Ed Sheeran’s ‘Shape of You’ was a giant hit of the last decade, a runaway success story – well almost. There is a fly in the ointment. Sami Chokri (a.k.a. Sami Switch) is claiming Sheeran and his fellow songwriters infringed his copyright, copying part of the chorus of ‘Oh Why’. You can hear the two disputed musical phrases here.
In order to support his claim Chokri wishes to plead evidence of Sheeran copying other songs. This is to prove Sheeran and his collaborators were in the habit of appropriation. The logic is if Sheeran can be shown to have infringed on other occasions, then it would be more likely he would infringe again. A judgment on 9 December, Sheeran & Ors v Chokri & Ors  EWHC 3584 (Ch), allowed Chokri to bring in these arguments, though it has not yet been decided whether they are made out.
Specifically, Chokri says Sheeran’s ‘Photograph’ copies ‘Amazing’ by Matt Cardle, ‘Strip That Down’ copies ‘It Wasn’t Me’ by Shaggy and ‘Shape of You’ copies ‘No Scrubs’ by TLC.
The logic and assumptions involved in establishing and relying on this type of evidence deserve scrutiny, raising questions such as the following:
Sheeran does not concede he copied the three other songs. Is it efficient for the court to resolve one dispute by deciding three other disputes?
Chokri notes Sheeran settled or cleared rights with the other songwriters, agreeing to a royalty split, which he says gives rise to an inference of copying. Is this inference correct? Parties may choose to settle for a variety of reasons. In the case of music, the collecting society PRS for Music will not pay out royalties in the event of a dispute, not paying interest on the monies held in suspense. Drawing negative inferences from settlements could undermine the value of parties settling – is this what the courts should be aiming for?
Even if it were demonstrated Sheeran had copied elements of three other songs, does it follow he is more likely to have copied Chokri’s song? Sheeran’s success means he is much more likely to receive claims than others. Perhaps three instances of potential copying is less than the average. It would be necessary to ascertain how often unsuccessful songwriters arguably imitate other songs.
To what extent has the ‘Blurred Lines’ case led to a rise in music plagiarism disputes and an incentive to settle which is reflected in the number of Sheeran disputes?
What is the rate of copyright ‘reoffending’?
It is insufficient to say other examples of infringement increase the probability of infringing again unless the probability is quantified. Would the probability of infringement increase by 0.00000001%, 1%, 10%, 50%…? How should the probability be calculated?
As Hugh Laddie wrote: ‘It is tempting to say “once a thief, always a thief”, but I suspect that this type of material is logically of minor probative value in many cases.’
This article can also be found at the The Kluwer Copyright Blog.