April 30, 2022 Musicology No Comments

Two songs that sound a lot alike. Is it copying or coincidence?

Anyone remember a Seinfeld episode in which Elaine and a visiting character, Zara, argued about whether some coincidences are bigger than others, or all coincidences just 100% coincidences? If you agree with Elaine, that there are degrees of coincidences, which is about perception of probability, maybe you think this is a pretty big one.

Probability, or the perception of probability, is very much the point here.

Palmer Rakes and Frank Fioravanti are suing Dan + Shay and the Biebs because they feel 10,000 Hours, which was a big hit in 2019, infringes on their copyright to “The First Time Baby Is A Holiday” which they wrote forty or so years ago.

On the day this lawsuit became news I was busy and could only offer a hot take. I said this looks more like coincidence than copying. I haven’t changed my mind. And I know some will be thinking, ”But it’s sooo the same!” “And there’s all that math in the complaint about percentages of notes that are identical!” “And they said something about “similarities so striking that 10,000 Hours simply cannot have been independently created.”

And melodically, they are a lot the same, it’s true. I certainly hear what you hear.

Unless you haven’t heard it. Here, two very similar melodies.

First up, 10,000 Hours By Dan + Shay and Justin Bieber.

And here’s First Time Baby, by Palmer Rakes and Frank Fioravanti.

Ok, they definitely sound a lot alike. They ARE alike. I think I remember a remark in the complaint about how it’s hard to overstate it. It kinda would be. But that by itself isn’t quite infringement. Infringement is going to require access and copying of original expression. And that last part has both “copying” and “originality” inside, so what arguments would substantiate both? Let’s consider.

In terms of tempo and key center, neither of which matter a whole lot, but I still get them said and out of the way first, because they’re the first thing I figure out when I put the tracks up, these two songs are in different keys and at different tempos. But they’re both laid back “moderato” tempos.

And in terms of form, similar there too, but similarly mundane. These are a couple of decidedly straightforward works. 10,000 Hours pretty much goes Verse, Chorus, Versus, Chorus, Bridge, Chorus; nothing novel. And First Time Baby is pretty much the same thing but lacks a bridge. Also nothing novel, and because First Time was written in the day of such things, after that second chorus there’s a solo instrumental section (over the verse material) and then the big final chorus and fade out. It was kinda funny to be expecting it, thinking — “is there gonna be a little verse solo followed by the last chorus?” — and then get it. Heartwarming even. I rather liked this track. Both songs have intros and outros too, but whatever. One way more involved than the other. None of this is interesting.

Here’s what’s interesting: First Time Babys’ verses sound a ton like 10,000 Hours’ choruses.

That’s the focus, the complaint shows these two sections side by side and finds that, depending upon how they slice it, 72% or 83% of the melody notes in these sections are identical in pitch and metric placement. Metric placement is the beat the notes land on. Both % numbers are defensible. And both sound pretty damning. That’s a lot of one of the main things we look for — identical pitches occurring at identical times.

Identical times though are a bit in the eye of the beholder. Here, I’d say they didn’t have to inflate it. It was on a silver platter. Moreover, I’d say the notes that aren’t identical are often “pickups” and “passing tones” and have a little less impact than the ones that are the same, so if anything, the percentages tell a slightly conservative story. If I were to do one of my high-value “money notes” types of illustrations here, where I filter out the notes that have less impact, the percentage might be closer to 100%!

Lyrically, the next place you look, not similar, but the plaintiffs will point out that both songs end these sections with the phrase “the rest of my life.” Not going to dwell on this. “It’s a time I’ll remember for the rest of my life.” vs “If it’s ten thousand hours or the rest of my life.” These are obvious, common phrases.

Let’s next consider the chord progressions — the harmony beneath these melodies. I’m sitting at an unfriendly keyboard and I want to type as few special characters as possible so I’m going to transpose both of these into the key with no sharps or flats, C major, and illustrate them that way.

First time: | C     | F  C | C     | Bb  F |
10k:        | C  Am | F  C | C  Am | F   C |

Not the same, nor especially similar, and both very basic and common. If you’re looking for trouble, you make something of the fact that both return to C via F in that second measure. That’s a plagal cadence, which is by no means rare, but if you’re looking to stack similarities, it is one. On the other hand if you’re looking for differences, that Bb to F is also by no means rare, but it’s the only remotely interesting choice made in either track, and it isn’t in 10k. In terms of harmonic rhythm, which is where the chords occur, part of what we call “groove,” in both tracks, the chords aren’t equidistant. They are about a half a measure each, but not exactly. The second chords in every measure of 10k occur just a little ahead of beat three. And this is also true of the second and fourth measures of First Time, which are the only places where it contains two chords in the measure.

And this follows the fact that the melodies of both songs similarly emphasize those metric placements.

If you’re not quite understanding “ahead of the beat,” think of the Biebs track and pay attention to the first syllable in “hours.” That little bit of grooviness there is the “and of two.” Both songs have that groove going. So although First Time just sits on a C major chord in the first and third measures, these grooves are similar.

And let me veer off for a second… the folks suing Ed Sheeran would probably say this is significant. This same sort of “and of two groove” is found in “Thinking Out Loud,” and in Marvin Gaye’s “Let’s Get It On” and I’m quite sure they’re going to try to make an incredibly big thing of it in that case.

So where are we? These are two very similar melodies, accompanied by not very similar chord progressions, but with similar harmonic rhythm. It does indeed look like a good amount of intrinsic similarity, especially to a layperson.

And the complaint says something interesting, given the climate of copyright infringement that we’re living in.

“… Musicological comparisons have in many cases involved comparisons of as few as
6 to 7 consecutive notes in the subject musical compositions. Here, in contrast,
several 47-note sections of 10,000 Hours are virtually identical to parallel sections
of First Time.”

I interpret that to mean they would distinguish and distance themselves from Dark Horse and Oh Why and other high profiles cases for which the courts, also my interpretation, are perhaps losing their collective patience. And the court of public opinion may be as well.

Okay, so why is this not a slam dunk?

Remember first that we aren’t looking at similar lyrics. (Again, that “rest of my life” thing is, I think, not worth much consideration.) It’s pretty much just that the melodies have a ton of identical pitches landing on identical beats. If I transcribe both parts onto vellum and we hold them up to the light, they’re gonna line up very very well.

But, what if I then wrote up parts of other songs you might know, also on vellum, and we hold them up to the light against First Time and 10k and they have lots of similar notes too, strings of them. The idea is simply that if the melodies are common and appear in “prior art” that came before “First Time Baby Is A Holiday,” then why especially do we think 10,000 Hours was inspired by First Time? And even if it was, we might say it doesn’t matter, since the material isn’t First Time’s to claim.

How on the nose does a prior art example need to be? If you happen to have read my blurb from when the news first broke, I made a quip at the end about a train a coming. It was the first thing that popped in my head. Then a Sting tune popped into my head.

There are only 12 notes. Sometimes there are even fewer.

Something we heard in the recent Ed Sheeran case was that the Oh I and Oh Why’s in that case were sung to pentatonic scales. It was one of the reasons Sheeran won. Pentatonic scales are super common, and scales are common property, so it’s an argument against originality. Well, these two melodies both contain nothing except pentatonic scale material. Penta. That means five. If you’ve heard it said, “there are only 12 notes!” And then someone like me scoffs, as I usually do, “Twelve is plenty. Write something good.” Well, five is not a lot. It’s Do Re Mi Sol La. That’s it.

Nevermind reading music; just follow my Do Re Mi’s. The first six notes of 10,000 Hours and This Time Baby, are La Sol La Mi Re Do La.

This is that Sting song. Same list. La So La Mi Re Do La. It’s in a cool 9/8 time signature too, try to enjoy that. Just listen to the first few lines.

The other thing that popped into my head, “People Get Ready” was written by Curtis Mayfield and recorded by the Impressions. Six notes, so it’s hexotonic. It sounds a bunch like 10k Hours and First Time Baby.

Yes, I know this is a live version of a remake. I used to watch plenty of MTV. I cherry-picked this. Stewart elects to begin the melody on La.

In the legal process (I am not a lawyer) there’s an idea around “filtering out” prior art to distill what’s original and therefore protectable by copyright. So how many notes would “People Get Ready” filter out? More than a few, but actually fewer than you might think. It would take a good bit of that “I vi IV V” 10,000 Hours chord progression with it, groove and all. But the similarity you might really be picking up on is melodic arc. In each of these songs the shape or arc that the melodies follow is one measure downward, then one upward, then one downward, and then a cadential statement that caps off and resolves the phrase.

Prior art “filtering” arguments will almost certainly be forthcoming. Filtering can be coarse in its application, and I’ve been known bristle at it. But rather like the famously inefficient “inverse ratio rule,” filtering out prior art makes sense when applied appropriately and when the approach is both honest and skillful. Reluctantly, I might describe it as a bit of an art. In this case, in my opinion at least, it helps to illustrate the truth, which is that these melodies are made from very common expressions which makes it harder to infer copying.

But I think the better argument is a bit like one I made in the Dua Lipa case. This isn’t nearly as big a coincidence as it at first might appear.

The plaintiffs made that point about this being a longer melody than those in some other infringement cases for which musicologists have argued copying took place. And yes, these two melodies that we’ve focused on are eight measures and about fifty notes long. But across those whole eight measures of “First Time Baby” measures one, three, five, and seven are essentially identical. So are two and six. So are four and eight. The phrases in 10,000 hours are four-measure phrases, repeated twice. So when you’re trying to calculate the probability that they were independently created, this presents very different math compared to 34 notes out of 47 which I’ll now call the raw data. Is this so different from the short, simple, and commonplace sorts of melodies that were contested in the trials over Dark Horse and Oh Why? It’s more, yes, but the raw data doesn’t tell the whole story.

Probability can be confounding for many people. I’m faced with this daily.

There’s a familiar illustration of eternity, or of the expanding infinite universe, that says eventually, somewhere, another Shakespeare will write the same works, by coincidence. What’s the probability that after forty years, someone else would have written this same melody? This is the essential question and opinions are going to differ. But given the simplicity of the material along with my perception that it’s a shorter string of similarities than the complaint argues, I would say the probability is high.

And an aside. I’ve heard all kinds of things about current tin pan alley strategies. One is certainly that songwriters, or songwriting teams, or songwriting factories are intentionally, even retroactively basing things on public domain tracks as a mitigation strategy against claims of copying. And I can easily imagine that someone must’ve said to the 10k Hours people, “wow this sounds a ton like “People Get Ready.” Because it does! But I dont think more than 20% of the notes in 10k’s chorus are actually at the same pitches as “People Get Ready” while “First Time Baby” is around double that. So am I saying the 10,000 Hours writers moved notes around to try to ensure they didn’t infringe on People Get Ready?

No. But would I be surprised?

Written by Brian McBrearty