Archive for March, 2014

Garcia v Google

by Daniel Gervais on Saturday, March 1st at 12:31 AM

The majority opinion in Garcia v Google (Ninth Circuit, February 26, 2014) seems to stand for the proposition that an actor has copyright in her performance. The case was described as horrific and has generated a very significant amount of traffic on listservs and social media.

Chief Judge Kozinski wrote, inter alia, the following:

  •  An actor’s performance, when fixed, is copyrightable if it evinces “some minimal degree of creativity . . . ‘no matter how crude, humble or obvious’ it might be.” Feist Publ’ns, Inc. v. Rural Tel. Serv. Co., […]. That is true whether the actor speaks, is dubbed over or, like Buster Keaton, performs without any words at all.
  • Where, as here, an actor’s performance is based on a script, the performance is likewise derivative of the script, such that the actor might be considered to have infringed the screenwriter’s copyright. And an infringing derivative work isn’t entitled to copyright protection.
  • As the above discussion makes clear, any analysis of the rights that might attach to the numerous creative contributions that make up a film can quickly become entangled in an impenetrable thicket of copyright. But it rarely comes to that because copyright interests in the vast majority of films are covered by contract, the work for hire doctrine or implied licenses.

 

The dissent makes a number of important points, including:

  • Section 101 of the Act is also instructive, because it differentiates a work from the performance of it. It defines “perform a ‘work’” to mean “to recite, render, play, dance or act it.” 17 U.S.C. § 101 (emphasis added). Given this provision, it is difficult to understand how Congress intended to extend copyright protection to this acting performance.
  • An acting performance resembles the “procedure” or “process” by which “an original work” is performed. Therefore, “[i]n no case does copyright protection” extend to an acting performance, “regardless of the form in which it is described, illustrated, or embodied in” the original work.

The dissent is correct in pointing out that the US Copyright Act draws a distinction between works and performances. But that is not dispositive. In this comment, I do not address all aspects of the case, far from it. I leave aside in particular the remedy issued by the court.

Let me begin by positing that actors can (and usually do) make what one might describe in everyday language as creative contributions to audiovisual productions such as a motion pictures. This might lead to the suggestion that actors should be protected against some unauthorized uses of their creative contributions.  In copyright terms, actor arguably make “creative choices,” the common test for originality in many countries and the Berne Convention, though there are variations in its application.

A similar analysis applies to music. Is Yehudi Menuhin’s version of the J.S. Bach’s sonatas not “creative”? Does it not exhibit “creative choices”?

From an international perspective, the debate seems a bit surreal. The protection of music performers was discussed in the 1950s.  A view held by many (mostly European) negotiators at the time was that at least some music performances could be considered “works” in their own right, but many also felt that parsing those worthy of work status from more “mechanical” performances (that is, not exhibiting sufficient “originality”)–combined with strong pressure from certain associations of composers–doomed the case for recognition of authors’ rights for performers. Eventually the 1961 Rome Convention was signed, protecting the rights of performers, broadcasting organizations and producers of sound recordings, generally referred to as neighboring or related rights (because they are neighboring on or related to authors rights). In part because the United States had issues with rights being recognized for performers and broadcasters, a separate convention (the Phonograms or Geneva Convention of 1971) was signed 10 years later. It protects only rights of record companies. The US joined that Convention in 1974–after the introduction of sound recordings protection under federal copyright law. The US never joined Rome.

As a result of the creation of related rights, most countries do not have a “Garcia issue.” They protect musical and audiovisual performances under a related right. For actors, this often includes a loss of exclusive rights after consent to the incorporation of the performance in the audiovisual production (or some other arrangement meant to allow the producer to exploit the work).

The term “performers” in Rome and the WPPT includes actors.  However, the WPPT essentially applies to “aural performances or performances fixed in phonograms.”  The Rome Convention also essentially excludes audiovisual performances (art 19), but only “once a performer has consented to the incorporation of his performance in a visual or audio–visual fixation,” which might not apply to the peculiar Garcia v Google fact pattern .  The same can be said of the recent Beijing Treaty, which provides a series of rights for performers, including fixation and distribution but also provides (art. 12(1)) that a “Contracting Party may provide in its national law that once a performer has consented to fixation of his or her performance in an audiovisual fixation, the exclusive rights of authorization provided for in Articles 7 to 11 of this Treaty shall be owned or exercised by or transferred to the producer.”

With the WIPO Performances and Phonograms Treaty (to which is US is party), arguably performers have most of the rights that authors enjoy under Berne, including a moral right (WPPT, art. 5).  Note that the WPPT singles out two of three Rome categories (performers and record producers) and is still a ‘related rights” instruments.  A possible treaty on the third Rome category (broadcasters) is under consideration at WIPO.

The US statute does not officially recognize a related rights category. In practice, it does of course.  There are three levels of copyright in the statute:

  • A full (Berne level) copyright for authors of certain works of the visual arts, which benefit from the rights in sections 106 and 106A.
  • A set of copyright rights which is more or less the equivalent of Berne economic rights for all works not affected by VARA/106A and other than sound recordings
  • A “related right-like” right for sound recordings with the 106 rights, excluding public performance but the (more recent) addition of rights in digital transmissions.

 

The Ninth Circuit’s intuition that (a) an actor should have rights where consent to fixation is in doubt and (b) that her contribution is “creative” are both correct. But because the only vehicle available under US law is “copyright,” the court was more or less compelled to apply Feist and the notions of derivative work, work-for-hire and joint work to the situation at hand. That is in large part why the outcome seems messy to a number of commentators.  A related rights regime typically does not have to address those issues.

The related rights approach is not a legal panacea. Indeed the exclusion of performances has been criticized and seem arbitrary (2 RICKETSON AND GINSBURG, at ¶¶19.01 and ssq). The border between unfixed choreographies (protected under the Berne Convention since 1967) and a performance is fuzzy. Though this is for another post, my intuition is that the border can be drawn using the distinction between imitation and reproduction. I reproduce a choreography (while varying stylistically as in the Anne Teresa De Keersmaeker v Beyoncé example) but I would say one can imitate not reproduce Clark Gable in Gone with the Wind. This is of course an issue that would need much more in depth treatment. [Thanks to Danielle Chien for suggesting this example]. By contrast, the fact that a performance is derivative seems a less rueful avenue because translations and any other works are necessarily derivative.

A final point is that the US statute is perhaps stuck in its own contradiction(s) here. Considering a sound recording or a movie as a work (which other countries prefer to consider as a related right) means that it must be original (Feist). In the case of a sound recording, that originality can perhaps follow from the studio engineer’s work (up to a point) but let’s be clear: it really comes from the performance. A motion picture is a mix of directing & acting (and many other contributions) on the basis of a screenplay. The screenplay is a discrete work. The case is harder for direction and acting. It was correct, therefore, for the court to say that the performance exhibited the traits of a creative contribution. Whether it is separately copyrightable (with possible ownership issues when it is not a work-made-for-hire) is another matter, but at least as a matter of international law the absence of the actor’s consent leads to the recognition of a right under both Rome and under the Beijing Treaty.  No doubt someone will be quick to point out that US has yet to ratify either one of those. True, but the underlying normative heft of requiring consent for the fixation to exhaust the performers rights strikes me as valid.

 
Saturday, March 1st at 12:31 AM