Gould’s estate or Carroll (the transcriber of Gould’s spoken words) held the copyright in respect of Gould’s spontaneous oral communications during the interview. It was held, by both the Ontario Superior Court of Justice and on appeal to the Ontario Court of Appeal, that Gould’s estate had no copyright claim in respect of the interview answers as they were merely casual, unstructured utterances that were not intended to attract any proprietary interest.28 Carroll was considered the author of the book, including the interview excerpts, photographs, and other literary elements, which taken together as a whole, constituted an original work attracting copyright protection.29 Further, the court agreed that once Gould had consented, without limitation, to be the subject of a journalistic piece, he (or his estate) did not have any proprietary claim over the finished product, nor could he restrain any further reproduction of the photographs or prevent the journalist from producing additional material about him.30 Since Carroll had received an assignment (from the magazine publisher) of all rights in the magazine article published in 1956, he was the sole owner of copyright in the entire book (published years later).31 Although the issues in Gould Estate could, in the words of Finlayson JA, be decided on the basis of ‘conventional principles relating to copyright’,32 and that there was no need for recourse to the relatively new tort of ‘misappropriation of personality’, the ‘intellectual property’ analysis regarding oral communications provided in the judgment of the Ontario Court of Appeal is surprisingly brief.33 There is considerable emphasis given to the fact that there was no contract, express or implied, between Carroll and Gould and to the fact that Gould had ‘clearly consented’, without limitation, to being photographed and interviewed in Nassau.34 It was also highlighted that Gould could not have been said to have commissioned the photographs of himself, since Gould had not paid for the photo-taking sessions with Carroll, who was in any event not a photographer for hire in the ordinary commercial sense.35 It is submitted that the copyright issue for the photographs, captions and commentary in the book is relatively uncontroversial, since in the absence of any contract to the contrary, Carroll would be the first owner of copyright in any original material that he had personally written or produced. These would include the photographs (which Carroll had composed and taken), the descriptions thereof, and the written reflections, all of which constitute Carroll’s original expression. What is more problematic from the perspective of copyright law is the legal entitlement to use the selection of words spoken by Gould, which did not ‘originate’ from Carroll in the authorial sense. With the aid of recording technology to back up any ‘slips’ in stenographic skill, and even if he had been ‘slavishly’ transcribing words without the assistance of a machine, Carroll could not have been said to have been exercising ‘creativity’ or ‘judgment’ in merely transcribing words uttered by another individual. Although the judgment as a whole is illuminating, the two points raised by Finlayson JA in relation to the interview transcript regrettably do not provide much substantive guidance on the fixation debate in Canada. Interestingly, although Gould Estate has been cited as an authority on fixation in the secondary literature,36 the word ‘fixation’ is not once mentioned in the judgments of either the Ontario Court of Appeal or the Ontario Court of Justice. One point raised by Finlayson JA, albeit with some degree of caution and scepticism, relates to the relaxed, spontaneous nature of the setting in which Gould gave the interviews. In this regard, it was noted, from the transcript of the holding by the motions judge, that Gould was not delivering a ‘structured lecture’ or dictating to Carroll, but rather was engaging in ‘easygoing conversation’.37 The second point raised is less clear and perhaps somewhat troubling. Finlayson JA asserted, without providing much by way of analysis, that it was ‘evident from this record’ that Gould did not have a copyright with respect to the transcriptions of his oral utterances.38 Without clearly distinguishing between the portions derived from Gould’s oral ‘contributions’ and the elements composed by Carroll, the court simply ruled that Carroll was the author of the ‘text and captions’ in the biographical account of Gould’s life and therefore was the copyright owner of the written material in the book. In particular, the basis of Carroll’s copyright entitlement to use Gould’s words is not articulated or fleshed out in the judgment; it is regrettably an unresolved question whether Carroll had become the author of the transcribed words because he had ‘transformed’ them into original expression of his own, or whether Gould’s estate had an inferior claim to the words as copyrightable subject-matter because Gould had not personally written them down in material form at or prior to the interviews. If the latter interpretation is adopted, it would bring Canada dangerously close to a doctrine of ‘self-fixation’ in copyright law. The existence of a ‘self-fixation’ requirement for casual oral communications39 in Canada appears to find some indirect support in the secondary literature, with two leading commentators suggesting that Gould’s oral statements did not attract copyright protection because ‘there was no fixation’40 (presumably by Gould himself). In relation to the first point raised by Finlayson JA, it is highly debatable whether ‘structure’ should be the determining factor in determining whether a spoken work constitutes protectable literary expression. There is a very fine line between formal and casual expression, and the distinction between the two is often a question of degree. Section 2 of Canada’s Copyright Act specifically mentions ‘lectures’ as an example of subject-matter that can qualify as an original (literary) work, and the section goes on to stipulate that the term ‘lecture’ includes addresses, speeches and sermons (corresponding to the French term ‘conférence’). In this vein, Finlayson JA emphasized in his judgment that some forms of ‘structured’ oral expression, such as lectures and sermons, could be protected, but not others. Yet there are significant dangers in assuming the presence of a ‘bright line’ between structured and unstructured works that have been reduced to material form. Valuable works of literature may be excluded from the pantheon of copyright protection if the ‘structured/unstructured’ dichotomy is applied too strictly. Some of the most enduring works in English literature, such as James Joyce’s Ulysses, do not conform to a ‘formulaic’ narrative structure. To some critics, Joyce’s work may read more like a spontaneous ‘stream of consciousness’ than a work with a clearly delineated narrative.41 Although Joyce’s novel was not a ‘speech’, if Canada is willing to allow copyright protection for certain forms of speeches but not others, then ‘structure’ alone provides a highly unsatisfactory basis for determining what to exclude from protection. There is a concomitant danger that judges may be making assessments of literary merit when they exclude works based on how ‘structured’ they are. Regardless of whether one is considering a written work or a transcribed oral work, the test for whether copyright subsists in the subject-matter should be driven by established principles of originality, rather than by the judge’s perception of ‘structure’ per se. Although the eventual decision by the Ontario Court of Appeal was probably justified in the circumstances, the analysis is unsatisfactory. The judgment is laudable in that it highlighted the public interest of allowing Carroll’s work to be published and circulated, thereby providing valuable insight into the life of a celebrated musician. However, the reasoning vis-à-vis ‘ownership’ of Gould’s words is awkward from a doctrinal perspective, and the same public interest considerations could have been achieved through the use of copyright defences or other doctrinal mechanisms validating Carroll’s use. The primary concern with the judgment relates to the manner in which the ownership of copyright in relation to Gould’s spoken words was dispensed with. The relationship between originality and fixation should also have been ‘disentangled’ in the judgment. A plausible argument could be made that despite their lack of ‘structure’, Gould’s interview responses were original literary works that not only reflected his choice of words, but also bore his ‘personal stamp’ as an author, revealing dimensions of his personality that would convey knowledge, information and pleasure to their intended audience. There is some authority from outside Canada, such as the Dutch case of Endstra Tapes,42 which suggests that spontaneous oral conversations can constitute works of authorship if they are the result of human creative choices.43 As for the fixation issue, whether Gould’s words were ‘fixed’ should not be a subject of any dispute, bearing in mind Carroll’s extensive ‘recording activities’ on site. Since it might be argued that Gould’s oral statements were both original and recorded in material form, one would need a plausible basis upon which to deny Gould copyright protection over his expression. One way of doing this would be to adopt an explicit ‘self-fixation’ requirement, which is merely implied in the judgment. However, in the absence of such a requirement, a more doctrinally coherent way of justifying Carroll’s copyright ownership in the work would be to employ the principle of ‘transformative use’. By selecting portions of the quoted statements and then commenting on them, and embellishing them with insights of his own, Carroll had effectively transformed the text containing portions of the interview transcript into an original literary work of his own.44 However, even original ‘derivative’ works can infringe if they borrow a qualitatively significant portion of text from another source.45 Nevertheless, if Carroll had only taken reasonable portions of text from the interview transcript for his book, he might have been able to avail himself of a fair dealing defence under Canadian copyright law.46 ‘Fair dealing for the purpose of criticism or review’, which is enshrined in section 29.1 of the Copyright Act of Canada,47 would provide Carroll with some shelter against liability for infringement of any copyright that might have subsisted in Gould’s oral responses, provided of course that Carroll adds insight of his own to the quoted statements, and includes clear attribution to the source of the interview answers in the published version or content. A further way of justifying a decision in favour of Carroll might be to imply the presence of a licence to use the interview transcript in future publications. Although there was no contractual agreement between Gould and Carroll, Gould had specifically invited Carroll to join him at the location in Nassau and acquiesced to the recording of notes for the purpose of writing a story about him for promotional purposes. Since there were no limitations placed on the extent to which the interview could be used in future publications, the unqualified ‘consent’ implied in this case might have provided Carroll with a defence against a copyright infringement claim. It is not entirely clear that Gould’s estate would fare any better under the Australian copyright system, if the facts were to occur today on Australian soil. Assuming, hypothetically, that Gould was a ‘qualified person’ in Australia, the primary stumbling block standing in the way of Gould’s claim to authorship would be the Australian judiciary’s interpretation of the fixation requirement. Bearing in mind that authors are defined as individuals who ‘bring the work into existence in its material form’,48 it would be difficult to argue that Gould had taken steps to ‘gather or organize the collection of material’ and to ‘select, order and arrange its fixation in material form’.49 An interesting parallel between the Australian case of Telstra and the Canadian case of Gould Estate is the weight given to ‘structure’, or ‘organization of material’ in the Australian context. In this respect, Gould could hardly be said to have been ‘organizing’ the transcription of his words, since he had neither hired Carroll to do the recording of his interview responses nor supplied him with the means or equipment to do so. Gould’s participation in the recording process can be more accurately likened to ‘passive acquiescence’ than active gathering of data. Even if it could be argued that Gould was ‘ordering’ the sequence of thoughts in his head, he would not be the party performing the act of fixation in material form. In this regard, Carroll’s role appears to be more akin to that of a ‘gatherer’, if not for the original ‘slavish’ transcription, then in the eventual compilation and selection of quotations for his published book. The acts of transcription and expression appear to be treated as an indivisible whole by the Australian courts, which echo the submerged assumptions driving the decision by the Ontario Court of Appeal in the Gould Estate case on the rights over recorded spoken words. Although Gould Estate is a case that governed casual oral utterances, the issues raised in that dispute have larger ramifications for the regulation of rights in (initially) unfixed spontaneous expressions that are subsequently recorded by third parties. The copyright statutes of Canada and Australia share a common weakness in that they do not specify with sufficient clarity the identity of the author in such situations. As demonstrated in the preceding analysis of the Gould Estate case, the Canadian and Australian jurisprudence appear to endorse the notion that fixation and expression are indivisible parts of the same creative act for oral works of a literary nature. While such a principle would arguably be justified for artistic works, it would create difficulties for literary works that are generated ‘spontaneously’ by an orator. The next section of the article will canvass some of the challenges that a doctrine of self-fixation would pose if applied uncritically to all works of creative expression. IV. A return to first principles: protecting original expression rather than mere labour in copyright law One of the arguments raised in the previous section relates to how a self-fixation requirement provides a highly unsatisfactory basis for discriminating between protectable and unprotectable oral expression. Although Gould could easily have recorded his interview answers as he was uttering them, the argument against imposing a ‘personal fixation’ requirement by an author could also be made in respect of works produced by individuals who can only (or choose only to) generate creative literary or musical output through unfixed, spontaneous performance. A gifted writer who weaves together tales of great intrigue by reciting lines of prose that are meticulously composed in his head, or a musician who can only produce stirring melodies from unwritten scores performed from memory, are examples of creators who would fall within this category. Without a recording of these works, the creative input of these spontaneous ‘creators’ might be lost forever. Although Gould Estate provides only one instantiation of how oral works can lead to disputes over intellectual property rights, the way in which the fixation issue is dealt with by courts and policymakers has significant implications for the regulation of rights in works that are ‘fixed’ by individuals other than those responsible for their first expression. The concern here is not that third-party transcribers can acquire authorship status by engaging in transformative use of recorded works, thereby creating original derivative works. Transcribers, such as Jock Carroll in the Gould Estate case, who exercise skill and editorial selection in producing derivative works reflecting their own creative choices, can and should be recognized as authors. However, the question here is whether speakers whose expression is recorded should be granted any rights in the recorded expression as literary works. If the law does not confer copyright on speakers in respect of their original oral works, but allows transcribers to assert claims over carefully edited selections of those recorded oral works, it runs the risk of creating a situation where an amanuensis is treated as the ‘default first author’ of expressions originating from another party. The interpretive ambiguities in the Canadian and Australian copyright statutes vis-à-vis third-party fixation might give rise to just such a scenario. A clearly delineated legal position identifying the author in cases of spontaneous expression would therefore help to obviate some of the uncertainties raised in Gould Estate and in the examples described in the preceding paragraph. Under the current vagaries of the Canadian and Australian jurisprudence, spontaneous ‘creators’ may, in some cases, find their identity as authors ‘usurped’ by individuals who meticulously and conscientiously ‘arrange for’ and perform the act of recording the work in material form. In other cases, such creators may find that their authorship of the oral work is not recognized by the law, even if it has been recorded by a third party. Imposing an implied ‘self-fixation’ requirement (à la Gould Estate) on authors would effectively deny ‘spontaneous creators’ any copyright in their work unless they fix it first. Unless we are dealing with subject-matter where the fixation is intimately connected with the act of expression (as in photography or visual art), granting first authorship of recorded spontaneous creative work to third-party recorders and editors, while disregarding any claim of the spontaneous ‘composer’, would severely undermine the doctrine of originality in copyright law. The test for whether a person can be considered an ‘author’ is strongly connected with the test for originality.50 A key criterion for authorship lies therefore in whether the expression contributed is of the ‘right kind and quantity’.51 In view of the strong connection between authorship status and originality, it would arguably be odd to disregard a speaker’s expressive contribution to a recorded speech, while recognizing a transcriber’s editorial contribution to a revised version of the same. V. The social utility of a third-party fixation clause The uncomfortable doctrinal oddities outlined in the preceding section suggest a need for statutory clarification on the identity of the first author in the case of (initially unfixed) oral works and other forms of spontaneously-generated content in the Canadian and Australian contexts. For simplicity, the term ‘third-party fixation’ or ‘TPF’ clause will be used in this context to refer to a statutory provision that stipulates the circumstances under which a work can be fixed by a person other than the author, in order for first ownership of copyright to vest in the author. This section of the article will discuss a number of ways in which an explicit TPF clause would help to promote social utility, and contribute to the preservation and dissemination of culturally valuable works that might otherwise be lost or forgotten. An important social benefit of having a TPF clause in copyright legislation is that it would provide clearer guidance and coordination to authors and publishers on the extent to which creators of initially unfixed spontaneous expression can restrain its subsequent publication by third parties. By identifying the first author of TPF works, such a clause would facilitate the negotiation of licensing, recording and profit-sharing arrangements between authors, publishers and broadcasters. The legal recognition of their rights in spontaneously-generated creative expression would encourage prospective authors of oral and unfixed works to enter into recording and distribution contracts with third parties, thereby ensuring that their intellectual legacy survives in some recorded form. Having a legal framework in place to facilitate the licensing of rights in relation to TVP works would incentivize both the creation and the recording of spontaneous expression. First, authors in whom copyright status is vested would be in a stronger bargaining position to dictate the conditions under which their work can be shared and disseminated in return for financial remuneration or other consideration, thereby increasing the likelihood that they might authorize the recording of their work by third parties. Second, media enterprises, publishers and individuals who are commissioned by the author, either before or after the fact of recordation, to assist in the dissemination of the work would have an incentive to maintain and preserve the work in a form which is suitable for public access. In the latter scenario, the incentive to record, preserve and disseminate the work would arise principally from the financial benefit derived by a third party from a recording or licensing contract with the author. The twin incentives operating on both sides of the ‘author-transcriber’ equation for commissioned oral works would facilitate the creation, dissemination and transmission of the works to the general public, thereby contributing to the growth and advancement of knowledge and culture in the longer term future. It is of course important to note that not all authors are necessarily eager to ‘showcase’ and disseminate their spontaneously-generated works to the general public. It is equally important to emphasize that in the US copyright system, unfixed spontaneous works that are subsequently recorded by third parties without authorization would not generally attract copyright protection. Hence, in such a scenario, where copyright protection is not available, unauthorized recordings might be shared and disseminated against the author’s wishes. However, it is submitted that even in the US context, a TPF clause plays an important role in enhancing social utility by clarifying and coordinating the dynamic interplay of rights enjoyed by creators of spontaneous content, publishers and third-party users. By providing public notice of the distribution of rights and responsibilities concerning the ownership and use of recorded oral works, an explicit TPF clause can help to reduce the incidence of litigation and legal disputes arising from an uncertain legal climate. Understanding the conditions under which the material form requirement must be satisfied in order for copyright to arise would also facilitate risk management by would-be authors in tailoring, adapting and arranging for the mode of fixation to suit the particularities of the copyright regime in which they produce their oral works. For instance, in a US-style TPF regime, an author wishing to limit access to a work could benefit from the guidance that the US TPF clause provides by taking risk management strategies to ensure that the expression is only made accessible to commissioned persons for the purpose of recordation, and shielded as far as possible from unauthorized third parties. The benefits associated with the ‘coordinating’ function of a TPF clause would also apply to third-party use of recorded oral expression. By providing clearer guidelines on when recorded expression falls within the scope of copyright protection, and when such expression enters directly into the public domain, a TPF clause can perform an important public notice function in helping users of information distinguish between legally permissible and infringing use. Fostering a climate of greater legal certainty and transparency with respect to the allocation of rights in TPF oral works would also have the added advantage of facilitating and legitimizing the broader dissemination of unprotected works to members of the public. In this regard, by providing guidance on the extent to which initially unfixed oral works are protected by copyright law, and when such works enter into the public domain, an explicit TPF clause would play an important role in circumscribing the circumstances under which dissemination of the works to the public would be legally permissible. Making such recorded oral works available to the public in a manner allowed by copyright law would, in turn, stimulate innovation, cultural growth and development by creating a more permissive and conducive environment for the production of derivative works, parodies and user-generated content, and contribute to the flourishing of creative expression, thereby promoting discourse and communication in contemporary civil society. Critics who are opposed to the idea of ‘decoupling’ expression and fixation might argue that granting ‘speakers’ copyright protection over their uttered words could have a chilling effect on ordinary discourse and communication by restricting the republication of those words by third parties. However, it should be borne in mind that, as for other categories of protected expression, copyright does not confer an absolute monopoly over TPF works, and is subject to important limitations and defences, such as fair dealing or fair use.52 In particular, an appropriately construed ‘users’ rights regime’ would help to ensure that reasonable limited uses of oral works can be made even during the term of copyright protection. Quoting selected excerpts from an interview that are subsequently commented upon in a biographical work of a well-known pianist would, for instance, potentially qualify for the fair dealing provision for criticism or review. Likewise, a newspaper journalist who reproduces a reasonable amount of text from a spontaneous oral speech delivered by a public figure for the purpose of writing an article on current affairs would enjoy some degree of shelter from liability under the fair dealing provision for news reporting.53 VI. Charting the way forward in clarifying the scope of a provision on TPF: future challenges Despite the social and doctrinal arguments in favour of an express statutory provision recognizing the authorial status of a ‘speaker-creator’, the statutory insertion of a TPF clause into copyright legislation is a challenge that should be approached with care and caution, taking into account the insights that might be gleaned from the US and UK experiences. This section of the article will highlight a number of residual concerns and remaining uncertainties in the US and UK approaches to third-party fixation—issues that would deserve closer scrutiny if Canada and Australia were to consider amending their copyright legislation to incorporate an explicit TPF clause. In updating their copyright legislation to address issues of third-party fixation, one potential issue that will likely arise in the Canadian and Australian contexts is whether authorization of fixation by the ‘speaker-creators’ of spontaneous or oral works should be made a necessary criterion for authorial copyright to arise in TPF works. This is a matter on which the legal positions of the USA and the UK diverge sharply. If Canada and Australia were to follow in the footsteps of the USA, and make authorization of fixation a requirement for copyright to subsist in TPF works, then the range of spontaneous oral works that would qualify for legal protection would be much narrower. In particular, impromptu speeches that are not fixed with the authorization of the speaker would fall outside the scope of copyright protection. The US approach to fixation would require not only that the author give consent to the recording, but also assumes an intention to create a work, since authorization to record expression necessarily entails a conscious desire to have that expression reduced to material form by a third party identified by the author. Such an approach would suggest that an intention to create a work serves as a prerequisite for copyright protection in the USA. This position has, however, been criticized in the European context as being too strict. In the Dutch case of Endstra Tapes,54 the Supreme Court of the Netherlands held that transcribed conversations that were the result of human creative choices, and that were neither banal nor trivial, could be considered works of authorship. It has been observed that a similar result would likely be obtained by applying the European standard of ‘the author’s own intellectual creation’, as articulated in the Infopaq case,55 to the facts of Endstra Tapes.56 In the European context, requiring an author to demonstrate a conscious desire to create a work would likely be viewed as unfairly burdensome.57 From this perspective, imposing too onerous a standard of proof relating to the mental state of an author, or on the purpose of creating a work, might arguably cause many otherwise original works to fall outside the scope of copyright protection. In view of the consequences of setting the authorship threshold too high, it is possible that important oral works of aesthetic value that are recorded without the permission of the ‘author’ or created without ‘authorial intent’ may simply be denied protection in the USA. Although this means that copyright would not be available as a tool to restrain the circulation of oral works recorded without authorization in the USA, it also means that would-be ‘authors’ who generate intellectual content spontaneously may be more reluctant to showcase their creative talent in venues where their expressions may be recorded by third parties. Further, denying a work copyright protection simply because the recording was not authorized by the person ‘expressing’ it quite simply ignores the latter’s intellectual contribution to the existence of the work. Any intellectual judgment and creativity that went into the work’s composition, even if done spontaneously, would, under the US copyright model, not be viewed as giving rise to copyright protection if the transcription had been done furtively or illicitly by an unauthorized third party. While the US approach to the authorship of initially unfixed TPF works may raise concerns about pitching the standard of ‘conscious creation’ too high, it is important to note that attempts at transplanting the UK model into Canada and Australia should also be approached with caution and careful forethought. If Canada and Australia decide to follow in the footsteps of the UK and enact a TPF clause styled after section 3(3) of the CDPA, some consideration should be given to the interpretive uncertainties surrounding the extent to which section 3(3) protects the neighbouring rights of transcribers and reporters in servile copies of recorded spontaneous works. In incorporating a UK-style approach, one of the challenges facing the drafters of an explicit TPF clause in Canada and Australia relates to determining the appropriate scope and nature of the neighbouring or entrepreneurial right (if any) to be granted to a ‘third-party fixer’ in relation to a mechanical transcript or record of a TPF work (as opposed to the underlying literary, dramatic or musical work in which creative copyright subsists and which vests in the author-speaker). The key issue here lies in elucidating the interplay of overlapping authorial and entrepreneurial rights that might subsist in an original oral work which is recorded through a mechanical process by a third party. If the third-party fixer has exercised original thought and selection in preparing the record, then the record may be considered an original derivative work in its own right. However, if the record is merely a servile rendering in material form of an initially unfixed oral work, an important question that needs to be answered is whether the mechanical record attracts a separate neighbouring right that vests in the transcriber, particularly if the record is reduced to a material form or tangible medium that is not expressly classified in the statute as copyrightable subject-matter. Although section 3(3) of the UK’s CDPA addresses the issue of neighbouring rights obliquely, there are unanswered questions and remaining uncertainties that will be discussed in the ensuing paragraphs. The CDPA currently recognizes eight categories of copyrightable subject-matter, four of which relate to original authorial works, and four of which relate to recorded or transmitted media in which entrepreneurial rights commonly subsist. The eight categories of copyrightable subject-matter are set out in sections 3 through 8 of the CDPA and comprise the following: literary works (including databases), dramatic works, and musical works (sections 3 and 3A), artistic works (section 4), sound recordings (section 5A), films (section 5B), broadcasts (sections 6 and 6A) and published editions (section 8). The latter four categories of subject-matter, namely sound recordings, films, broadcasts and published editions, are strongly associated with derivative content or formats, based on underlying works or expressions, that are prepared by entrepreneurial entities, such as recording companies, production companies, television stations and publishing houses. The possibility that overlapping authorial and entrepreneurial rights may subsist contemporaneously in recordings of TPF works is alluded to in the final part of section 3(3), which provides that ‘nothing in [section 3(2)] affects the question whether copyright subsists in the record as distinct from the work recorded’. Hence, literary, dramatic and musical works as defined in section 3 may be fixed by third parties in formats or media that are recognized by the CDPA, and the Act leaves open the possibility that derivative entrepreneurial rights in such recordings may be enjoyed independently by third parties. Although section 3(3) itself does not provide a list of permissible media on which TPF recordings may be fixed, the sections following section 3 provide specific categories of media in which copyright may subsist. One question that may be raised in this context relates to whether the CDPA provides a comprehensive and inclusive framework for media on which third-party fixation may occur. None of the above categories, for instance, relate to servile handwritten transcripts. As an example, it is not entirely clear whether a reporter who makes a handwritten copy of an impromptu speech delivered by a public figure, in a way that might be reminiscent of the classic case of Walter v Lane,58 will be granted any entrepreneurial or neighbouring rights under the CDPA in respect of ‘mechanical’ handwritten copies of the speech so produced. This hypothetical scenario presupposes that the reporter does not qualify for original authorship of a literary nature, but merely produces a ‘servile’ handwritten copy that seeks to be as faithful as possible to the oral work produced by the author-speaker. Although section 3(3) of the UK’s CDPA seems to allow for the possibility that neighbouring rights of third parties can subsist in recorded copies of initially unfixed expression, its language has been couched in relatively ambiguous terms.59 One commentator has interpreted the section to mean that section 3(3) does no more than leave open the question whether a reporter enjoys a separate (copy)right in his or her record.60 Whether or not a separate right is enjoyed would ultimately depend on whether the record falls within one of the recognized categories of protected subject-matter in the CDPA. Hence, if the reporter has made a sound recording, or a film recording, or a published edition of the recorded oral work, then the reporter’s record would be protected under the corresponding sections of the CDPA respectively. In these situations, the reporter’s neighbouring right in the recording or edition would subsist alongside the speaker’s authorial copyright in the literary work embodied in the recorded speech. However, what is less clear is whether handwritten transcripts (and transcripts fixed in other unclassified types of media) that are principally servile records of the transcribed (oral) work will qualify for concurrent protection as entrepreneurial works, since there is no specific category in the CDPA which recognizes reporters’ transcripts as protectable subject-matter per se. The above question has provoked an animated debate among UK copyright scholars, reflecting a range of views on how to treat the servile handwritten transcript as a possible candidate for (derivative) copyright protection. It has been suggested, for instance, that a record of a speech by hand qualifies as a ‘sound recording’, just as an audio recording would, under section 5 A of the CDPA.61 Another commentator has rejected this view, arguing that a sound recording must enable the reproduction of recorded auditory sounds, rather than the mere words of a literary work;62 and hence a handwritten, verbatim transcript of spoken words which falls outside the eight categories of subject-matter in the CDPA might not be eligible for any protection.63 In view of the limited reach of the eight extant categories of protected subject-matter, it has been proposed that a new category relating to ‘mechanical copyright’ in literary transcripts be introduced into the CDPA, so that verbatim handwritten records of TPF works can be protected alongside other entrepreneurial subject-matter such as sound recordings and typographical arrangements of published editions.64 In this vein, it has been observed that the need for copyright law to accommodate the protection of written transcripts would be particularly acute ‘where the result is a product of genuine cultural interest that might otherwise remain inaccessible to the public’.65 The extent to which entrepreneurial rights in recorded spontaneous works should be expressly protected by legislation is an issue that should be taken into account by Canadian and Australian policymakers when drafting a TPF clause that would be suitable for their domestic copyright systems. Handwritten transcripts constitute, of course, only one example of an entrepreneurial ‘product’ that is not specifically categorized in the CDPA as protected subject-matter. As technology develops, new forms of media will likely be made available in the future to facilitate the recording of information. The larger challenge of enacting a TPF provision à la section 3(3) of the CDPA lies, therefore, in providing clearer guidance on the types and categories of recording media that can potentially trigger neighbouring rights in third-party recordings of spontaneous oral works. In considering this challenge, a possible way forward might be to enact a TPF provision that is technology-neutral, which expressly elucidates the interplay of authorial and entrepreneurial rights in TPF material, and that is broad enough to encompass emerging forms of transcription and recording media in the digital era. Interestingly, some guidance can be derived from §102 of the US Copyright Act on copyrightable subject-matter in drafting a technology-neutral clause on fixation, by making reference to works and recordings that are ‘fixed in any tangible medium of expression, now known or later developed, from which they can be perceived, reproduced, or otherwise communicated, either directly or with the aid of a machine or device’ (emphasis added). If applied to the TPF context, a technology-neutral provision that reflects the general ambit of the quoted segment from §102 of the US Act would arguably be broad enough to encompass derivative entrepreneurial works that are produced by hand, as well as those recorded with the aid of a machine or other technology, now known or later developed. In learning from the experiences of both the USA and the UK with their own TPF provisions, it is accordingly important to recognize the twin challenges of stipulating the circumstances under which authorial copyright can subsist in TPF works and clarifying the range of permissible media through which third-party fixation can be effected for neighbouring rights to arise. In revising their respective copyright statutes to meet these twin challenges, incorporating an explicit TPF clause to regulate the interplay between authorship and entrepreneurial rights for TPF works in Canada and Australia would provide authors, reporters and other ‘entrepreneurial’ transcribers with more concrete guidance on the circumstances under which their efforts in producing and recording TPF works will be legally protected. A clearly defined statutory framework for the legal protection of overlapping authorial and entrepreneurial rights in TPF works would also play an important role in demarcating respective rights and responsibilities, in facilitating the negotiation of licensing agreements, and in providing third-party transcribers with incentives to reduce spontaneously-generated works to material form, thereby helping to ensure, at least in some small measure, that (initially) unrecorded works of cultural significance can be preserved and shared with others for the benefit of future generations. VII. Conclusion The doctrine of fixation continues to generate a fair degree of controversy in copyright law, and scholars opposed to a fixation requirement in copyright law might even suggest that the doctrine is facing an ontological crisis.66 For the time being, however, fixation appears to be firmly rooted in the copyright legislation of many common law jurisdictions, and is not likely to be removed as a requirement for the copyright protection of literary, musical and dramatic works in those jurisdictions, at least in the immediate foreseeable future. What is less clear is whether expression and fixation have to be done by the same party in order for copyright to subsist in spontaneously expressed creative output of a literary nature. This article has sought to highlight uncertainties in the copyright frameworks of Canada and Australia, both of which do not define with sufficient clarity the identity of the author when expression and fixation of oral works are performed by different entities. This article suggests that the lacuna in the copyright frameworks of Canada and Australia vis-à-vis the identity of the author in cases of ‘unfixed expressions’ can be remedied by means of a statutory provision, which expressly accommodates the possibility of third-party fixation for spontaneous oral works. Such a provision would clarify some of the ambiguities arising in Gould Estate and similar cases, and codify the principle that ‘self-fixation’ should not be a requirement for authors to enjoy copyright protection in their spontaneous oral expressions. A clarificatory provision on third-party fixation would help to foster greater doctrinal clarity in the protection of spontaneous oral expressions that are subsequently recorded, facilitate the negotiation of licensing and distribution agreements, and help to mediate the balance of power between authors, entrepreneur-recorders and subsequent users of those recorded spontaneous expressions. The in-built statutory limitations to copyright protection also serve as an important safeguard against excessive monopoly control over the recorded content, and ensure that reasonable uses can be made of the oral work even during the term of copyright protection, thereby facilitating the preservation and dissemination of valuable expression that might otherwise be lost, and enriching the public domain in the long term. Jock Carroll’s biography of Glenn Gould provides us with a rare and fascinating glimpse into the life of one of Canada’s leading classical musicians, and Gould’s own recorded words have, through Carroll’s efforts, been preserved as a precious testament to the musician’s life, personality and creative genius. A clearer, more doctrinally coherent analysis of the third-party fixation issue in the case would, nevertheless, have helped to solidify the interpretive legacy of this interesting decision on the materialization requirement in copyright law. Yet, despite its limitations, this case may provide valuable learning opportunities for jurisdictions such as Canada and Australia, and perhaps a focal point for reform as they continue to refine and update their copyright legislation for the modern information society. Footnotes 1 Gormley v EMI Records (Ireland) Ltd [2000] 1 IR 74. 2 Gould Estate v Stoddart Publishing Co [1998] 39 O.R. (3d) 545 (Ontario Court of Appeal). 3 See, for instance, ss 1(1)(a) and 3(2) of the Copyright, Designs and Patents Act (CDPA) of the UK, as well as 17 USC §101 (on the definition of a ‘fixed’ work) and §102(a) of the Copyright Law of the USA, Tıtle 17 of the United States Code. 4 Routine phrases and expressions commonly used in daily life are, on the other hand, less likely to be viewed as original. See Hector MacQueen, ‘“My Tongue is Mine Ain”: Copyright, the Spoken Word and Privacy’ (2005) 68 Mod L Rev 349, 363, where it is noted that routine words and catch phrases ‘with which we navigate our way through daily life may well fall short of even low standards of originality and literariness’. 5 See, for instance, Apple Computer v Computer Edge [1984] FSR 481 (FCA), 521, where it was held that in order for a literary work to convey information or instruction, it must be able to convey an intelligible meaning. 6 Art 2(2) of the Berne Convention for the Protection of literary and Artistic Works (1886) 828 UN Treaty Ser 221, to which most nations are signatories, provides considerable flexibility for members by stating that it shall be ‘a matter for legislation in the countries of the Union to prescribe that works in general or any specified categories of works shall not be protected unless they have been fixed in some material form’. See also Elizabeth White, ‘The Berne Convention’s Flexible Fixation Requirement: A Problematic Provision for User-Generated Content’ (2013) 13 Chi J Int’l L 685, 687, who describes the Berne Convention’s reference to fixation as being expressed in ‘non-binding and arguably unhelpful’ terms. 7 Gould Estate v Stoddart Publishing Co Ltd 39 OR (3d) 545; 161 DLR (4th) 321; 80 CPR (3d) 161; [1998] OJ No 1894 (QL); 114 OAC 178. 8 See Daniel J Gervais and Elizabeth F Judge, Intellectual Property: The Law in Canada (2nd edn, Thomson Canada 2011) 51, noting that there was no copyright in Gould’s unstructured oral responses at the time of the interview because there was no fixation. 9 See Elizabeth Adeney, ‘Authorship and Fixation in Copyright Law: A Comparative Comment’ (2011) 35(2) Melbourne Univ L Rev 677, 696, who also uses the expression ‘immediate peer group’ when comparing Australia’s approach to the notions of ‘work’ and ‘authorship’ to those of other jurisdictions, particularly those with which Australia has strong legal, cultural and historical connections. 10 See, for instance, s 3 of the UK’s CDPA and s 22(1) of the Copyright Act of Australia. 11 It is interesting to note, however, that some critics of the fixation doctrine have suggested that copyright should subsist in any ‘perceptible expression that emanates from the intellect of a person and is intended to appeal to the aesthetic sense or intellect of others, regardless of materiality’. See for instance David Brennan and Andrew Christie, ‘Spoken Words and Copyright Subsistence in Anglo-American Law’ (2000) 4 IPQ 309, 349, who argue that there is nothing inherent in the objectives of copyright law that requires the imposition of a material form requirement. Imposing such a requirement may, in their view, lead to outcomes contrary to the objectives of copyright protection. 12 Paul Torremans, Holyoak & Torremans Intellectual Property Law (8th edn, OUP 2016) 190. 13 S 3(3) prescribes, inter alia, that it is immaterial whether the work is recorded by or with the permission of the author. See also Lionel Bently and Brad Sherman, Intellectual Property Law (4th edn, OUP 2014) 92 as well as Torremans (n 12) 239 (noting that speeches can constitute original literary works once they are recorded, and that it is not necessary for speakers to do their own recording). 14 See Adeney (n 9) 677, 681. 15 But it is noteworthy that s 3(3) also mentions that it does not affect the question of whether copyright subsists in the record as distinct from the work recorded. 16 See §101 of the Copyright Law of the USA, Tıtle 17 of the United States Code. It ought to be borne in mind, however, that federal law does not necessarily pre-empt all aspects of state law. The basic principle enshrined in art VI clause 2 of the US Constitution provides that federal law will prevail over state law in the event of a conflict between the two. 17 As an example, the definition of ‘dramatic work’ in s 2 of the Copyright Act of Canada includes ‘any piece for recitation, choreographic work or mime, the scenic arrangement or acting form of which is fixed in writing or otherwise’ (emphasis added). In the case of musical works, the 1993 amendments of the Copyright Act removed the express fixation requirement for such works. See Gervais and Judge (n 8) 49–50. 18 See for instance section 15(1)(a) of the Copyright Act of Canada, which makes reference to copyright protection of unfixed performances. See also Adeney (n 9) 677, 689, where it is noted that ‘unfixed works’ fall within the concept of authored works in Canada. 19 [1954] Ex. C.R. 382, 20 C.P.R. 75 (Can. Ex. Ct) para 28. 20 See, for instance, Australian Law Reform Commission, Copyright and the Digital Economy, Report No 122 (2013) (published on 13 February 2014), <http://www.alrc.gov.au/publications/copyright-report-122> (accessed 23 April 2018). One of the key areas of inquiry in this report relates to whether the exceptions and statutory licences in the Copyright Act 1968 (Cth) are adequate and appropriate in the digital environment. 21 See s 32(1)(b) of the Copyright Act 1968 of Australia. 22 IceTV Pty Ltd v Nine Network Australia Pty Ltd (2009) 239 CLR 458 (High Court of Australia) (IceTV). 23 See Telstra Corporation Ltd v Phone Directories Co Pty Ltd (2010) 264 ALR 617, 624, [20], referring to dicta expressed by French CJ, Crennan, Kiefel, Gummow, Hayne and Heydon JJ in IceTV (n 22). 24 See Alexandra George, ‘Reforming Australia's Copyright Law: An Opportunity to Address the Issues of Authorship and Originality’ (2014) 37(3) UNSW L J 939, 941. 25 ibid. 26 See Adeney (n 9) 677, 680, 689–91. 27 The estate also sued the photographer and book publisher for breach of contract and misappropriation of personality. However, this article will focus on the legal significance of the copyright action only. 28 See Gould Estate v Stoddart Publishing Co [1998] 39 O.R. (3d) 545 (Ontario Court of Appeal) at [22]. The Ontario Court of Appeal relied on a case from the US, Falwell v Penthouse International Ltd, 215 U.S.P.Q. 975 (1981, Vir.Dist. Ct.) 977. It was held in Falwell that no copyright subsisted in unstructured remarks made to members of the media by the Reverend Jerry Falwell. See also Gould Estate v Stoddart Publishing Co [1996] 30 O.R. (3d) 520 (Ontario Court of Justice (General Division)) at [31]–[32], noting that the interview was conducted in informal, relaxed and casual settings, and Gould’s responses were not part of a private conversation or a structured speech, but instead were ‘offhand comments’ which he knew could find their way into the public domain. 29 Gould Estate v Stoddart Publishing Co [1998] 39 O.R. (3d) 545 (Ontario Court of Appeal) [24]. 30 ibid [26]. 31 ibid [24]. 32 ibid [5] and [27]. 33 ibid [26]–[27]. 34 ibid [26]. 35 ibid [21]. 36 See Gervais and Judge (n 8) 51. 37 Gould Estate v Stoddart Publishing Co [1998] 39 O.R. (3d) 545 (Ontario Court of Appeal) [22]. 38 ibid [23]. 39 The irony of the situation would, of course, be that an oral utterance would no longer be ‘casual’ or ‘spontaneous’ if it had earlier been written down in material form by the particular ‘author-speaker’ delivering it. 40 See Gervais and Judge (n 8) 51, where they write: ‘The later fixation of those conversations in the interviewer’s book was copyrighted expression belonging to the author-interviewer, but Gould’s oral speech at the time of the interview was not protected by copyright because there was no fixation.’ Gervais and Judge do not expressly mention a ‘self-fixation’ requirement. However, they assert that copyright protection was denied because Gould’s expressions were unfixed. This reference to the lack of fixation should probably be interpreted as being unfixed by the speaker, since it would be untrue to say that Gould’s words had not been reduced to material form (in view of Carroll’s avid on-site ‘recording’ activities). 41 See JA Cuddon, A Dictionary of Literary Terms (Penguin Books 1984) 660–61, where stream of consciousness is defined in literary criticism as a narrative method that seeks to depict the multitudinous thoughts and feelings which pass through the mind. 42 Zonen Endstra v Nieuw Amsterdam [2008] EIPR N73 (Endstra Tapes). (Netherlands Supreme Court, 30 May 2008). It should be borne in mind that the Endstra Tapes case was decided under the ‘European’ standard of originality as articulated by the European Court of Justice in Infopaq Int v Danske Dagblades Forening, Case C-5/08 [2009] ECR I-6569 (ECJ, Fourth Chamber). 43 The case concerned whether the contents of an interview by the police with an informant could constitute protectable subject-matter. In overturning the decision of the Court of Appeal of Amsterdam, the Dutch Supreme Court held that the informant’s conversations met the standards for protection, as they reflected human creative choices that were neither banal nor trivial. See also Bently and Sherman (n 13) 110. 44 Such a proposition would also be supported in the case law of the UK. See for instance Express Newspapers v News (UK) [1990] F.S.R. 359, in which it was held that a journalist who had selected extracts from a recorded interview for a report had exercised the necessary skill and judgment for copyright to subsist in the report. See also Torremans (n 12) 239, who notes that reporters who edit a work by selecting parts of a speech and adding their own comments are applying skill, labour and judgment to their report, which constitutes an original literary work and attracts its own copyright. 45 See Redwood Music v Chappell [1982] RPC 109, which articulates the principle that a derivative work can be both original and infringing. 46 See CCH Canadian Ltd v Law Society of Upper Canada [2004] 1 S.C.R. 339, in which the Supreme Court of Canada held that fair dealing was a ‘users’ right’ to be interpreted broadly and liberally, and was not merely an ‘exception’ to copyright protection. 47 The analogous provisions in the UK on fair dealing for the purpose of criticism or review can be found in s 30 of the CDPA. 48 Telstra Corporation Ltd v Phone Directories Co Pty Ltd (2010) 264 ALR 617, 624, [20]. 49 ibid. 50 See Bently and Sherman (n 13) 126, who note that the test for whether a person who contributed to the production of a work can be considered an ‘author’ is similar to the criteria used to evaluate whether a contribution is ‘original’. 51 Bently and Sherman (n 13). See also Gordon J in Telstra Corporation Ltd v Phone Directories Company Pty Ltd (2010) 264 ALR 617 at 685, who notes that: ‘Authorship and originality are correlatives. The question of whether copyright subsists is concerned with the particular form of expression of the work.’ 52 See, for instance, ss 29 and 30 of the UK’s CDPA (which protects activities such as research, private study, criticism, review and news reporting) and §107 of the US Copyright Act. 53 See, for instance, s 30(2) of the UK’s CDPA. 54 See Zonen Endstra v Nieuw Amsterdam [2008] EIPR N73 (Netherlands Supreme Court, 30 May 2008). 55 Infopaq Int v Danske Dagblades Forening, Case C-5/08 [2009] ECR I-6569 (ECJ, Fourth Chamber). 56 See Bently and Sherman (n 13) 110, where it is argued that there is ‘no reason to think that any different result would have occurred from applying the European standard of “author’s own intellectual creation”’. 57 ibid. 58 [1900] AC 539. 59 The position of the reporter who writes down a spontaneously delivered speech but who does not have any sound or film recording was expressly considered by Lord Beaverbrook in the debates on s 3(3). See, for instance, Hansard, HL, Vol 495, cols 610–611 (29 March 1988). Interestingly, it was noted by Lord Beaverbrook, in reference to opinions expressed by Lord Merton and Lord Williams, that the door may have deliberately been left ajar so as to allow courts the flexibility to decide in a particular case whether a reporter had expended sufficient skill and judgment to attract original copyright in the transcript. 60 See Nigel P Gravells, ‘Reporter’s Copyright and Sound Recordings: A Reply to Professor Derclaye’ (2013) 2 IPQ 91, 92. 61 See Estelle Derclaye, ‘Debunking Some of UK Copyright Law’s Longstanding Myths and Misunderstandings’ (2013) 1 IPQ 1, 10: ‘Section 5A is broad and includes a record of a speech by hand as it is no less a sound recording than a record of the same speech in analogue or digital audio format.’ 62 See Gravells (n 60) 91, 95. 63 ibid. 64 See Justine Pila, ‘An Intentional View of the Copyright Work’ (2008) MLR 535, 549, 557. 65 ibid 549. 66 There is a strand in the copyright literature which calls for the removal of the fixation criterion as a requirement for copyright protection. See for example Brennan and Christie (n 11) 309. © The Author(s) 2018. Published by Oxford University Press. All rights reserved. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)

Journal

Journal of Intellectual Property Law & PracticeOxford University Press

Published: Oct 1, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off