The Musical Work in Copyright Law

Abstract

Music is a special art form and a special object for copyright protection. In contrast to other, more static, forms of art, such as most of the visual arts, music is fashioned time and therefore constitutes change and movement; it also exists only in its performance. This article shows how copyright constructs a normative device, the ‘musical work’, to be able to protect this fleeting and transient art form. The article also discusses how music is analysed to ascertain which of its elements obtain copyright protection against infringement, and whether a ‘hearing test’ is an adequate method for establishing infringement.

I Music and musical work: Legal definitions

‘Works of copyright’ are not factual descriptions adopted by the law, but normative legal creations which particularly resemble works of art (in a broad sense) as they are generally understood in the real world. An artistic work, according to law, includes paintings, drawings or sculptures in the ordinary sense, independent of the specific meaning in the law. However, graphs, charts, diagrams, plans or certain photographs that would not commonly be regarded as artistic in nature are also covered by the category of ‘artistic work’.¹ The large category of literary work comprises some prominent examples of artificial legal categories, such as the computer program, which is supposedly only another instance of a literary work alongside novels or poems.² Yet the musical work is something special, a fact which lawyers do not seem to give much consideration. What is protected by copyright is strictly speaking not music, but an artificial legal construct that is normatively defined as a true representation of the music. The protection mechanism is a legal creation of a representation of what is commonly regarded as music, and this representation subsequently becomes the reference point for copyright protection as if it were the music itself.

In this article it will not be argued that copyright does not manage to protect music adequately because of its particular nature. It will rather be shown that the legal creation of the ‘musical work’, and the assumptions behind this creation, enable acceptable protection in most cases: it is worth looking into this construct that lawyers employ without much awareness of its nature. Admittedly, some legal practitioners may dismiss as irrelevant more theoretical discussions about why and how something operates if it works reasonably well, but such an approach is too unsatisfactory from an academic perspective. Problems, arguably irresolvable ones, nevertheless arise when less traditional forms of music are concerned, such as serial music and aleatoric music. However, since the most commercially relevant forms of music are in the area of pop music (broadly understood), legal actions are brought almost exclusively in that sector. As pop music, despite its generally vanguard attitude, usually operates with more traditional techniques of the musical craft than modern classical music, problems of copyright protection which would be raised by modern classical music do not come before the courts. There are court decisions in relation to musical sampling,³ which has been an innovative form of music-making (initially particularly in rap and hip-hop), but that issue is strictly speaking not one of the ‘musical work’ and its legal construction in copyright law, because sampling cases are infringement and fair use cases which presuppose the protected pre-existing musical work.

It is mostly pop music that has an important commercial value, and in copyright countries the music entertainment industry is stronger and litigation with regard to music more common than in droit d’auteur countries of the European continent. This is a reason why the present article will focus significantly more on copyright systems, although droit d’auteur systems will frequently be referred to where relevant.

Before the specific protection method of ‘musical work’ is considered in more detail, it is worth looking at some definitions of ‘musical works’ in several copyright and author’s rights laws as a starting point. When considering the copyright systems first, in the US, the Copyright Act 1976 refers to ‘musical works’ as one category of works which are the subject-matter of copyright, whereby any accompanying words are included in the musical work-category.⁴ Otherwise the ‘musical work’ is not defined further. Obviously, from a musician’s or composer’s perspective, ‘music’ does not include accompanying words, so factual ‘music’ and normative ‘musical work’ overlap but do not coincide.⁵ In the UK, the musical work is normatively separated from the literary work of the text (lyrics), a dramatic action or dance. Under the Copyright, Designs and Patents Act 1988 (CDPA 1988), a ‘musical work means a work consisting of music, exclusive of any words or action intended to be sung, spoken or performed with the music.’⁶ There are no further indications as to what ‘music’ actually is, of which the musical work supposedly consists. Although, in contrast to the US, the non-musical category of text/poetry is excluded, the ‘musical work’ category of the UK CDPA still does not coincide with ‘music’ in a musician’s understanding, as we shall see.⁷ The Irish approach is that of the British one.⁸

The statutes of the author’s rights systems do not give much guidance about musical works either. In France, ‘musical compositions with or without words’ are protected, as well as dramatic-musical works (typically operas or musicals),⁹ without any further explanation about the nature of the musical work. The comprehensive understanding of the musical work that can include words or action is reminiscent of the United States. The Italian definition is almost identical to the French one.¹⁰ The Dutch regulatory approach is also similar.¹¹ Polish law restricts the musical work to music with or without words.¹² The German Author’s Rights Act (Urheberrechtsgesetz) lists ‘works of music’ as one category of protected works without further detail.¹³ The same applies to Austria (‘Tonkunst’, literally ‘musical art’).¹⁴

In Belgium, the musical work is subsumed in the general category of ‘literary and artistic works’.¹⁵ In Sweden, musical and works of scenography are in one category.¹⁶ This is interesting, because one would probably expect literary, artistic, scenographic and musical works to fall into separate groups: one would think that scenography, for example, is closer to the artistic work, where the factor of time is less relevant than in music, and yet scenography is obviously connected to performance and the dramatic work or the theatre play, and therefore the time element matters, similar to music. It will be explained later that the ‘musical work’ has nevertheless special features which distinguish it from the artistic work, the dramatic work, the film and arguably even the choreographic work.¹⁷ The legal technique of categorisation is, however, not as relevant as it seems, because in droit d’auteur countries the categories of work are normally a demonstrative, not conclusive, list (Austria is a theoretical exception),¹⁸ so that the work in question need not fit into a specific category to obtain protection. The CJEU echoed that principle.¹⁹ In the UK, however, there is theoretically a fixed list of copyright work categories (‘category approach’ or ‘pigeonhole approach’),²⁰ but generally with insignificant practical differences to the situation in droit d’auteur countries.

Certain features become apparent from the definitions of ‘musical work’ of both the copyright and the author’s rights worlds. The normativity of the term ‘musical work’, departing from common use, becomes particularly obvious in cases where, depending on the individual jurisdiction, the definition of ‘musical work’ can comprise text or action in connection with the music: a composer or musician would not normally regard lyrics, acting or dance as parts of the music in an ordinary or artistic sense – more precisely a Western composer, because, for example, in the many forms of African traditional and modern music the distinction between poetry, music and dance is much more blurred than in the European traditions.²¹ Furthermore, there is no legal definition of music. The copyright countries sometimes offer an exclusive definition (‘music without …’ or ‘exclusive of …’) but leave the question of what constitutes music to statutory interpretation. Both in copyright and in author’s rights countries, the definition of ‘musical work’ is referred to judicial practice.²²

It is commendable that copyright statutes across the world refrain from a definition. Music is generally more frequently a work of art in the usual sense. There are rarely purely utilitarian works in music, like compilations, diagrams or charts that are included in the literary or artistic work categories, so that a work of music typically contains an element of artistic novelty or originality – in this context not necessarily to be understood in the technical legal meaning of these terms. It would be counterproductive if a legislator tried to pre-empt artistic novelty and development or freeze artistic freedom and innovation with a certain definition. ‘Music’ is something to be determined by the courts, who need to rely on the understanding of the music world for their assessment. However, such an expert opinion by a musician or musicologist, as opposed to a layperson’s hearing experience, can cause problems.²³

Yet, even if we ascertain with sufficient clarity for the purpose of the law what music is, that definition of ‘music’ does not provide a definition of ‘musical work’. ‘Music’ in reality and ‘musical work’ in law are not the same. The reason for that lies in the specific nature of the art form of music itself.

II Making music ascertainable for protection

1. The ascertainable object of the property right

The major problem for the copyright protection mechanism is that music is fashioned time. But the copyright protection system is based on property rights which ultimately presuppose static space. Copyright (in the following, this term encompasses author’s rights) is a property right, either formally, such as in the US,²⁴ UK²⁵ or France,²⁶ or functionally-economically (called ‘Immaterialgüterrecht’, ‘intangible goods right’), as in Germany²⁷ and Austria.²⁸ Property rights in general are characterised by a fixed list or numerus clausus (at least in fact, even if there is no codified system),²⁹ so that contracting parties cannot create their own property rights but have to resort to a catalogue of types of property rights that the law provides. In this way, third parties who are bound by this right erga omnes can identify it: in intellectual property, these are particularly patents, trade marks, copyright and designs, created and defined by statute. Furthermore, to qualify as a property right, it must have a certain permanence or stability. The famous dictum in English law in National Provincial Bank v Ainsworth³⁰ by Lord Wilberforce gives good heuristic guidance:³¹

‘Before a right or an interest can be admitted into the category of property, or of a right affecting property, it must be definable, identifiable by third parties, capable in its nature of assumption by third parties, and have some degree of permanence or stability.’

This case was in relation to the assessment of the nature of a right in land law, thus referring to tangible property. With intangible property, such as copyright, the same idea applies, but the practical implementation is more difficult. A copyright protects (and is a property right in), say, a painting, that is, the specific realisation of an artistic work, as it materialises in paint on a canvas. That painted canvas is subject to rules on tangible and moveable property separately, but the copyright is denoted indirectly by that as well, although the copyright is a separate property right and typically in separate ownership, with separate transferability of that ownership.³² The copyright benefits from the static nature of this reifier: the painting is a stable, permanent physical object, perceivable by anyone’s own unaided senses. But music? What can serve as a (physical) permanent reifier that denotes the copyright/property right in a piece of music? And does such a permanent reifier permit the music to be perceived by one’s own unaided senses?

2. Music in contrast to other forms of art

As already said, music is fashioned time, or if one is inclined to make an aesthetic statement, embellished time. Music is a temporal structure, that is, an intrinsically time-bound process, not a structure in time, that is, merely subjected to time, like a typical artistic or architectural work. That means music is not static at all but moves and changes all the time – if time were to stop, music would cease to exist. One can see music as a phenotypical expression of time (or, following Husserl, an example of the inner awareness of time³³), like the apparent cyclical movement of the fixed stars, a ticking clock, the growing of a plant and other changes.³⁴ In such a case, the reference point for a property right becomes difficult to establish. A work of architecture is static and does not change. Music is a questionable reference point for such a permanent right: after all, third parties bound by the copyright property right erga omnes need to be able to ascertain wherein the right lies and where its limits are. This is a problem with all, particularly intangible, property rights: for example, a trade mark right must be clearly ascertainable for third parties, which can become difficult for trade marks that are not graphically represented (such as olfactory marks),³⁵ although the express requirement of graphical representation has recently been dropped in EU trade mark law.³⁶ An artistic work represents its appearance and features, and thereby its limits of protection, visually and graphically. A painting which appears visually sufficiently different from a pre-existing painting will not infringe copyright.³⁷ The similar or different features in question are static and immediately ascertainable.

However, the visual arts have great problems with depicting and capturing movement and change. They may show a (dramatic) moment frozen in time, particularly in the Romantic period, such as some paintings by J. M. W. Turner (1775-1851) that depict storm and boats in a rough sea. In the first third of the twentieth century, the cubist, futurist and surrealist movements experimented with the phenomenon of movement, change and fluidity in painting and sculpture, such as works by Marcel Duchamp (1887-1968), ‘Nu descendant un escalier n° 2’ (1912), Umberto Boccioni (1882-1916), ‘Forme uniche della continuità nello spazio’ (1913),³⁸ or René Magritte (1898-1967), ‘La Durée poignardée’ (‘Time Transfixed’, actually: ‘Time Stabbed with a Dagger’) (1938). But all these experiments only emphasise the attempt at going beyond the limits of the art form and rely on the spectators to mentally complete the movement which the artwork hints at. An exception was perhaps František Kupka (1871-1957), who wanted to express a genuine visualisation of the time experience in the three dimensions of visual art.³⁹ An illusion of movement, for example through Eadweard Muybridge’s (1830-1904) ‘Zoopraxiscope’ discs of the 1880s and 1890s, or the moving images created by the swiftly passing frames of a film (in the old days before film recording became taped and digitised), does not change that essentially static nature of the elements of the artwork that create the illusion.

The contrast between the static work of the visual arts and the time-dependent flow of the work of music is obvious.⁴⁰ A painting, a sculpture or a work of architecture is static, space-bound, visual. But what about forms of art where time evidently plays an important role: the dramatic work (e.g. a theatre play), a recital of poetry or a speech (literary works), happenings, events and performances as forms of visual art, video art, the film (cinematographic art), dance and choreography? Like music, they contain change based on the flow of time. What is (or represents?) here the property object to which copyright attaches? Or could one also say that the nature of the art form is not really relevant for the protection mechanism of copyright, because copyright protects the abstract intellectual creation as it may manifest itself in the real would but not that manifestation as such, so that it is not important whether the object of protection is static or time-bound, or even fashioned time, as with music?

As to the last point, copyright does protect intellectual creations, but their manifestation, being perceivable by the human senses, is a necessary prerequisite for protection and the reference point for the right. This manifestation does not have to be in the sense of fixation, as the copyright systems require, but it must be an emergence in the physical world.⁴¹ How this manifestation materialises reveals the nature of the art form and its potential differences to music. With paintings, sculptures and works of architecture, the manifestation is invariably also fixation in the copyright sense.⁴² Dramatic works (in copyright laws where this separate category exists)⁴³ are manifested through writing and thus resemble literary works (which can also be recited). It is a philosophical question what the text is: the actual work itself (similar to the situation with artistic works), or (like in music) an instruction for a performance. The latter is certainly more relevant to a dramatic work than to a novel, for example. However, in relation to the dramatic work, and even more so with regard to the literary work, the text represents (or: is) the work itself, and the scope of the performance is comparatively narrow, determined by the text as its borders. One could surmise that the written text is not a very accurate record for the rendering of the literary or dramatic work. However, the text of the language is far more precise than the music notation. The written text of a poem or a play gives a narrow gauge for a performance, while musical notation is much more imprecise as a basis for the performance. We will see that the score is not the music itself.⁴⁴

Where performance plays a much more dominant role is in dance and dance choreography. The concept of dance tends to resist fixation⁴⁵ and shares with music that it can exist perfectly well without recording. It also seems to be that copyright is not at the forefront of dancers’ concerns; they are more interested in the creative process than in the protected product.⁴⁶ If a choreographer is considered as author and copyright owner, it can become complicated to ascertain wherein the protection actually lies. That concerns methods, styles, techniques and traditions as well as the personal (choreographer’s and possibly dancer’s) input and raises questions of originality and co-authorship.⁴⁷ Case law is scarce.⁴⁸ Countries which have choreographic works as a separate protected category⁴⁹ are also confronted with these problems of ascertainment of the existence and extent of the right. France resorts to the requirement of copyright-like fixation (against the general droit d’auteur principle) to ascertain and prove the choreographic right in question.⁵⁰ A sophisticated philosophical approach may consider the visual imagery of dance as something that can be dissected into distinct, and ultimately static, phases or stills which together produce the performance, while such a dissection to reach a ‘static frame’ is not possible for music, not even down to a single tone.⁵¹ Despite dance being real movement and change, it can provide static moments of rest and immovability, even torpor (like the tableaux vivants of the early twentieth century⁵²), because it is visual. In any case, it appears that while dance and music share the performance as the central work-creating element, the technique of the legal construct that enables the protection under copyright law is much more elaborated with music than with dance.

The term ‘performance’ seems vaguer than one would perhaps think: for example, in English law the Court of Appeal said that a film that used a jump-cutting technique for editing cannot be performed physically and therefore cannot be a dramatic work, while the showing of that film itself is a performance (therefore making the film a dramatic work).⁵³ This interpretation is doubtful and confuses the meaning of performance:⁵⁴ a film can contain a performed dramatic work (the film script), but it is not a performance itself. Films are moving images, starting from experiments in the nineteenth century, like those by Eadweard Muybridge, and it is irrelevant whether the recording of the images is on celluloid or electronic/magnetic. In every version they are recorded static images, and the time-bound movement is created by an illusion relying on the physiological inertness of the eye. But each frame is static and in this microstructure not that different to pictures in the visual arts. The same presumably applies to video art, as long as it contains a film sequence as it is traditionally understood – but any experimental art can only be assessed individually.

Music, in turn, consists genuinely of movement and change throughout and cannot provide a static moment which would be a safe reference point for a property right and its extension or limits. This even applies to the microstructure of music. Physically-acoustically, even a single tone, which would only be a small segment of the complete work of music anyway, cannot be ‘static’ and detached from time, otherwise it cannot exist. In reality even a single tone makes micro-movements in frequency and amplitude in fractions of a second that are characteristic of the individual musical instruments, as a spectrogram would reveal.⁵⁵ Only one sound is, apart from its time-dependency, entirely immutable, the most artificial sound of all: the sine wave of the pure tone. Music is a living art, like a breathing animal: if there is no change, or put differently, no time, the animal is dead.

3. The need for the performance of music

This ‘vitalist’ aspect of the living art of music becomes apparent in the need for performance: there is no music without its performance. While we can look at a painting with our unaided senses, we usually need highly skilled musicians to realise the piece of music. The musicians create and recreate the piece of music with each performance, a slightly different individual animal of the same species: no cat is exactly the same but all belong to the same kind. The same applies to every performance of Mozart’s piano concerto No. 27 in B-flat. A sound recording of that piano concerto is strictly speaking a recording of a past musical performance, not the music itself, like a film recording of a cat, not the cat itself. The sound recording or phonogram is not an arrangement of sounds, thus not an original creation or its performance, but the fixation of such sounds as they occur in a performance which can then be replayed.⁵⁶

Music is therefore an organisation of sounds by the human mind that needs to be performed by human beings to be realised or brought into existence in a lapse of time. Music exists only in and through its performance, and it never exists at once in full (unlike a work of visual art⁵⁷) but emerges and vanishes gradually in the flow of time. So we do not have a ‘full picture’, a defined and static property object to which the property right of copyright clearly relates and which denotes the extent and limit of the right that third parties have to respect without having a contractual relationship to the titleholder. In addition, we do not have the property object at the outset, static or not, but have to recreate it every time through the act of performance by specialists to bring it into existence. Our unaided senses alone would not make it perceivable.

Copyright law separates performance from composition, the latter being protected by musical copyright, the former (depending on the jurisdiction⁵⁸) by performers’ rights,⁵⁹ although court decisions have obviously recognised the central importance of the performance for the making of the work of music, for example, in the US case Capitol Records v Mercury Records Corp.:⁶⁰

‘I also believe that the performance or rendition of a “musical composition” is a “Writing” under Article I, § 8, Cl. 8 of the Constitution separate from, and additional to, the “composition” itself. It follows that Congress could grant the performer a copyright upon it, provided it was embodied in a physical form capable of being copied. The propriety of this appears, when we reflect that a musical score in ordinary notation does not determine the entire performance, certainly not when it is sung or played on a stringed or wind instrument. Musical notes are composed of a “fundamental note” with harmonics and overtones which do not appear on the score. There may indeed be instruments – e.g. percussive – which do not allow any latitude, though I doubt even that; but in the vast number of renditions, the performer has a wide choice, depending upon his gifts, and this makes his rendition pro tanto quite as original a “composition” as an “arrangement” or “adaptation” of the score itself […]. Now that it has become possible to capture these contributions of the individual performer upon a physical object [a sound recording] that can be made to reproduce them, there should be no doubt that this is within the Copyright Clause of the Constitution.’

This passage contains a clear recognition of the role of the performance for the work of music beside the composition and also refers to the importance and limits of the musical text or score. However, when it comes to the ‘fixation’ of the performance through a sound recording, the reasoning becomes a little imprecise. The written text of the music, the score, is not the music, but a building instruction for the performance that creates the music. However, it is the written text of the music which enables the protection of the composed music as a musical work of copyright: the score is the reified evidence of the composed musical work which contains the music as the art form. With this reification, the work of music becomes fixed, an object to which the property right of copyright can attach and whose limits of protection can be ascertained. This is effectively a transmutation from the transient and evanescent time-bound art form of music to a static space-based object, similar to a work of the visual arts or of architecture. The copyright-property commodifies the fixed object that represents the music without being it: the music itself remains the live performance based on the score, if there is one.

The necessity of fixation of the musical work as a static medium that represents the music as a living art form is reflected in the requirement of recording or fixation of the work for copyright protection in copyright countries.⁶¹ Author’s rights countries do not have this requirement,⁶² hence an improvisation or impromptu playing of a piece of music is protected by author’s rights as such without any fixation/recording⁶³ (although in an infringement action the composer will have to prove the ‘features’ of his or her work of improvised music, and that will hardly be possible without a recording). Nevertheless, author’s rights countries also require the work to be perceivable by the senses of third parties: the mere thoughts of the composer which have not been expressed to the outside world obtain no protection.⁶⁴ Mozart was famous for having been able to compose whole works in his head,⁶⁵ but for copyright purposes the composer must either write down the work or have the performance recorded on tape (copyright countries’ recording requirement), or played without a recording (sufficient in principle for author’s rights countries). In any event, the work has to enter the outside world.

Writing down or recording is, however, not constitutive to the art form of music – it can perfectly well exist without any fixation, although notation is characteristic of Western art music and its polyphonic nature.⁶⁶ Works of visual art, however, need fixation; therefore, in the UK CDPA 1988 the recording requirement is stipulated for literary, dramatic and musical works, but not for artistic works,⁶⁷ because without a recording of some kind (paint strokes, carving out of a stone, etc.) the artistic work would simply not exist.⁶⁸ In non-Western cultures, performance without a written score as its basis is very common: most examples of non-Western, particularly African, classical or traditional music were passed on to future generations and preserved by way of oral tradition.⁶⁹ The same applies to the techniques of performances of (North) Indian classical raga music and its methods of communication with the audience,⁷⁰ and even Western performance techniques in baroque and classical music were handed down from teacher to pupil and sometimes preserved in contemporary or modern textbooks which are now consulted for an ‘authentic’, historically informed practice of performances of baroque music, for example.⁷¹ Where there is a (Western) art performance or happening (which breaks down classical art categories between visual art, music, literature and dramatic works anyway), any attempted copyright protection would face the same problems as with music as a non-static, evanescent art form.⁷²

4. Performance of music as its initial creation and joint authorship

One can see that performance has a role not only in the recreation of the music, but potentially also in the initial creation of the music. In the latter case, the performance is not an implementation of the existing building instruction of the score, but can be the creation of the musical piece itself for the first time. Here Learned Hand J’s reasoning in Capitol Records v Mercury Records Corp. becomes somewhat confused: he separates the composition and its rendering as performance and makes both separate copyright works if they comply with the fixation requirement (also in the form of a sound recording). In fact, however, a performance is the carrier and realising force of the composition and renders the copyright work; it is not an independent (musical?) work and separate from the composition. The same is true of musical pieces that are created for the first time through a performance (improvisations): in this case the performance creates and shapes the new music as the object of the copyright protection as a musical work. That can also apply to (perhaps ‘creative’ or ‘eccentric’) performances of existing music, such as ‘covers’ in the pop music sector or in classical music, where the standard of performances is normally more rigid, with less interpretative freedom: if the performance moves so far from the performed work that it obtains independent copyright-originality and becomes a separate work of music, it attracts copyright protection as a musical work (not as a performance) in its own right, provided the distance to the pre-existing work is sufficient and only the (unprotected) ideas or concepts of the pre-existing work can be considered as an influence.⁷³ Depending on the circumstances, such a performance can be the creation of a new work of music, even if it is perhaps an homage to an earlier work, but then the (new) composition aspect matters for copyright purposes,⁷⁴ not that of the performance. These considerations are also relevant in the discussion of variations and arrangements.⁷⁵ Where the performance, within the usual variations, is still a performance of an existing piece of music, there is no copyright protection for the performance as such but only performer’s rights protection as a related right, where applicable. Learned Hand J’s copyright construct of the performance may also have been prompted by the fact that at that time performers and sound recordings (as the fixation of performances) had no protection under US law.⁷⁶

The (re)creation of music through its performance is also a central question in the ascertainment of joint authorship. The practical scenarios can mostly be found in pop music. For example, a band consisting of a lead singer (A), a guitarist (B), a bass guitarist (C) and a drummer/percussionist (D) perform their songs on stage. The songs are written either (1) by A, who writes the lyrics and the music, or (2) by A (lyrics) and A and B (music). In these straightforward situations, in case of (1) A is author and copyright owner of both the literary work of the lyrics and the separate musical work (that depends on the individual jurisdiction’s definition of ‘musical work’),⁷⁷ and in case of (2) A is single author and copyright owner of the lyrics and joint author and owner with B in relation to the music.⁷⁸ C and D are mere performers, and all four, A, B, C, D, enjoy performers’ rights. But are C and D always performers only? What happens if C adds a bass guitar solo to the performances which (within a certain spectrum) remains the same in all performances, and D uses a special rhythm pattern and/or drumming style which is characteristic for the auditory experience of the performed songs, neither of which has been written by either A or B? The typical conflict situation that arises before the courts is that the band members have fallen out and C and/or D claim a share in the royalties, which requires acknowledgement of C’s or D’s joint authorship as the basis of this claim.⁷⁹

Some UK cases show the transition from mere performance to co-composition that leads to joint authorship of the musical work.⁸⁰ In Stuart v Barrett⁸¹ a drummer who joined a band later claimed joint authorship for his contribution. During the making of the song in question, he tried different drum beats and ended up with an off-beat drum pattern which seemed to fit well, and the other band members joined in with embellishments and changes. By the end of the session they had a completed piece of music, without words, which became the song in question. While the court refused to lay down any general rules which would apply to all group compositions, it had, in this individual case,⁸² ‘no doubt about the significance of the drum part to the whole of the work. The work is given shape and drive by the drummer and a good drummer, as I accept the plaintiff is, can significantly influence the whole composition.’

Although ‘the player of tuned percussion might be more readily recognisable as a contributor to a musical composition than a drummer’, a rejection of a drummer’s contribution in principle would be a misinterpretation of a drummer’s contribution to composition in contemporary (pop) music.⁸³ In the first instance decision⁸⁴ of Fisher v Brooker,⁸⁵ the subject-matter was a tune, played on a Hammond organ, for which joint authorship was claimed. The court found that Matthew Fisher, the former Hammond organ player of the pop group Procol Harum, was joint author and copyright owner of the song ‘A Whiter Shade of Pale’ (1967) for his famous eight-bar organ solo introduction which was recognised as a musical work and an original contribution to the song. Mr. Fisher did much more than copy what Mr. Brooker, the defendant, had earlier performed. Mr. Fisher was at most building upon ‘ideas’ used in the course of rehearsal. Mr. Fisher’s organ solo as heard in the first eight bars of the work and as repeated was sufficiently different from ‘what Mr. Brooker had composed on the piano to qualify in law, and by a wide margin, as an original contribution’ to the musical work.⁸⁶ In other words, the claimant Mr. Fisher did not just perform the defendant’s, Mr. Brooker’s, music, but performed and thereby composed an independent part of the work that forms part of the overall piece of music and becomes an original contribution.

For music, an art form that relies on performance for its existence, it is irrelevant whether the music was first constructed as a separate composition (on paper) and then performed, or whether it was created in a performance.⁸⁷ In pop music and jazz,⁸⁸ rehearsals and recording sessions which are effectively or deliberately improvisations and composition sessions are the norm anyway,⁸⁹ but even in classical music Mozart or Beethoven were performer-composers in relation to many of their works or parts of them. Only where the performance is a rendering of music composed earlier (rarely the case in pop music) with no original improvisation, joint authorship of the merely performing musicians can be ruled out.⁹⁰

These cases suggest the following interpretation: the performance crosses the line to composition where an element of the performance fulfils the originality requirement for copyright protection, so that this element of the performance becomes a contribution to the composition of the piece of music as a whole and a part of the musical work for copyright protection purposes (bestowing joint authorship if different authors were involved). The required criteria of originality depend on the jurisdiction in question, such as the originality requirements in the UK⁹¹ or the US,⁹² or the author’s rights definitions of originality, such as a personal contribution or a stamp of the author in France,⁹³ a personal intellectual creation in Germany,⁹⁴ intellectual works of creative character in Italy⁹⁵ or, at the EU level, ‘own intellectual creation’, that is, a minimum level of creativity,⁹⁶ at least for some types of copyright work.⁹⁷ This issue goes beyond the present discussion.

Where court decisions paint a slightly confused picture is when they try to use aesthetic criteria to ascertain whether a contribution by a performer is original for copyright purposes and can therefore confer joint authorship. This happened, for example, in Hadley v Kemp, where it was said that in order not to be regarded as mere performance or interpretation of musical works, ‘the contributions need to possess significant creative originality’⁹⁸ to qualify for joint authorship. In Stuart v Barrett, the court said that ‘a good drummer […] can significantly influence the whole composition’,⁹⁹ which suggests that the drummer’s contribution must have a certain (superior?) quality, presumably deriving from his ‘good’ drumming skills.¹⁰⁰ Could a defendant who is faced with a joint authorship claim then perhaps concede the contribution but also argue that it ruined the artistic quality of the work and in this way defeat the claim?

Difficult as this is may be to implement in individual cases, a legal definition of originality, even in the sense of ‘creative originality’ which is – as an ideal type¹⁰¹ – the more common criterion of originality in author’s rights countries, is not to be influenced by aesthetic decisions of what ‘good music’ is.¹⁰² It is trite knowledge that banal and commonplace tunes are almost always original enough to obtain copyright protection, otherwise much of the music industry would face serious problems. The recent UK case Edward Sheeran v Sami Chokri could be seen as an example of mass-produced music, but originality and subsistence of copyright were rightly not at issue, only infringement.¹⁰³ When Ed Sheeran won the case in April 2022, he released a video statement in which he said that ‘there are only so many notes and very few chords used in pop music; coincidences are bound to happen’.¹⁰⁴ That may not be artistically-aesthetically satisfying¹⁰⁵ but it underlines the ultimate purpose of copyright protection in the music business: the creation of musical works as a legal category, that is, merchantable objects of property.

5. Non-traditional forms of music composition and performance: experimental music

The confrontation with modern ‘art music’ or modern ‘classical music’ quickly shows the limits of copyright protection for music. The problem is, however, a theoretical one, partly because commercially this branch of music is rather irrelevant and so there is no benefit in starting lawsuits,¹⁰⁶ partly because lawyers hardly have any connections with this type of music, so they would not think much about possible copyright problems, and partly because modern composers and performers (in modern music-making these groups have merged more again, as until the nineteenth century) are concerned with creating art, and are little interested in academic divisions of art forms and a fortiori very little concerned about (and with) copyright. A proper examination of this issue would have to be referred to a separate discussion beyond this text, because it is more a question of (legal) philosophy than of copyright law how to deal with the legal institution of copyright if a significant proportion of those whom this institution seemingly protects view it as only dampening and hindering their creativity and art-making. For example, the Fluxus movement,¹⁰⁷ both in relation to visual art and to music (e.g. John Cage), rejects the notion of the work of art altogether, which is supposedly the focal point of copyright protection. Bazon Brock proposed the following about the visual arts in 1976:¹⁰⁸

‘Traditionally, artists produce in the studio and then publicly exhibit the results of their work as works of art. [...] With the maxim “work is a discarded tool” (Werk ist abgelegtes Werkzeug) we aimed at a different identification of the results of the process. [...] I called this for the international linguistic usage Cognitive Tools, German Erkenntnismittel. [...] “Create problems, not works of art” or: “Polemicize vividly instead of disseminating works of art, so that we have something to think about instead of merely admiring what is only dust and paint.”’

John Cage said in a lecture (Composition as Process, II. Indeterminacy) in 1958:¹⁰⁹

‘This is a lecture on composition which is indeterminate with respect to its performance. The Klavierstück XI by Karlheinz Stockhausen is an example. The Art of the Fugue by Johann Sebastian Bach is an example. In The Art of the Fugue, structure, which is the division of the whole into parts; method, which is the note-to-note procedure; and form, which is the expressive content, the morphology of the continuity, are all determined. […] In the case of the Klavierstück XI, all the characteristics of the material are determined, and so too is the note-to-note procedure, the method. The division of the whole into parts, the structure, is determinate. The sequence of these parts, however, is indeterminate, bringing about the possibility of a unique form, which is to say a unique morphology of the continuity, a unique expressive content, for each performance.

The function of the performer, in the case of The Art of the Fugue, is comparable to that of someone filling in colour where outlines are given. […]

The function of the performer in the case of the Klavierstück XI is not that of a colourist but that of giving form, providing, that is to say, the morphology of the continuity, the expressive content. This may not be done in an organized way: for form unvitalized by spontaneity brings about the death of all the other elements of the work.’

These two passages, if consequently implemented, would blow up every concept of musical/artistic criteria that are relevant to copyright protection. To what extent is the aleatoric piece Klavierstück XI (1956) by Karlheinz Stockhausen influenced by John Cage’s Music of Changes (1951)?¹¹⁰ Would this influence be relevant for copyright purposes? What really constitutes the work of music? What exactly is protected in either piece (what about melody, harmony, rhythm)? On which basis could one protect John Cage’s famous 4’ 33’’ piece (1952) at all?¹¹¹ The artistically correct answer to all these questions presumably is: why is that relevant? (And these ‘modern’ examples are already over 60 years old.) This shows that copyright was initially designed for, and conceptually modelled upon, the protection of what today would be named ‘classical music’ from the late eighteenth century onwards – the application of the UK Statute of Anne of 1710 for literary works was extended to music by the decision of Bach v Longman in 1777;¹¹² author’s rights protection then started on the European continent from the 1800s onwards. One could say, in the spirit of Adorno,¹¹³ that copyright was made for older ‘classical’ music, and current popular music is made for copyright – if it is truly popular, that is, a commercially successful musical product. This reveals that the term ‘popular music’ is no artistic category but only a commercial classification, even if some musicologists may consider that differently. Any instance of contemporary music, whether more ‘traditional’ or experimental, style- and category-transgressing, crossover (not only ‘classic/pop’ but also into other categories of art, video art, performances involving dance or visual art creation) will have to be considered individually if a copyright issue comes about. It may not satisfy the academic scholar that he or she cannot develop general rules ex ante beyond broad copyright principles,¹¹⁴ but that is the reality of this art form. One will hope that any deciding courts will have good expert witnesses at hand,¹¹⁵ so that judgments may not damage the art with inept normative pronouncements. It is interesting to see that music, particularly in its avant-garde areas, remains true to its nature: this elusive and anarchistic art form ultimately keeps escaping from any detailed ascertainment by the law.

III. The difference between music and the musical work

In the discussion so far we have already carefully divided music as the art form (with the piece of music or work of music as an instance of this art form) from the musical work as a normative category of copyright law. This distinction is necessary from the perspective of legal theory and the theory of music, but it also has a practical relevance in copyright law. In music theory and philosophy, there is a conception of the musical work which seems to have developed in the early nineteenth century, with the onset of Romanticism.¹¹⁶ This view has been criticised, particularly by early music specialists who claim that ideas of the musical work had emerged in the Renaissance already.¹¹⁷ This controversy among musicologists is not relevant to the legal definition of the ‘musical work’, also because the law normatively decides what qualifies as a musical work, and that may not coincide with a musician’s or musicologist’s understanding. Leaving some music specialists’ subtle academic definitions aside, for the musician a musical work is a separate musical piece: that piece can have a shorter form (e.g. a song, by far the most relevant form in court actions) or a longer one, such as a symphony or violin concerto, but it is always a separate piece containing music as a distinctly expressed use of the craft of the musical art (melody, harmony, rhythm, etc., if applicable),¹¹⁸ and the musical work consists of and contains conclusively this individually expressed incidence of the musical craft. In law, the musical work as a normative category can go beyond that.

The UK case of Sawkins v Hyperion¹¹⁹ is a rare example where the difference between the music (in a musicological sense) and the musical work (in a legal, copyright sense) can be studied. Leaving aside those facts that are not material to the present discussion,¹²⁰ the case had to decide whether a performing edition of seventeenth-century baroque music, prepared by a musicologist (the claimant Sawkins), can obtain copyright protection on which that musicologist’s claim for royalties against a record company for the use of his performing edition in recordings of this baroque music can be based. A musician would presume that the music in question is seventeenth-century baroque choir music which is not (and in fact never was) in copyright. The performing edition is an assembly of the various manuscripts in different libraries to a score for the realisation of the performances, but no new music. Even where the musicologist has recomposed missing parts in the style of the seventeenth-century composer, one may consider this a kind of restoration work, but not new music for which copyright protection could be claimed. No copyright would subsist in the edition as a musical work itself. This is exactly what the defendant recording company Hyperion argued.¹²¹

The court held in favour of the claimant musicologist that his edition of the seventeenth-century music is a musical work and is copyright-protected. For the classical music business this finding has come as a surprise, presumably also because of the wrong assumption that baroque music would now be copyright-protected. However, from a copyright lawyer’s perspective, this decision is straightforward. The claimant created a recorded (fixed) original musical work which the performing edition constitutes and therefore he obtains copyright protection for it (not for the music which this edition contains). The musical work as a legal definition can go beyond what a musician commonly considers as music: a performing edition, not being the music itself in the ordinary understanding but rather a vehicle for the performance of the music, can be a musical work. The court did not make an express terminological distinction between ‘music’ and ‘musical work’, but effectively applied such a distinction, following the reasoning from the first instance judge:¹²²

‘I do not accept the narrow approach advocated by Hyperion [the defendant] as to the type and nature of the work required to attract musical copyright.

Hyperion’s emphasis is on the necessity for composition or re-composition of new notes of music, as in a musical adaptation or arrangement, in order to produce an original combination of sounds appreciated by the ear, rather than on the exacting scholarly exertions of Dr Sawkins [the claimant] on the notation of the scores and the inclusion of performing indications and directions. […]

Patten J. pointed out ([54]) that the 1988 Act did not define what is meant by “music”. As he observed, however, that is what this case is really about:

“... the real issue which divides the parties is whether a musical work includes items such as the figuring of the bass, ornamentation and performance directions or is really limited for copyright purposes to the notes on the score, so that in the case of an existing work nothing less than significant rearrangement of, or significant additions to, the melody will create a new copyright in the edition of a musical work.”

I agree. I am also with the judge in rejecting Hyperion’s contention that Dr Sawkins did not acquire any copyright by virtue of the considerable effort, skill and time spent by him on the task of creating the performing editions of the three works in question. […]’

The musical work therefore includes items such as the figuring of the bass, ornamentation and performance directions, that is, not the notes and sounds (that is, ‘music’) only. The subject-matter of copyright protection is not only original music, here in the form of an adaptation or arrangement of a pre-existing piece of music, but also a performing edition and restoration of old music, whereby the restoration by its very nature does not seek artistic creativity but deliberately submits to the original composer’s personal artistic style, in the same way as a picture restorer would approach the matter. This avoidance of artistic novelty still constitutes originality for copyright purposes (again a normative term that may depart considerably from the usual understanding of originality among artists). That the musicologist’s effort in preparing the performing edition is sufficiently original for copyright protection is not due to a ‘lower’ originality threshold for copyright in the UK (at that time, ‘skill and labour’), but also applies in France, an author’s rights country with a more ‘creativity as originality’ approach. The same musicologist sued on the same facts in France and also obtained author’s rights protection for this performing edition.¹²³ This slightly earlier French decision did not, however, impress the English judges as a persuasive authority.¹²⁴ The French decision is not unusual for author’s rights countries. In art restoration cases, restorers frequently obtain author’s rights protection for their restorations.¹²⁵ Furthermore, critical (scientific) editions enjoy author’s right protection.¹²⁶

Another important aspect of the Sawkins decision was its discussion of the relationship between the music and its written record. As said earlier,¹²⁷ the written text of the music, the score, is the reified evidence of the composed musical work which contains the music as the art form. The music score enables the protection of the composed music as a musical work of copyright because the property right of copyright can attach to it as a fixed item that is not inherently time-bound. The Sawkins judgment does not say that specifically, but it does state clearly that the score is not the music, the art form itself:¹²⁸

‘In the absence of a special statutory definition of music, ordinary usage assists: as indicated in the dictionaries, the essence of music is combining sounds for listening to. […] Music must be distinguished from the fact and form of its fixation as a record of a musical composition. The score is the traditional and convenient form of fixation of the music and conforms to the requirement that a copyright work must be recorded in some material form. But the fixation in the written score or on a record is not in itself the music in which copyright subsists.’

The musical work is represented and evidenced by the record, typically a music score, particularly in Western classical music, but this kind of evidence – which additionally serves as the necessary fixation of the work in copyright countries – can also be achieved in the form of a sound recording of a past performance, which is obviously a much more precise representation of the music than a music score could be. In contrast, the content transported by language and alphabetic characters can reflect a novelist’s or playwright’s decisions and wishes much more clearly than musical notation can do for composers, although modern music notation (more or less fully developed since the early 1600s)¹²⁹ is highly sophisticated. The imprecisions of musical notation are particularly relevant for issues of interpretation. While modern composers tend to fill their scores with a lot of performance directions, composers of the seventeenth and eighteenth centuries only gave very terse instructions,¹³⁰ so that the performer is often left with the pitch and length of the tones only and, perhaps, an indication of the tempi (‘adagio’, ‘andante’, etc.) for the whole piece. A simple example for the problem of notation is the waltz (3/4 time): the standard German type of waltz would be a steady beat of 1 – 2 – 3, while the Viennese waltz would have the distinct rhythm of 1 – 2 …| 3, thus an extended 2 followed by a belated 3, which nevertheless has to come at just the right moment so that it still serves as the third beat in the bar; hence, an extensive rubato would destroy this. Nothing of that sort would appear in the musical notes of a waltz. But one hears the difference immediately if one listens to a Viennese orchestra playing a waltz by Johann Strauss and a German orchestra trying to do that as well. It seems to work with Austrian and Czech orchestras, but not really with German or Hungarian ones.

It is a philosophical question whether the music score (which can also be a performing edition, as we have seen) is the representation of the musical work in law or is the musical work, but the actual music that the musical work refers to and denotes only becomes real in its performance, for which the musical work or music score provides the necessary instructions. What ultimately matters, also in an identification of whether there has been copyright infringement, are the sounds one can hear in the performance, not what is written in the score. For the purpose of musical copyright, the score is not an artistic work: similarity of notation without similarity of the listening experience points away from infringement. However, what exactly is protected in a piece of music depends on the use of the elements or building blocks of the art form of music and their role in the legal assessment of protection.

IV. The elements of music and their relevance in copyright infringement actions

1. The building blocks of music and the idea-expression dichotomy

To ascertain the subject-matter of potential copyright infringement in music, one needs to eliminate those areas and elements which are not protected by copyright, but which constitute the building blocks of the architecture and craft of the art of music in general, the toolkit of composers and performing musicians, as it were. These elements of what constitutes Western classical music, pop music and jazz are: tone¹³¹ (pitch, timbre and dynamics which are determined acoustically-physically), melody, rhythm, harmony, counterpoint and form (which are all determined artistically-aesthetically).¹³² By no means all criteria may appear in any one piece: for instance, counterpoint as such is found in classical music mostly until the end of the eighteenth century, but even in pop music the (counter)movements of voices ultimately hail from choir music and counterpoint. Particularly in the field of pop music (the most important area for litigation) the following criteria have also been suggested: harmonic and melodic design, rhythmic design (sound colour or timbre, arrangement, dynamics), structuring (including phrasing, repetitions).¹³³ The tone, with its pitch, timbre and volume (dynamics), can relatively easily be recognised and ascertained as well as distinguished in musical works that are compared in infringement actions. Furthermore, they are absorbed in the category of melody. As a working definition, ‘melody’ is an organised sequence of sounds which are perceived as intrinsically connected, or, following the German nineteenth-century music theorist A. B. Marx (1795-1866), melody¹³⁴ can be defined as a tonally and rhythmically organised series of sounds.¹³⁵ The melody is practically the most relevant, but not the only, feature considered in musical copyright infringement actions.¹³⁶

These elements or building blocks¹³⁷ constitute what copyright lawyers would call ‘ideas’, ‘techniques’, ‘methods’ in copyright law, and these techniques of composition cannot be protected. This is a case of the idea-expression dichotomy, a concept that is well established in copyright countries,¹³⁸ but applies in effect in author’s rights countries of the European continent as well.¹³⁹ The idea-expression dichotomy also forms part of national¹⁴⁰ and international¹⁴¹ legislation, and features in the EU Software Directive¹⁴² and in CJEU decisions,¹⁴³ one of which recently indicated that the CJEU in its application of the idea-expression dichotomy is beginning to transcend the Software Directive.¹⁴⁴ Examples of ‘ideas’ and ‘expressions’ in music are: the use of a harmonic progression from tonic to dominant and back to tonic, the use of two voices or parts in contrary motion or the construction of a musical piece in A-B-A-C-A form (a rondo form) does not entitle the composer to copyright protection. However, these techniques combined with a certain rhythm and a distinct melody will typically be protectable ‘expression’.¹⁴⁵ Thus, progression of chords/harmony, typical rhythmical patterns and commonplace melodic motifs are unprotectable ideas, while melody (and rhythm) using a harmony/chord progression or a melody (and harmony) using this typical rhythm would normally be protectable expression. Melody (or more broadly, a sequence of sounds) is usually the most important criterion.¹⁴⁶ If work B shares the same techniques (‘ideas’) of harmony and rhythm with work A but uses them in a different way (in particular a different melody), work B does not infringe the copyright in work A: the ideas match, but not the expressions. According to the merger doctrine, where ideas and expressions merge, there is no protection available.¹⁴⁷ In that case any creative choices which could confer originality¹⁴⁸ are impossible. Merger is a very rare situation in music.¹⁴⁹

It has just been said that the idea-expression dichotomy applies in effect in author’s rights countries as well. However, divergent and not always consistent terminology may lead to the erroneous conclusion that the idea-expression distinction is of limited relevance in author’s rights countries, for example in Germany.¹⁵⁰ It does not alleviate the problem that the meaning of ‘idea’ cannot be determined precisely in the Anglo-American systems either.¹⁵¹ ‘Idea’ can best be defined (loosely) with regard to the specific art form in question (and examples from it), as for music above. ‘Idea’ refers to method, technique, ‘grammar’, style, musical concepts as analysed by music theory or the craftsmanship aspect of making art, with its rules and traditional restrictions.¹⁵² ‘Idea’ does not denote the possibly creative idea in the composer’s head¹⁵³ (which, if detailed enough, may already be an expression that has to materialise in the outside world to obtain protection:¹⁵⁴ either informally, by playing/improvising the piece, or specifically, through fixation as in the copyright systems). ‘Idea’ and ‘expression’ do not correspond to the terms ‘form’ and ‘content’ (‘Form und Inhalt’) either,¹⁵⁵ a dichotomy which has frequently been used in German law in the past but is much less relevant today.¹⁵⁶ ‘Content’ is often also ‘idea’, particularly if it is information, a conventional turn of phrase (also musically), something commonplace without any detectable individual character – it always depends what the content in question consists of. One can see the connection to originality, a concept that should actually be separated from idea-expression in theory.¹⁵⁷ But it is largely the original use and combination of the building blocks of music (‘ideas’) – an original individual use that refers back to the maker-composer and leads to copyright protection – which means music creation in reality. A subset of that idea-expression relationship to originality can also be the German ‘distance theory’ (Abstandslehre) with regard to adaptations of works, whereby the second work must be sufficiently remote from the first work to avoid infringement, so that the first work fades away in the second work.¹⁵⁸ This is typically (not always) the situation where the second work uses ideas (in the copyright sense) but not expressions (denoted particularly by the appearance of creative individuality) of the first work, and therefore the second work does not constitute an infringing adaptation, but is a separate non-infringing new work in its own right, inspired by the first. Pre-existing works always have influenced artists (and they usually learn their craft from these works¹⁵⁹), and some of their works may be adaptation, some fruits of inspiration.

Of course, all that sounds neat in theory, but in a concrete musical copyright infringement action the necessary ascertainment may prove horrendously difficult. In particular, it is practically impossible to establish general principles as to what extent an expression must be reduced notionally to determine solely the unprotectable idea in music,¹⁶⁰ a problem resembling the Kantian ‘Ding an sich’.¹⁶¹ Nevertheless, the principal rule stands that a work that (actively) copies a substantial part¹⁶² of a musical expression, not just of the idea, from another work infringes that other work’s copyright. The problem remains how one ascertains in conflicting musical works whether there is similarity and whether a substantial part has been taken.

2. Ascertainment of infringement and the hearing test

The way in which expressions in two different musical works are compared and similarities are established is through a hearing test. This, however, poses certain difficulties.¹⁶³ The usual test in the US and other jurisdictions, used without much reflection in the courts, is the lay listener test:¹⁶⁴

‘The test to be applied […] depends not upon external criteria, but instead upon the response of the ordinary reasonable person to the works. […]. “Analytic dissection” and expert testimony are not called for; the gauge of substantial similarity is the response of the ordinary lay hearer.’

Judges’ statements are similar in other decisions: ‘when the two songs were played to us, it was immediately apparent, to me at any rate, that the effect on the ear was one of noticeable similarity’;¹⁶⁵ ‘I listened to some of the tapes produced in Court and I am in no doubt about the significance of the drum part to the whole of the work’;¹⁶⁶ ‘The link passage is, to my ear, an original and cleverly worked adaptation of the song’s melodic line’;¹⁶⁷ ‘I accept that [the claimants’ and the defendants’ musical phrases] involve a vocal chant of some kind, I do not agree that the [claimants’] phrase is in a low register: to my ear both the low and high harmonies can be heard equally.’¹⁶⁸ Or, especially if copyright subsistence is at stake: whether there is ‘individual aesthetic expressiveness’ which gives the work the necessary originality, that ‘shows itself also for the layman in the field of music in the fact that he grasps this melody on repeated hearing as known and assigns it to the plaintiff’.¹⁶⁹ The music (in that last case Stevie Wonder’s Happy Birthday song) operates apparently as a kind of signature tune here, as an indication of origin, similar to a trade mark. However, in the reasoning of the courts, the actual comparison of songs does not rely on the lay listener only but frequently contains significant musical expert analysis, particularly for eliciting what the unprotectable musical building blocks or ‘ideas’ are,¹⁷⁰ and potentially also where elements of the work are not sufficiently original and lack protection.¹⁷¹

At the EU level, the Pelham decision gives some indirect indication about the listener test that the CJEU seems to have in mind,¹⁷² at least in relation to sound sampling.¹⁷³ Pelham copied, that is, electronically sampled, about two seconds of a rhythm sequence from the song Metall auf Metall by the music group Kraftwerk for his song. Following a reference from the German Bundesgerichtshof (BGH), the CJEU decided that the reproduction of even a very short sound sample of a phonogram must be regarded as a reproduction ‘in part’ of that phonogram within the meaning of Art. 2(c) of the EU Information Society Directive¹⁷⁴ and is therefore subject to the producer’s permission as the exclusive right holder. If, however, the user uses a sound sample in a modified form unrecognisable to the ear in a new work, then this is not a ‘reproduction’ in the sense of Art. 2(c) and no permission is needed. Freedom of the arts allows an artistic expression in the form of a sample taken from a phonogram and modified to such an extent that the sample is unrecognisable to the ear in the new work as a distinct artistic creation.¹⁷⁵ The question is: unrecognisable to whose ear? The lay listener’s or the expert’s? The Advocate General’s opinion does not refer to a listener test at all.¹⁷⁶ The CJEU does, only briefly, and since there is no further information, the test appears to be a lay listener test, in practice presumably exercised by the judges. However, it can arguably not be inferred that expert listeners are always ruled out.

Craig and Laroche conclude, having composed music containing elements of other well-known musical pieces¹⁷⁷ as an experiment for their critical study of the lay listener test, that ‘Without sufficient appreciation of the musical significance of apparent similarities, the fundamental distinction between independent creation and unlawful copying cannot be satisfactorily drawn. […] [Music theory] is vital to upholding some of copyright’s most foundational norms. The lay listener test circumvents music theory, thus bypassing critical steps in the infringement inquiry’;¹⁷⁸ and that ‘the recognition of similarity is an acquired skill, not a stable binary yes/no response.’¹⁷⁹ After an experiment testing lay listeners’ reactions to hearing allegedly infringing musical compositions, Lund concludes in relation to a similar method of performance of two pieces that, although the manner of performance has an effect on listener perception of similarity, ‘it is not so determinative as to eliminate the effect of actual structural musical similarities on listener perception’. An instruction of a jury in the experiment showed limited success in attenuating bias in listener perception. For Lund, the lay listener test is poorly suited to assessing the ‘substantial similarity’ of musical compositions, as it directs the lay listener too much towards the performance, rather than towards the compositional elements.¹⁸⁰

However, for composers such a lawyerly/scholarly approach in court is usually rather artificial; they want to create an aural effect on listeners, irrespective of how knowledgeable they or anybody may be about the academic theory of music behind it. One does not need a music degree in order to compose, and ultimately music itself is a material for the creative artist, no matter who has created it. However, composers and musicians are never lay listeners; they invariably know the craft of music, whether academically or intuitively.

What should then be the determining approach? Is it the sound of the music to the lay listener or is it the design and technique (‘Compositionswissenschaft’, as Joseph Haydn would have called it in the eighteenth century),¹⁸¹ and with them the musical work, to the expert musician and musicologist who can appreciate how the piece has been constructed and crafted (rhythm, harmony, etc.)? In most cases, the design and technique determine and influence the sound. The so-called extrinsic similarity test (similarities in the score and the musical techniques, also following musicological expert analysis or ‘dissection’) and then the intrinsic similarity test (similarities in the lay listener’s experience) in US case law¹⁸² are supposed to reveal the actual situation, but it is not so clear whether that two-step test can be carried out with exactitude in practical circumstances; the two steps may well coalesce. Although the lay listener may not know what prompts the sounds in their particular way, a different composition technique would lead to different sound impressions and sound patterns, also for lay people – at least if they have any susceptibility to music at all. That can obviously be problematic, not only because of a proverbial tone-deafness, but also because of an inability to concentrate on acoustical similarities and differences in the hearing experience, particularly if one does not know the underlying composition techniques responsible for these different effects. The assessment result can then depend on the individual judges (or, depending on the jurisdiction, jury) and their sophistication as lay listeners and is insofar a risk for the parties to the action. Particularly where different composition techniques nevertheless lead to a near-identical hearing experience, at least for the unskilled lay listener, some expert opinion by musicologists must be appropriate.¹⁸³ In special cases, where the musical work is somewhat different from the music according to a layperson’s understanding, expert witnesses are inevitable.¹⁸⁴

The decision is ultimately a normative, not a factual one: what matters is more the normative musical work than the music in it, although they often coincide. The music is fixed and represented (typically) by the score, and this normative musical work, as a static representation of the fleeting art of music, enables property rights (copyright) to be attached to it.¹⁸⁵ However, for copyright infringement, not the recording and preservation of the music by the musical work but the music itself matters, perhaps recreated from such a record or score. Since the subject-matter of copyright protection is the music, the sounds, not the score,¹⁸⁶ one would think that the underlying technique, even if different but leading to the same aural impression, is not decisive, and the court could find infringement if the other requirements are met. That may be acceptable where expert listeners would be able to distinguish in their assessment of similarity the sound experience from the technical craft for its creation. But a lay listener cannot be expected to have this sophistication. If an expert listener hears a difference and a lay listener does not, what applies? How ‘good’ or ‘capable’ must a lay listener be to qualify as a standard listener? Such a decision would be a normative one and could theoretically follow a legal definition or test. The situation appears vaguely similar to trade mark law: in EU law, when likelihood of confusion between two trade marks has to be ascertained, reference is made to the ‘relevant public’. The test is that the relevant public must be deemed to be composed of the average consumer, reasonably well informed and reasonably observant and circumspect.¹⁸⁷ Outside the EU, the test is similar.¹⁸⁸ However, these trade mark tests cannot be adopted for copyright because their rationale is a quite different one: there are no consumers to be addressed (and to be shielded from deception), but a work of music is to be protected because of the more than trivial and commonplace efforts of its maker. In this regard, trade mark law centres on the consumer, copyright on the author. Market behaviour is not relevant in copyright law, nor is – to a certain extent – the effect on those being confronted with the work of music: creation is more relevant than perception. Furthermore, copyright, unlike trade mark law, has no priority rule. Therefore a double-creation of identical tunes is possible: if the composer of tune 2, who may be accused of having taken over tune 1, can show that he/she has had no access to tune 1 at all and tune 2 was an independent and original creation which just happens to resemble tune 1, both composers have copyright.¹⁸⁹ That also applies if both composers have used the same composition techniques.¹⁹⁰

A definition for the qualities of a standard listener in musical copyright infringement cases can probably not be given (something like, for example, the ‘person skilled in the art’ for ascertaining inventive step or non-obviousness in patent law).¹⁹¹ However, in less straightforward cases, an entirely unsophisticated lay listener may not be sufficient, and some musicological expertise should be (and usually will be) admitted by the courts, and frequently one of the parties of the trial will request this. Where the aural effect of two works is the same for the usual lay listener, but the musical techniques are different, as ascertained by music specialists, that would rather count against infringement, particularly if the expert listener is able to detect relevant aural differences.

V. Conclusion

The term ‘music’ is not defined in the copyright systems, but follows a general understanding (or dictionary definition) of music. Music is fashioned time, that is, an intrinsically time-bound process, and as such not directly suitable for property protection, which would presuppose a static reference point, an object of property, such as a physical thing, or for example for copyright protection for artistic works, a physical object (a painted canvas, a sculpture) that can serve not only as tangible moveable property but also as a vehicle for signifying the copyright in it. This is achieved by the ‘musical work’, a normative term which is not identical with music, although it often coincides with it in practice. The musical work provides the static features to which the property right of copyright can attach, against the actual nature of music as a fleeting art that exists only in its performance without the need of fixation. The musical work fixes the music through a score, and that score, although not being the music, serves as a representative of the music for the law because the music can be recreated from it. The content of the musical work is often merely the score, and with it indirectly the music which it denotes. However, the musical work can go beyond the music, and encompasses something musicians would not consider as music, for example performing editions of works of music, whereby copyright protection covers the performing edition only, not the music in it. In some jurisdictions, the musical work in copyright also comprises lyrics or action and therefore also exceeds the common understanding of music by a (Western) musician.

Musical copyright infringement cases reveal that the musical work in the form of the fixation as a score itself is not the subject-matter of protection, but only the music, the sounds, that it denotes. However, the hearing tests and the underlying assessment for infringement, either by lay listeners or by experts, of the unprotectable and protectable elements of music (‘ideas’ and ‘expressions’) and their influence on the listening experience, as well as the ascertainment of the (dis)similarities of melodies, are fraught with difficulties. Furthermore, since music is a living art, performance (and improvisation), conceptually separated from composition by copyright law, can be a contribution to the composition of the music and therefore a shaping element in the making of the music and, with it, the musical work. In such situations, performers are also authors and can become joint authors of the musical work, together with the main composer(s).

The concepts of copyright have been developed for more static objects of protection, such as texts and works of visual art. Trying to restrain music, this highly abstract, ephemeral, ungovernable and unpredictable art form, by copyright-property rules is like the futile attempt to hold one’s own cat against its will. But that is part of the beauty of music.

Footnotes