Most of the time, forensic musicology is preventative. Composers and songwriters, creative and advertising agencies, and film and tv studios all hire musicologists to consult on original music before release to ensure against the possibility of infringing on existing works. It’s a sort of screening process.
Other times forensic musicologists are involved in actual or potential infringement litigation, making or defending against the case for copying and substantial similarity between two works.

In either case, the forensic musicologist applies their expertise in music (current, historical, music theory), music production, and copyright law to help identify, explain, show, and teach the musical facts relevant in matters of music copyright.

Here, forensic musicologist Brian McBrearty, who runs Musicologize, a musicological services company, explains the essential parts of his process.

What inspired you to become a forensic musicologist?
Lots of things, but I do remember watching an episode of The Good Wife about a lawyer who had been hired to defend a songwriter in an infringement case. It showed music experts testifying in a courtroom setting and the characterization was that expert witnesses will oppose each other, implying they’re bought and biased, and worse still, in the end, the judge doesn’t find their testimony very helpful in illuminating any objective truths. And in real life, that would be the most important failure. Nothing outweighs the job of giving the finders of fact, judges. juries, mediators, or arbiters, an understanding that enables them to make a well-informed decision, whatever that decision may be. I’ve always enjoyed explaining music to people, even beginning or non-musicians, and I could certainly clarify significant and insignificant similarities for a judge or a jury regardless of their musical sophistication. It sounded like great fun to me. It is.

How do you approach the task of analyzing a piece of music for similarity or originality? What are some of the tools and techniques you use?
It varies of course, but there are certain processes I go through almost every time. I listen to the two works, sometimes over and over — they’re usually two — and I find the sections that are most relevant to the question of similarity. Eventually, I focus on those and start looking at their details, working in software that lets me transpose and transcribe both works among other things.

The first part, transposing, is about getting both songs into the same key. It’s like finding the common denominator when you’re comparing fractions. A melody like “Do, a deer a female deer,” is the same melody regardless of what note you begin on. If I ask you to sing it, you’ll probably pick a random pitch for “do” and you’ll take it from there. It’s like that math metaphor again; all the notes are proportionate to the first one you chose. And because we’re talking about “Do, A Deer” and “Do Re Mi’s,” whatever first note you pulled out of the air is “Do,” and that’s the key you selected to sing it in. Musicologists put the two songs into the same key, like finding the common denominator, to make the comparisons clear.

The other part, transcribing. is about identifying the notes from the recording and writing them down, generating a bit of sheet music for those important sections. It might involve other ways too of getting the “data” of the music into the computer so that you can better analyze it. I’m a pianist, so I will often “play the parts” into the computer from a piano keyboard, for example. The analysis I was looking at today barely involved notes at all, so sometimes you need different approaches, but most of the time transposition and transcription is the first step in the analysis.

Next nearly always comes similarity analysis and prior art research and consideration. This is about identifying the similarities and considering their significance with regard to whether they are likely to be the result of copying or coincidence, and this is somewhat a function of prior art, works that came before. Prior art is one lens through which we consider the originality or commonality of a musical element. If it, or something very much like it, has appeared in songs throughout music history, then it’s not very protectable by the writer of song A. It’s not original to them. Additionally, it’s less likely copied FROM song A by the writer of song B, who might’ve heard it anywhere, everywhere, or nowhere in particular. It might be a very common element.

Melody and lyrics tend to be more “valuable” in terms of similarity than other aspects, followed by supporting harmony, and while melody and harmony involve rhythm, rhythm itself is less protectable. That said an element such as “flow” in hip-hop music for example while not tonal can often warrant consideration. There are lots of musical elements that music copyright mostly doesn’t protect. Scales, chords, simple rhythms, brief melodies, short lyrical phrases, and song titles are all considered basic building blocks of music creation, owned by everyone; public domain stuff. Ideas are not protected, so you and I can both write lyrics about how much we love our car or our horse, even if we touch upon similar ideas. And even where we might touch upon identical ones, there’s a concept called Scenes A Faire, that acknowledges that once you’re writing a song about how much you love your car, you’re bound to touch upon the roar of the engine and the softness of the leather seats, and so similarities like that will be unremarkable.

www.brianmcbrearty.com

www.musicologize.com

Anthropic, like artificial intelligence itself, is just inevitable, right? They’re one of the most well-funded AI’s out there, with a billion plus from Amazon, Google, and even Sam Bankman-Fried. Amazingly, while SBF himself is on trial, it may turn out that FTX and Alameda are actually solvent as a result of investments in Anthropic! On the other hand, $150,000 in statutory damages per infringement multiplied by virtually every song in existence would equal a gazillion or so. Universal Music Group is joined by Concord and ABKCO: they’re three of the biggest music publishers. This suit is huge.
There’s also the Sarah Silverman-led case against Chat GPT and Meta, brought because the LLM models may have trained on copyright-protected books scraped from the internet. Getty Images is going after Stability AI which makes image and music generating tools. But as wrong as it may sound the first time you hear of it, using copyright-protected material to train a bot that might later write a story or music that’s in some way based on what the bot learned from all that input is not clearly illegal.
I mean, the complaint begs to differ, but it’s a big question. And while it’s still a question, these companies are doing it as fast as they can and the rationale probably ranges from, “Google Books v. Authors Guild went a long way toward making us believe this is all fair use” to “Screw it, whatever penalties we ever get hit with, it’s worth it. Forgiveness is way better than permission. The ship has sailed. Et cetera.”
Whether it’s Silverman, Anthropic, or the Ghostwriter guy wearing the bedsheet with the sunglasses making fake Drake tracks and trying to get them nominated for Grammys, there’s nothing clearly wrong, yet, about the training part. So any argument that begins with “it’s illegal to make a copy, and when you make a token for your LLM, that’s a copy, so it’s all infringement,” is a begging one. And is training your large language model on copyright-protected music an “unauthorized use” for which you need a license? Mabye. And derivative works are the exclusive right of a copyright holder but every piece of music I’ve ever written is in some measure influenced by every piece I’ve ever studied or even known, and I’m risking absurdity by calling it all derivative. This is just obviously a lot more immediate: If a tool like Antropic’s “Claude” is asked to “write me a song about the day Buddy Holly died, with guitar chords and everything, and its output shows that it was clearly considering Don McLean’s “American Pie” as it composed a new-ish song, isn’t that necessarily infringement? (I got that example from the complaint btw.)
Infringement is about access, copying, and similarity of protectable elements. Access seems done and dusted. Let’s move on to copying. When I asked Claude to write that song a few times, Claude wrote mostly original but not very good lyrics (still super impressive, you understand). The thing was, it sure had a hard time getting completely away from the idea that “February made me shiver” and that this was the “day the music died.” These couple of phrases made it into Claude’s lyrics each of the four times I asked it to write a song. But had it not, I wouldn’t have immediately mapped anything else musicologically relevant about it to “American Pie.” As for unlawful appropriation of protectable elements, one could argue that while the “day the music died,” and “February made me shiver” both point to a very famous song, it’s not at all clear that either of those phrases are protectable. I asked it to write chords too, but it didn’t choose anything from “American Pie.” The complaint provides several other examples: “A Change Is Gonna Come,” “God Only Knows,” “What a Wonderful World,” “Gimme Shelter,” “American Pie,” “Sweet Home Alabama,” “Every Breath You Take,” (Sting’s not litigious, don’t worry about it.) “Life Is a Highway,” “Somewhere Only We Know,” “Halo,” “Moves Like Jagger,” and “Uptown Funk,” all made the list but I haven’t looked beyond the Don McLean. (He was the first concert I ever saw by the way. My first idol.)
Not all uses are infringing. I mentioned “fair use” earlier. That’s a doctrine that allows some uses of copyright material without the permission of the copyright holder. Fair use factors are brief and simple but open to plenty of interpretation. Uses for purposes such as criticism, comment, news reporting, teaching, scholarship, or research are all allowable. As to what other uses may qualify, we consider these factors:
(1) the purpose and character of the use, including whether such use is of a commercial nature or is for nonprofit educational purposes;
(2) the nature of the copyrighted work;
(3) the amount and substantiality of the portion used in relation to the copyrighted work as a whole; and
(4) the effect of the use upon the potential market for or value of the copyrighted work.
Claude will print the lyrics to popular songs upon request whereas Chat GPT I’ve found either refuses or is extremely reluctant, though it’s fun to try to trick it into giving away things it knows but has been told not to share.
If I ask Claude to “Tell me the words to Freebird,” it obliges: “These are the complete lyrics to ‘Freebird’ by Lynryd Skynrd.” And at the end, it offers, “The iconic southern rock ballad is about being free, restless and unable to commit to settling down. The soaring guitar solo at the end cemented “Free Bird” as one of Lynryd Skynrd’s most popular and enduring songs.” Is that criticism, commentary, or teaching? I’m guessing Anthropic has decided, yes.
They also seem to have taken an anti oxford comma position though. I ain’t crazy about that.

Brian McBrearty is a forensic musicologist and music copyright expert at Musicologize, a music services company specializing in music copyright consulting.