Back to BlogEducation

Who Owns the Song You’re Singing?

By Lance GlasserJune 29, 20269 min read

In the last post, we landed on one rule above all others: before you publish a recording, ask whether you’re actually allowed to. Easy to say. This post is about how to answer it — how the music licensing tangle actually works, and why it’s far less terrifying than it looks once you see the single fact that organizes the whole thing.

First, plainly: I’m an engineer and a founder, not a lawyer, and none of what follows is legal advice. This is a map for asking better questions — not a substitute for an actual attorney, your denomination’s licensing body, or your own performing-rights organization. Licensing law varies by country and shifts from year to year. Use this to learn what to look up, then go look it up — or ask someone qualified to.

With that said, here is the fact that makes the rest make sense.

There are two copyrights, not one

Every piece of recorded music you have ever heard contains not one copyright but at least two, and they are owned, traded, and licensed separately.

There’s the song — the melody and the words, the thing a composer wrote. And there’s the recording — one particular performance of that song, captured by one particular set of people on one particular day. Think of the song as a recipe and the recording as a cake baked from it. The person who wrote the recipe and the person who baked that cake are usually not the same person, and they each own their own thing.

This is why recording yourself doesn’t make you clean, the point we kept hammering last time. Bake your own cake and you own the cake — but you did not invent the recipe. Play a copyrighted song at your own piano and you own your recording of it; you still don’t own the song. Two copyrights. You have to clear both, or make sure both are free, before that file goes anywhere public.

Once you see the two-copyright split, every confusing acronym in this world snaps into one of two columns: it’s either about the song or about the recording. So let’s sort them.

Four rights, and which ones you actually trip over

Copyright isn’t a single permission; it’s a bundle. For our purposes, four strands matter.

The performance right lives with the song. It’s triggered when music is performed in public or streamed — a service, a concert, a livestream. You almost never license this one song-by-song. Instead, the venue or organization holds a blanket license from the performing-rights organizations, or PROs: in the United States, ASCAP, BMI, SESAC, and GMR. Between them they represent essentially the entire catalog of popular and sacred music, and a blanket license is a season pass — perform anything in their repertoire, report what you used, done. Most churches, schools, and concert halls already hold these. You may be covered for this one without knowing it.

The mechanical right also lives with the song. It’s triggered when you reproduce a composition in copies you distribute — pressing a recording, selling a download. In the US this one is unusually friendly: it’s compulsory, meaning the songwriter can’t refuse you, only collect a set fee, administered these days through the Mechanical Licensing Collective. If you’re handing out recordings of copyrighted songs, this is the strand to understand.

The sync right lives with the song too, and it’s the thorniest of the four. It’s triggered when you marry music to moving images — any video. Unlike mechanicals, there’s no compulsory rate; you negotiate directly with whoever controls the song, and they can simply say no. If your plans involve video, this is the corner to take seriously and ask a professional about.

The master-use right is the only one that lives with the recording. It’s triggered when you use someone else’s specific recording. And here’s the good news hiding in it: if you make your own recording, this entire strand disappears. There’s no master to license, because you’re the one who made it.

The shortcut hiding in plain sight

Look at those four again and a pattern jumps out. Three of them attach to the song; one attaches to the recording. Which means there are exactly two ways to make a problem vanish rather than license it.

Record it yourself, and the master-use right evaporates — you own the recording outright. Use a song that’s in the public domain, and the song-side rights evaporate too — no performance, mechanical, or sync clearance needed on a composition nobody owns anymore. Do both — your own recording of a public-domain work — and you are, in the cleanest case, holding something with no copyright left to clear at all.

That’s not a loophole. It’s just what’s left when you stop using other people’s recordings of other people’s songs. It is also, not coincidentally, exactly the kind of backing track we kept pointing at in the last two posts.

The license your community probably already has

If you’re not in the public domain, the next-best thing is a blanket license built for your world, and two of them are worth knowing by name.

For Christian worship, there’s CCLI — Christian Copyright Licensing International. A congregation’s CCLI license covers reproducing song lyrics and sheet music for the congregation, and a CCLI streaming license extends that to livestreamed and recorded services. It does not cover everything in every situation, but for a church singing contemporary worship music, it resolves an enormous amount in one annual payment.

For Jewish congregations, there’s a direct parallel that far fewer people know exists: JLicense, run through the Reform movement’s music arm. A synagogue reports its usage of Jewish liturgical music under its JLicense number, much as a church reports to CCLI. If you’re a cantor or a music director in a Jewish setting and you’ve been quietly worrying about this, that’s the door to knock on.

The pattern is the same in both: your institution holds the license, not you personally, and a great deal of what felt risky turns out to already be covered. Step one is often just asking your administrator what licenses the organization already holds.

Public domain, done right

The public domain is the cleanest source there is — and the easiest to get subtly wrong, which is why it earned its own “clean if public domain” status back in the catalog.

In the United States, the rule is refreshingly crisp: musical compositions published in 1930 or earlier are in the public domain, and the line advances one year every January 1 — works from 1930 joined the moment 2026 began. A song that old, as a song, is yours to use freely: no performance, mechanical, or sync clearance owed on the composition itself.

But — and this is the trap that catches careful people — a public-domain song is not the same as a public-domain arrangement of it. This is the single most common way to get the public domain wrong, so let me state it flatly. When someone takes a free old melody and writes a new choral setting, a fresh harmonization, an accompaniment, or a modern edition with its own editorial markings, that new layer is a brand-new, fully copyrighted work. The 1600s hymn tune is free. The contemporary SATB arrangement of it sitting on the octavo in your folder is not — and using it without clearance is exactly as much an infringement as photocopying any other in-print piece of music. The melody being centuries old buys you nothing if the version in your hands was arranged last decade.

The same logic applies to recordings: an old song, newly recorded, carries a new copyright on that recording. So “it’s in the public domain” is a claim about exactly one thing — the underlying composition — and it tells you nothing about the arrangement or the recording you’re actually holding. Free tune; quite possibly not a free chart, and not a free track. When in doubt, work from an original public-domain source, or from a setting explicitly marked as public domain, rather than a modern published edition. (Sound recordings also run on their own separate timeline from compositions, which is one more reason to check the recording independently.)

What this means for singing together

Here’s where it lands for anyone trying to sing together across distance.

The model we’ve been describing this whole series — everyone performing against a shared backing track — pairs naturally with the cleanest licensing posture available. Use a track you recorded yourself, or one built on public-domain music, and you’ve already removed the master-use right and often the song-side rights along with it. What can remain is the composition’s performance right when you stream or perform publicly — and that’s the strand your organization’s blanket license, the CCLI or the JLicense or the PRO coverage, is most likely to already handle.

That’s not legal advice, and it’s not a promise about your situation. It’s a sane default: prefer self-made and public-domain material, find out what blanket licenses your institution holds, and bring in a professional the moment video, distribution, or anything you’re unsure about enters the picture. Most of the fear in this area comes from not knowing which of two columns a question belongs in. Now you do.

The whole map, in one breath

Three posts, one idea. Every musical tradition reached, independently, for a shared reference to sing against. The references come in many forms, and choosing the right one is a handful of plain questions. And clearing the one you choose comes down to a single fact — the song and the recording are two different things — and a short list of blanket licenses most communities already hold.

Reach for a reference. Choose it well. Clear it cleanly. After that, the only thing left is the hard part we actually built Lyrekos to solve: getting everyone to hold onto that reference at the same instant, across a slow and crowded internet, no matter how far apart they are.

That part, at least, you don’t have to figure out yourself.

A last reminder, because it matters: none of this is legal advice. Laws differ by country and change over time, and your situation may have wrinkles this map doesn’t show. When the stakes are real, ask a qualified attorney or your licensing body. This series is here to help you know what to ask.

Ready to put a clean backing track to work with singers in different places? Lyrekos is in beta now — come try it and sing together, wherever your voices are.

Lance Glasser

Lance Glasser

Lance is CEO and Co-founder of Kinetic Audio Innovations. He was previously a faculty member at MIT, Director of Electronics Technology at DARPA, and CTO at KLA. He also makes sculpture, which has nothing to do with audio but explains the hundreds of pounds of bronze in his house.

Sing in Sync, Anywhere

Clear your reference cleanly, then let Lyrekos handle the hard part — keeping every singer locked to it across any distance.

Join our CommunityTry it