When does one pop song infringe the copyright of another pop song? US case law on the topic – which more or less mirrors Canadian law – shows the unpredictability of how music copyright infringement determinations are made. In two recent high profile US cases, one jury found for the plaintiff where the two songs had almost nothing in common (Gotta Give it Up and Blurred Lines) whereas in the other case, the jury found for the defendant even though there was significant borrowing from the plaintiff’s composition (Taurus and Stairway to Heaven).
There are many points of complexity in music infringement cases which no doubt account for such wildly different outcomes. In this column, I unpack some of the more difficult issues within the US and Canadian context. These difficulties include: (1) how music is compared (2) the legal test for copyright infringement and (3) the relevance of lay versus expert opinions about the comparison. I conclude with some ideas for making the infringement analysis better in music copyright cases.
How Music Is Compared
We begin with a technical, though critical, point. The Blurred Lines and Stairway to Heaven cases were both about (as these cases overwhelmingly are) infringement of the musical composition – i.e. the melody, harmony and chords that make up the song. This is distinct copyright from sound recording or performance elements of a song – i.e. the arrangements, choice of instrumentation, the production feel – which have their own copyright but have no bearing on the musical composition per se.
These accoutrements can have a profound impact on lay perception about the similarity. In one study, songs which compositionally had little in common were more likely to be considered similar if they shared the same genre or “sound”. Conversely, compositionally similar songs were less likely to be viewed as similar if they were played in different styles. This finding was reflected in Blurred Lines in which the jury sided with the plaintiff even though the two songs, at the end of an extensive analysis, shared only the same “production feel.” At the summary judgment stage of the Stairway to Heaven case, the court rightly discounted expert reports which drew on similarities in production or performance.
So how do you make the comparison? One might think that sheet music would be a media neutral means of comparison. However, such a comparison is not as accessible to lay jurors nor, in pop music circles, is that necessary the way in which a song comes into being – i.e. sheet music is not usually the first medium of expression. For these reasons, courts do not usually allow triers of fact to listen to sound recordings of the songs. Rather, the comparison is made using a stripped down rendition of the composition on a piano or a guitar to offer an accessible means of comparison while eliminating production and performance elements.
The Legal Test for Copyright Infringement
The test for infringement is whether there has been a substantial qualitative taking of the plaintiff’s song. This assessment is unavoidably impressionistic. The bottom line is that the threshold for this is not particularly high. Infringement may occur when a part of the melody or chorus, or of a riff or a hook, is borrowed.
Adding to the confusion is whether, when examining what is taken, one considers the whole of the borrowed elements or only its copyright protected parts. In Stairway to Heaven, for example, the alleged infringement was a “descending chromatic bass line” that lasted approximately two minutes at the beginning of each song. The plaintiff alleged this was entirely original expression (copyright protected) whereas the defendants maintained that is a “centuries-old” musical convention (not copyright protected but in the public domain).
In Canada, our Supreme Court in Cinar decided that, when making comparisons, the whole of the work (protected and unprotected elements) is to be considered and not just that which can be filtered out as protectable copyrighted expression. This assessment favors plaintiffs since unprotected elements do not need to be discounted. In other words, there can be infringement when what is borrowed is in whole or large part from the public domain.
In Cinar, the Supreme Court also affirmed that the proper inquiry for copyright infringement is what is taken from the plaintiff’s work, not how it is used by the defendant. This is critical in a case like Stairway to Heaven, in which the borrowed elements make up a quantitatively and qualitatively large part of the plaintiff’s work but a relatively small part of the defendant’s work. If the defendant’s use of the borrowed elements was deemed relevant to infringement analysis, there would a much clearer, and convincing, basis for the outcome in that case.
Lay Versus Expert Opinion
The last question is whether expert evidence is relevant and helpful to infringement proceedings. On the one hand, 90 per cent of music infringement cases involve pop songs. If we accept that copyright serves primarily as a means of compensating artists, then the opinion of lay people (the consumers of pop music) should be mostly what matters. If they do not view the songs as similar, and most certainly if they do not view them as substitutable in the marketplace, then perhaps there should not be infringement.
But in fact there can – depending on the complexity of the music – be latent or structural similarities between songs that are not perceptible to the lay person. In Cinar, the Supreme Court showed a strong willingness to admitting expert evidence when there may be structural similarities owing to special nature of the intended audience (children), comparisons between two media (written script v. TV production) or different stages of development (basic script v. full scale TV production). Arguably for many pop songs, such reasons for seeking expert opinions do not exist.
The Way Forward
Music copyright infringement analysis will always be difficult. Certain safeguards and rules can be put in place to make enable a more accurate and rigorous analysis.
First, if it is a composition copyright infringement case, then stripped down versions of the songs – which remove performance and production elements – must be basis for comparison.
Second, our Supreme Court should redefine what it means to holistically compare two works for substantial similarity analysis. If experts can convincingly demonstrate that a work like Taurus is merely a new performance of an old compositional convention, then this should mean that there is no basis for copyright infringement. The court might also want to reconsider whether use of the borrowing by the defendant, and not just the amount taken of the plaintiff’s work, should also be a factor in infringement analysis.
Third, expert evidence will often be helpful in making determinations about what is original and subject to copyright and what is a musical convention in the public domain. Experts can also explicate deeper similarities between works. In these ways, such evidence will often illuminate issues about what and how much is taken. The trier of fact will remain in the position of accepting or rejecting (or choosing between) expert evidence.