Lately there’s been a blizzard of litigation about violations of song copyright, all based on the melody of pop tunes. I thought the Chiffons had a point when they claimed “My Sweet Lord” infringed “He’s So Fine”— listen to them back to back and I bet you’ll agree— but now there are so many online music platforms, and we’ve seen amateur musicians who’ve uploaded their work to some site or other claiming that major songwriters have infringed their melodies, whether those songwriters have ever heard of them, heard the music, or even heard of the uploading site.
And sometimes it just gets insane, as when Lana Del Rey’s “Get Free” was accused of violating Radiohead’s “Creep,” which had in turn been accused of violating a Hollies song called “The Air That I Breathe.”
The problem is that there are only 12 equal tuned notes per octave, and there are a finite number of ways they can be arranged. Which limit has now been reached by musician and technology attorney Damien Riehl and his partner Noah Rubin, who developed a brute-force algorithm that has now composed every possible melody from now to the end of time, which you can now carry on a memory stick in your pocket.
(We are not talking about microtones, because they aren’t used much in pop music, which is the only music that generates enough money to be worth suing over.)
The computer-generated melodies were promptly copyrighted by Riehl and Rubin, who then made them available to the entire world via a creative commons license.
Though if you are, say, the Chiffons, your copyright of “He’s So Fine” predates R&R’s copyright of the same tune, so you’re still okay and we can look forward to being entertained by infringement suits from now till doomsday. (Though Rielh is apparently prepared to argue that notes are nothing but math which can be displayed as such in a MIDI, and that since all math existed from the beginning of time, copyrighting a melody is impossible.)
Assuming that the courts agree with R&R’s copyrights, that means that you can no longer copyright your brand-new melody. (Though you can copyright the lyrics, the arrangement, etc.) And if the courts agree that music is nothing but math, all copyrights of melodies are hereby void, and a whole lot of expensive attorneys lose their work.
Of course there will have to be an enormous amount of expensive litigation before this is all settled, which is interesting for a project intended to relieve the world of expensive litigation.
In case any of you are wondering how this applies to prose, it pretty much doesn’t. As Rielh states in his talk, music may have 12 (ish) notes, but the English language has hundreds of thousands of words, so brute-forcing your way through every possible paragraph is probably impossible (for at least the next five to ten years).
I think it’s a splendid idea, but what’s copyrightable is creative work and in this case that’s the algoritm, not its output. The output is just finding what’s already there (and possibly selecting what’s “reasonably” pleasant to listen to?).
As for text, well, there’s already thisarticledoesnotexist.com (with its companion http://www.thispersondoesnotexist.com ) which could probably be extended to fiction so that every reader gets to read a short story nobody has ever read nor written.
BTW my copy of Quillifer the Knight arrived Thursday, so this is a good weekend.
1. Algorithms are not copyrightable (a particular web page or paper describing an algorithm would be, but not the algorithm itself). They may or may not be patentable in certain circumstances.
2. As Riehl points out in the excellent FAQ section of his website, if the algorithm is just “finding what’s already there”, then the melodies are “mere facts”, and thus not copyrightable, any more than the decimal digits of pi are, or a list of prime numbers (or a list of telephone numbers, which I think is one of the classic examples from the court cases that have argued that you can’t copyright “mere collections of facts”).
I am now reminded of Spider Robinson’s 1983 short story “Melancholy Elephants”, which touched on some of these issues. (Specifically, the idea that there are a finite number of melodies, and lengthy copyrights — in the case of the story, a proposal for perpetual copyright — are inimicable to musicians.)
One of those is presumably The Ultimate Melody, as described in Tales From the White Hart.
I think it’s a splendid idea, but what’s copyrightable is creative work and in this case that’s the algoritm, not its output. The output is just finding what’s already there (and possibly selecting what’s “reasonably” pleasant to listen to?).
As for text, well, there’s already thisarticledoesnotexist.com (with its companion http://www.thispersondoesnotexist.com ) which could probably be extended to fiction so that every reader gets to read a short story nobody has ever read nor written.
BTW my copy of Quillifer the Knight arrived Thursday, so this is a good weekend.
Yay for litigation! This world needs more lawsuits!!!
@ Urban:
A couple of points:
1. Algorithms are not copyrightable (a particular web page or paper describing an algorithm would be, but not the algorithm itself). They may or may not be patentable in certain circumstances.
2. As Riehl points out in the excellent FAQ section of his website, if the algorithm is just “finding what’s already there”, then the melodies are “mere facts”, and thus not copyrightable, any more than the decimal digits of pi are, or a list of prime numbers (or a list of telephone numbers, which I think is one of the classic examples from the court cases that have argued that you can’t copyright “mere collections of facts”).
I am now reminded of Spider Robinson’s 1983 short story “Melancholy Elephants”, which touched on some of these issues. (Specifically, the idea that there are a finite number of melodies, and lengthy copyrights — in the case of the story, a proposal for perpetual copyright — are inimicable to musicians.)
Comments on this entry are closed.