April 2, 2020 Musicology No Comments

Protection for a combination of unprotected elements.

Musically, the “Led Zeppelin Stairway v Spirit Taurus” trial was at best an abstraction. Legally though? — it’s awesome and informs the starting point from which to look at the second half of the Dark Horse ruling’s consideration of “protection of a combination of unprotected elements.”

The Dark Horse ruling sets up by citing a slew of famous intellectual property cases, first Satava v Lowery, and quotes (about a protectable combination of otherwise unprotected elements):

“Those elements must be numerous enough, and their selection and arrangement original enough to warrant protection.”

The “Sativa,” case, by the way, was a guy selling glass jellyfish sculptures in glass tubes and seeking a monopoly on that idea. But you can’t get monopolies on ideas, so Swirsky wound up with copyright only to his original and not public domain expressions that he brought to sculptures of jellyfish in glass tubes.

Next, we have this, from the ninth circuit en banc on Stairway:

“Since… a selection and arrangement copyright protects… the particular way in which the artistic elements form a coherent pattern, synthesis, or design,” it is not enough to assert “a ‘combination of unprotectable elements’ without explaining how these elements are particularly selected and arranged.”

That bold italic emphasis above is mine — “the particular way” is the thing to which you might be able to get copyright. Selection and arrangement isn’t magic, and unprotectable work doesn’t “convert” to protectable work just as body fat doesn’t convert to muscle. If you make a sculpture out of soup cans, you don’t get a monopoly of all future soup can sculpting; just your own particular way of “selecting and arranging” the cans, and only to the extent that your way is original and not public domain. The soup can idea remains generally unprotectable.

Next up, “Swirsky,” a case brought against Mariah Carey by someone who thought their chorus was both sufficiently novel (it wasn’t) and substantially similar to Carey’s “Thank God, I Found You.” (Again, it wasn’t, but it’s worth noting that there was an inverse ratio rule expressly applied in that case, lowering the threshold for “similarity.”)

While the plaintiff’s musicology on Swirsky was probably wrong — observed similarities between Carey’s Thank God I Found You and Xscape’s One Of Those Love Songs were likely borne of either a mistaken analysis (my preferred hypothesis) or a failure of perspective in discounting common phrases likely discoverable in countless prior works given some time to investigate it — the point is that none of that was as wrong as the district court’s dismissal as a matter of law, essentially saying the expert and his methods were dismissable. The ninth circuit court of appeals reversed, essentially asking, “who are you to dismiss the expert’s testimony?”

Here’s why it matters: Xscape’s musicologist first of all was Dr. Robert Walser, of the musicology department Univ. of Califonia at LA. His approach was too high minded. He produced distilled transcriptions, selecting the notes that mattered a lot and omitting a bunch that did not. I see in this an educator at work, taking something complex and simplifying it for the trier of fact, but the court saw it differently. They misunderstood, I imagine, that the musicologist wasn’t cherry-picking to avoid notes that didn’t serve his argument. Instead, he was including and emphasizing pitches and rhythms that dominated the listener’s perception of melody while excluding and deemphasizing throwaway “ornamental” notes that in his view mostly obfuscate. The defense, of course, rationally transcribed every note and pointed to every non-similarity. I might call this the “maximize the denominator” defense. The Ninth Circuit applied and imposed due modesty.

Not in Dark Horse. Truly, is there any reason we can’t call into question summary judgment itself? Contrary to the preponderance of reporting on how normalizing this is, it’s really quite unsettling.

Importantly to Dark Horse, Swirsky spoke thusly to a “combination of elements.”

Furthermore, to disregard chord progression, key, tempo, rhythm, and genre is to ignore the fact that a substantial similarity can be found in a combination of elements, even if those elements are individually unprotected. See Satava v. Lowry, 323 F.3d 805, 811 (9th Cir.2003); Apple Computer, 35 F.3d at 1445. Thus, although chord progressions may not be individually protected, if in combination with rhythm and pitch sequence, they show the chorus of Thank God to be substantially similar to the chorus of One, infringement can be found. See Three Boys, 212 F.3d at 485; Satava, 323 F.3d at 811.

The Ninth was right — pitch sequence, for example, isn’t meaningful in a vacuum but acquires meaning in combination with other components. As I illustrated with familiar examples in part one, you can have unprotectable pitch collections, rhythms, harmonic accompaniments, etc combining to protectable song components.

The Ninth explains very well:

“In fact, concentration solely on pitch sequence may break music down beyond recognition. If a musician were provided with a group of notes identified only by numerical pitch sequences, he or she could play that music a number of different ways, none of them being substantially similar to each other. In order to perform a song exactly, the musician would need information about key, harmony, rhythm, and tempo — the type of information not included in the district court’s comparison.”

The Blurred Lines case (“Williams,” as it’s cited) is broadly thought to have been a travesty. Tread carefully. That it should be a model for a protecting a combination of unprotectable elements is tortuous. The premise, that in Williams a combination of (1) signature phrases, (2) hooks, (3) bass lines, (4) keyboard chords, (5) harmonic structures, and (6) vocal melodies “could receive copyright protection if sufficiently original,” is in some sense predicated on no better than a hypothetical since (for what it’s worth) I’m unaccepting of that those six elements existed in the plaintiff’s track, much less both it and the defendants’. When Sandra Day O’Connor said. ″A factual compilation is eligible for copyright if it features an original selection or arrangement of facts, but the copyright is limited to the particular selection or arrangement,″ she didn’t have to question whether Feist in fact had telephone listings in its publication. It just did. You could scaffold from there. In music copyright cases, courts need musicologists to illuminate for them what’s in the book, what’s not, what’s iffy.

And that’s true of LOTS of the unprotectable elements throughout the precedential (?) case law. These unprotectable elements were the stuff of largely ludicrous claims by plaintiffs that were rightly dismissed. “Hey, their song is about a truck, and ours was about a car. It’s the same thing!” – type stuff; less than substantive toward a principal that’s ultimately going to have a structure like, “things of that nature are unprotectable.”

No, things of that nature are silly. We learned little from this.

Next, there’s an idea brought to bear from Three Boys that a protectable combination should be concerned with elements present throughout the work as a whole and not just a portion of the composition. This is nonsense that both ignores that music is structured in diverse “sections” each with their own discrete makeup, while it simultaneously invites the errant protection of broader musical ideas like groove and instrumentation that copyright is specifically meant to preserve for the public domain.

In considering protection of a combination of elements, her honor, whom I’ve failed to name to this point — she is U.S. District Judge Christina A. Snyder — shares some precedent informing how many elements a protectable combination might need at a minimum, and seems to say Swirsky, Three Boys, Williams, and Stefani (yes, Gwen) taken together imply it’s as few as five. Then in that light, she considers Joyful Noise’s 8-note ostinato — whether its elements are numerous enough and arranged in a sufficiently original manner.

Her honor states “it is undisputed that the 3-3-3-3-2-2 pitch sequence,” “the resolution of that sequence with a 3-2-1-5 sequence,” “the even rhythm, and its development across a sparse texture is not a particularly unique or rare combination.” I don’t know of a 3-2-1-5 sequence in either track, first of all, but it’s pedantic, as the Ninth pointed out, to look at pitch sequences being protectable in a vacuum, and while “unique or rare” seems a high standard, I could’ve sworn the Joyful Noise expert’s testimony was that he couldn’t find this ostinato employed similarly in any prior art, which I took to mean, to his knowledge some combination of these elements is precisely “a unique combination.”

Again let me note, Justice Snyder reaches the right verdict, and the defense should’ve prevailed in the first place. This is my strongly held and almost certainly never-to-waiver opinion. Also, her honor is surely way smarter than I am.

That said…

Next, the judge considers the testimony of Perry’s expert, “unrebutted” testimony she adds, that the pitch sequence “3-3-3-3-2-2” appears in “Merrily We Roll Along” and “Jolly Old St Nicholas.” And this observation was repeated in the press and in viral youtube videos on the matter. Well, guess what. It does NOT appear in either “Merrily We Roll Along” or in “Jolly Old St. Nicholas.”

I know. You’re incredulous. “Impossible; that would be total bullshit. Someone woulda said something!” Here, the pitch sequence from Merrily.


Yeah, you can find the 3-3-3-3-2-2 in there if you squint (I bolded and italicized it for you) but it’s contextually so different as to be a stretch for that reason alone. But much more than that, Merrily is in a major key unlike either Joyful Noise or Dark Horse. So three’s and two’s mean different notes in Merrily. And the same can be said of “Jolly Old St. Nicholas.” Anybody not buying it, comment. I’ll add audio or something and prove it to your satisfaction.

It’s true, next, that the placement of the ostinato doesn’t make it original enough to warrant protection. But it’s not as silly as the cited Allen V Destiny’s Child where no authority was found to consider “that the location of a single common 3-note sequence was sufficient to support predictability.” These cases are musically NOTHING alike. The notes in Allen v Destiny’s Child aren’t even remotely the same! Dark Horse, while still bullshit, is lightyears more similar. Dark Horse is a series of valid observations that add up to weak sauce, while the other is a case of complete bullshit that gets less valuable as you stack it up. Allen v Destiny’s Child is irrelevant.

Then we again get, (I’m paraphrasing) “given the number of scales that exist, the fact that the plaintiff says they’re both taken from the minor scale doesn’t matter.” I think I said this in Part 1, but like Swirsky earlier, that’s not the musicologically interesting point. The point is two pitches taken from a scale might function similarly and possess some degree of interchangeability, and therefore non-identical tones can be shown to be more similar than their scale tone number might suggest to the layperson. That’s probably why he offered it; that’s what it means. It shouldn’t be so simply interpreted and used against him in a different context, particularly in the course of a reversal. That’s not “most favorable light.”

Then lastly we’re told it also doesn’t matter that both ostinatos are similar sounding synthesizer parts because synthesized timbres are “a common element in popular music.” Synthesizers though are capable of producing virtually ANY sound. How can we dismiss similarity in timbre on the basis that they’re both synthesized? If you want to say “two-oscillator sawtooth wave, detuned a few cents, with wide-open filters, a quick attack, short decay to a high sustain, and a little bit of release are a common element in popular music,” go right ahead. But just as we dismiss the protectability of “scale or mode choices” because of the limited number of possibilities, shouldn’t we consider synthesizer programming choices which have an infinite number of possibilities?

I’m glad Katy Perry is off the hook. It was a bad verdict. Now we have the right one, but we had to ignore a lot of expert testimony and accept some other expert testimony that the jury was unimpressed with, and piece together a bunch of very imperfect case law as authority to get it.

I hated this verdict. But I never expected it to get reversed. And I won’t be surprised to see it flipped back.

Tell me I’m wrong.

Written by Brian McBrearty