Tuesday, April 29, 2014
"Shadows" and "Innocence," copyright and performance
Earlier, I wrote about the Ninth Circuit’s recent Garcia decision, which is turning out to be the copyright Ishtar* of 2014. One take on what is so rank about the opinion is that it flouts a basic copyright principle that performances (separately from the works they are based on) are not copyrightable.
But earlier this month, just after Garcia was decided, the District of Nevada issued a far less remarked-on opinion entering summary judgment in favor of Teller (the silent, shorter member of the famed Penn & Teller duo), who argued that a YouTube video by Dutch magician Gerard Dogge infringed Teller’s copyright in his illusion “Shadows.”
At first glance, it’s hard to tell these cases apart. If Garcia was wrongly decided because (in large part) it erroneously held that performances are copyrightable, then shouldn’t Teller have lost as well? The answer is no, but it requires a closer look at the circumstances of these deceptively different cases. And the value of taking that closer look is to parse out more carefully what does, and does not, work about the “performance is not copyrightable” aphorism. More below the fold.
It’s worthwhile to give a quick sense of the infringement in Teller. Teller’s “Shadows” is a very elegant and affecting illusion that begins with a rose on a stand with a light in front of it causing a shadow of the rose to be projected onto a screen behind it. Teller then appears, and cuts away the petals of the rose’s shadow, and when he does so, each corresponding petal on the actual rose falls as well. In Dogge’s video, he performs a virtually identical illusion with nearly identical set dressing and materials (indeed, the name of his illusion is “The Rose and Her Shadow”), though there are some slight variations (Teller uses a rose in a vase, while Dogge’s is in a bottle; Teller’s and Dogge’s performances end with different performative flourishes).
That said, the way to reconcile Garcia’s wrongness with Teller’s (relative) okayness lies in the nature of the latter’s asserted copyright interest. Teller registered his work, “Shadows,” as a dramatic work (actually, "dramatic pantomime" in the registration certificate) with the Copyright Office in 1983 (though he had been performing the illusion since 1976). The registration comprises a detailed description of “Shadows” to the minutest detail. By contrast, Garcia’s purported copyright in “Innocence of Muslims” derived solely from whatever originality her performance added to the underlying words written by the screenwriter.
Moreover, Teller’s registration of “Shadows” also clarifies and simplifies his authorship status with respect to the work. He is solely listed as the dramatic work’s author, because he is—the entirety of “Shadows” is the product of his creative mind alone. Again by contrast, Garcia’s authorship status with respect to “Innocence” is a wreck. She is at best one of many joint authors of the work, though her relative contribution to the final product is vanishingly slim.
Still, these two cases raise a puzzle: Is performance copyrightable or isn’t it? I don’t think we need to get too Clintonian about this (i.e., no need to default to “it depends on what ‘performance’ means” hair-splitting). Teller didn’t hold that performances are generally copyrightable. It held that Dogge’s YouTube video amounted to an unauthorized public performance of Teller’s copyrighted dramatic work.
If you want to get really technical about it, Teller’s claim was not that Dogge’s video was substantially similar to Teller’s performance of “Shadows” (indeed, there are many different performances of “Shadows,” though they are all nearly identical), but rather that it was similar to the dramatic work “Shadows” that was embodied in the copy Teller filed with the Copyright Office back in ’83. A performance is something that you can do with a work, just as you can reproduce or adapt or distribute copies of it, it’s not the work itself, and only works are copyrightable.
Throughout this post, I’ve been saying that Teller seems basically right. The reason I’m equivocal lies in one part of the opinion that went largely unremarked. The court remarked that the defendant’s work was substantially similar to Teller’s in part because “both performances are based on the incredibly unique concept of a performer cutting parts of a rose’s shadow, thereby cutting the corresponding parts of a real rose.”
Putting aside the court’s problematic usage of “performance” and “performer”, what raises a red flag about this passage is the court’s suggestion that Teller’s copyright extends past his specific expression of the particular dramatic work articulated in the deposit copy he included with his copyright registration, and applies generally to the “concept” embodied in his dramatic work.
This phrasing seems to flout copyright’s good old idea/expression dichotomy, though as with all idea/expression issues, the distinction is a hard one to draw. I’m OK with the outcome in Teller because Dogge’s video mimicked Teller’s work down in detail with only a few exceptions. But one could imagine variation on “Shadows” that are not as slavish in their copying. Consider a variant where a garrulous magician cut the petals from the projected shadow of a sunflower, causing the real thing to fall. Or more abstractly, imagine a version where a talkative illusionist came out and cut the limbs off of the projected shadow of the human effigy of some great historical villain (Hitler, Stalin, Donald Sterling), causing the limbs on the actual figure to fall off. I think the latter two would be far enough from Teller’s work to be allowable, even though they are based on the “incredibly unique concept” that animates “Shadows.”
Finally, if Teller is (mostly) rightly decided, does that mean that magic tricks are copyrightable, contra the major premise of Jacob Loshin’s really cool article on informal means of protecting illusions in the magician community? No way. What Teller owned was the copyright in a dramatic work that happened to contain a magic trick. The underlying idea that animates the specific expression of the performance remains, in my opinion if not the D-Nev’s, fair game.
*This 1987 movie, featuring Warren Beatty and Dustin Hoffman lost in the Sahara Desert, was generally considered unspeakably awful and became a legendary Hollywood bomb. This tends to be my go-to reference for Hollywood disasters because, unlike Waterworld or Cutthroat Island or Heaven’s Gate, I actually saw Ishtar in the theater when I was a kid. God knows why. Maybe it was some form of punishment.
Interesting writeup. A couple thoughts:
1. Waterworld isn't as bad as people say - way better than Ishtar. (but maybe that's me)
2. I follow your analysis of Teller and agree with it. But I'm not sure how it distinguishes from Garcia. Garcia claims to be an author of a segment of the movie. Now, this is debatable for all sorts of reasons (some of which you mention). But if she gets over that hurdle, then YouTube is performing a version of her copyrighted work, just like in Teller.
3. Thus, it seems like the only distinguishing factor between Teller and Garcia is that Teller's fixed performance is protected as a dramatic, pantomime, or audiovisual work and that Garcia's fixed performance is not, either because it's not fixed by her, or because her acting the lines didn't contribute original authorship, or we believe in the Aalmuhammad view that you are either a joint author or you or nothing, or whatever your objection is.
Maybe that's another way of saying what you said, but maybe not.
Posted by: Michael Risch | Apr 29, 2014 12:15:50 PM
Fair point that the distinction between Teller and Garcia is hazy--that is the puzzle that inspired this post. The distinction I draw is quite formalist. Where, as in Teller, a dramatic work is registered by a sole author and described in careful detail in the deposit copy, then I think the major problems raised by Garcia are avoided.
But as you suggest, just because Teller is a clear case of ownership doesn't explain why Garcia is not a plausible case of ownership. So the point of my post is really a defense of Teller rather than a critique of Garcia, though I still think Garcia was awful and will get roundly rejected by the 9th Cir en banc.
And in terms of Waterworld and other Hollywood bombs, my sense is that "bomb" status is mostly a matter of a film's failing at the box office, which is a different thing entirely than not being a good film. Heaven's Gate destroyed Cimino's career and almost brought down Orion, but many film critics have said that the director's cut is actually a very good movie--just not one, apparently, that audiences wanted to see.
Posted by: DF | Apr 30, 2014 7:33:02 PM