1. The court concluded that Garcia’s performance is an independent work of authorship on her part, though she didn’t write the script. I think that’s right, just as a recorded performance of a song is an independent work of authorship on the singer’s part, though the singer didn’t write the song. (Disclosure: I clerked for Chief Judge Kozinski, and consider him a personal friend. I have also done some legal work for Google, such as writing this commissioned white paper, but not on any matter related to this case.)
The dissent disagreed, reasoning that “Under this line of cases, an actress’s performance in a film is more like the personal act of singing a song than the complete copyrighted works in [previous cases]. As a result, it does not seem copyrightable.” But while a singer’s unrecorded voice isn’t protected by copyright, a particular performance by a singer that is recorded with the singer’s authorization is copyrightable.
2. The court concluded that Garcia’s performance wasn’t a “work made for hire” under the Copyright Act. If I hire you as an employee to write (or act or sing) something for me, then the work becomes my intellectual property, not yours; but the court concluded that Garcia wasn’t an employee, but was rather a contractor asked to do a particular brief task. “Youssef hired Garcia for a specific task, she only worked for three days and she claims she received no health or other traditional employment benefits.” (The dissent disagreed, and the question on this point is a bit technical; I think the majority was probably right.)
Alternatively, if I have you contribute certain things to a broader work I’m creating, and have you do that as a contractor rather than an employee, your contribution could still be a work made for hire if there was a written agreement to that effect. Indeed, that’s how professional filmmakers often protect themselves against these sort of lawsuits. But Youssef, who was far from professional, never got his actors to sign any such agreement.
3. The court concluded that Garcia hadn’t licensed Youssef to use her performance in the movie. Now even in the absence of an express license of intellectual property rights, an actor’s participation in making a movie can constitute an implied license of such rights. (The Copyright Act requires a signed writing to transfer exclusive rights, but nonexclusive licenses can be conveyed even without such a writing.) But given Youssef’s deception of Garcia, the court says, any such implied license didn’t extend to the movie that Youssef actually produced:
A non-exclusive license may be implied from conduct and arises where a plaintiff “create[s] a work at defendant’s request and hand[s] it over, intending that defendant copy and distribute it.” … Any such license must be construed broadly. If the scope of an implied license was exceeded merely because a film didn’t meet the ex ante expectation of an actor, that license would be virtually meaningless….
Nevertheless, even a broad implied license isn’t unlimited. Garcia was told she’d be acting in an adventure film set in ancient Arabia. Were she now to complain that the film has a different title, that its historical depictions are inaccurate, that her scene is poorlyedited or that the quality of the film isn’t as she’d imagined, she wouldn’t have a viable claim that her implied license had been exceeded. But the license Garcia granted Youssef wasn’t so broad as to cover the use of her performance in any project. Here, the problem isn’t that “Innocence of Muslims” is not an Arabian adventure movie: It’s that the film isn’t intended to entertain at all. The film differs so radically from anything Garcia could have imagined when she was cast that it can’t possibly be authorized by any implied license she granted Youssef.
A clear sign that Youssef exceeded the bounds of any license is that he lied to Garcia in order to secure her participation, and she agreed to perform in reliance on that lie. Youssef’s fraud alone is likely enough to void any agreement he had with Garcia. See 26 Samuel Williston & Richard A. Lord, A Treatise on the Law of Contracts § 69:4 (4th ed. 2003). But even if it’s not, it’s clear evidence that his inclusion of her performance in “Innocence of Muslims” exceeded the scope of the implied license and was, therefore, an unauthorized, infringing use.
The situation in which a filmmaker uses a performance in a way that exceeds the bounds of the broad implied license granted by an actor will be extraordinarily rare. But this is such a case. Because it is, Garcia has demonstrated that she’s likely to succeed on the merits of her claim.
4. The court concluded that, in addition to being likely to succeed on the merits, Garcia had shown that the injunction would tend to prevent “irreparable harm” — here, the threat to her life from the display of the (infringing) movie:
Garcia has shown that removing the film from YouTube will help disassociate her from the film’s anti-Islamic message and that such disassociation will keep her from suffering future threats and physical harm…. Taking down the film from YouTube will remove it from a prominent online platform — the platform on which it was first displayed — and will curb the harms of which Garcia complains.
To be sure, the harm is different in nature than what is usually involved in copyright cases. Most copyright cases involve economic harm (defendant’s infringing work competes with plaintiff’s original work) or privacy harm (defendant is about to publish plaintiff’s unpublished letters). But the harm is certainly no less grave than typical copyright harm.
The dissent argued that, “while Garcia has provided undisputed evidence of past threats and injuries, she has failed to link her allegations of future harm to potential future viewings of the film on YouTube” (emphasis added). Therefore, the dissent argued, the court should defer to the trial court’s conclusion that there wasn’t sufficient showing that an injunction would avoid irreparable harm. This is an interesting question, but I’m inclined to the majority’s view on it.
5. Finally, the court concluded that the preliminary injunction doesn’t violate the First Amendment, because “the First Amendment doesn’t protect copyright infringement. Cf. Eldred v. Ashcroft, 537 U.S. 186, 219–220 (2003).” I have argued that preliminary injunctions in copyright cases — entered based on a mere likelihood of success — should often be seen as unconstitutional prior restraints, even though permanent injunctions or damages liability (based on a final judgment following a full examination of the merits) are constitutional. But I’m afraid that courts haven’t adopted those arguments, and the majority’s view of the First Amendment issue under current precedent is likely correct.
So I think that, on balance, the majority opinion is probably consistent with existing law, and probably the best interpretation of existing law. Note, though, some important limitations:
1. The opinion rests on Garcia’s copyright-protected performance. If the video is reposted with that performance removed, Garcia would have no legal basis for objecting.
2. Google made no fair use argument in its Ninth Circuit brief, and the panel didn’t discuss fair use. I think that it’s quite possible that, under the fair use factors, the posting of a movie might be fair use on the part of the poster and of Google.
The work is of important historical significance, and viewing it may be necessary to thoroughly understand the controversy related to the work. Garcia’s performance is only a small part of the work. The video does not harm “the potential market for or value of the copyrighted work,” namely Garcia’s performance within that video. (Garcia was harmed, but not in a way relevant to this fair use factor.)
The posting of the work may not be commercial on the part of the poster, and has only a small commercial benefit to Google. And while some factors — such as Youssef’s deception in getting Garcia to participate, the general nature of YouTube as a profit-making enterprise, the use of a large part of Garcia’s performance, and the creative (rather than factual) nature of Garcia’s performance — cut against fair use, I think on balance a fair use defense would be strong.
I think Google could still raise the fair use issue at the trial court, in the remaining litigation over whether the preliminary injunction should be made permanent. Moreover, if someone reposts this on another site, that site should have a fair use defense available to it. The entire video ought to remain available to those who want to get a full understanding of the controversy, and the fair use doctrine ought to be a mechanism for doing that.