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

Final (published) version of Plain Packaging & TRIPS

by Daniel Gervais on Wednesday, December 11th at 4:07 PM

The Article, coauthored with Professor Susy Frankel, was just published in the Vanderbilt Journal of Transnational Law. It is available on SSRN.

Wednesday, December 11th at 4:07 PM

TRIPS database of articles now has 1,075 entries

by Daniel Gervais on Monday, December 9th at 8:37 PM

Click on Articles link on right

Monday, December 9th at 8:37 PM

The Canada Seal panel report: Possible Impacts on IP Disputes?

by Daniel Gervais on Wednesday, November 27th at 6:59 PM

In a piece coauthored with Professor Susy Frankel, we examine the applicability of notions such as justifiability to diminutions of IP rights. The article argues that to interpret the scope of restrictions imposed by TRIPS, including article 20, on WTO member States, one should look beyond the text of each article and consider articles in context, with due regard for the object and purpose of the provision and of the Agreement.

In a panel report issued on the importation and marketing of seal products from Canada issued  on 25 November, 2013, and dealing in large part with the Agreement on Technical Barriers to Trade (TBT), a WTO dispute-settlement panel seems to agree with this approach, which also meshes well with previous panel and Appellate Body reports.

Some of the panel’s findings are illuminating. Here is a taste:

“[T]he object and purpose of the TBT Agreement to strike a balance between the objective of trade liberalization and Members’ right to regulate, the Appellate Body concluded that Article 2.1 should not be interpreted as prohibiting any detrimental impact on competitive opportunities for imports in cases where such detrimental impact on imports stems exclusively from a legitimate regulatory distinction.”

Turning to Article XX GATT, the panel noted this:

“The necessity of a measure within the meaning of Article XX(a) is determined through ‘a process of weighing and balancing’ of ‘all the relevant factors, particularly the extent of the contribution to the achievement of a measure’s objective and its trade restrictiveness, in the light of the importance of the interests or values at stake’. The more vital or important the values or interests furthered by a measure are, the easier it will be to accept that measure as necessary. According to the Appellate Body, ‘if this analysis yields a preliminary conclusion that the measure is necessary, this result must be confirmed by comparing the measure with its possible alternatives, which may be less trade restrictive while providing an equivalent contribution to the achievement of the objective pursued’.

In tackling the relationship between the TBT and GATT, the panel said this:

“Given the close relationship between the GATT 1994 and the TBT Agreement and the need to interpret relevant provisions under both Agreements in a consistent and harmonious manner, we consider that an analysis of a measure’s contribution to an objective under Article 2.2 of the TBT Agreement is also relevant to such analysis under Article XX of the GATT 1994. The Appellate Body in US – Tuna II (Mexico) recalled that in assessing the necessity of a measure under Article 2.2, a panel must assess the contribution to the legitimate objective actually achieved by the measure at issue as in other situations, such as for instance when determining the contribution of a measure to the achievement of a particular objective in the context of Article XX.”

The panel’s examination of whether the discrimination in the EU ban between commercial seal hunts and hunts by Inuit hunters is “justified” is similary interesting.  While some of that discussion may be used by Members in the TRIPS Art. 20 context, one should bear in mind that, as a substantive matter (that is, independently of burden of proof issues), an analysis of “arbitrary and unjustifable” under art. XX of GATT  is not necessarily the same as the justification required under art. 20 of TRIPS.

All of this also shows, it seems to me, that a civil discussion of applicable rules is possible–indeed desirable–even (or especially) where an important public policy matter is at stake .

Wednesday, November 27th at 6:59 PM

Four new items: TRIPS & Development, patentable subject matter, the three-step test and copyright formalities

by Daniel Gervais on Wednesday, November 27th at 6:13 PM

A brief decriptive paper on TRIPS & Development (chpater in upcomign book by Sage) is avilable here.

An updated version of my Patent Target article available here

A new paper on the Three-Step Test, coauthored with Martin Senftleben and Christophe Geiger, is available here.

Finally, a paper coauthored with Dashiell Renaud on renewed copyright formalities and the Berne Convention is available here.

Wednesday, November 27th at 6:13 PM

major updates to articles page (link on right), with many new articles and hyperlinks

by Daniel Gervais on Friday, July 19th at 10:42 PM

have a look!

Friday, July 19th at 10:42 PM

New (draft) chapter on the protection of Geographical Indications (GIs) as Certification Marks

by Daniel Gervais on Friday, July 19th at 10:35 PM



 A  Cognac After Spanish  Champagne

(For forthcoming book by Professors Jane Ginsburg and Rochelle Dreyfuss, Intellectual Property at the Edge (Cambridge Univ. Press).)

The draft chapter is available here.cognac

Friday, July 19th at 10:35 PM

New Book on Canadian Copyright “Pentalogy”

by Daniel Gervais on Thursday, May 2nd at 2:42 PM

Edited by Michael Geist.  The book can de downloaded at


My chapter looks at the five cases’ impact on collective management.


Thursday, May 2nd at 2:42 PM


by Daniel Gervais on Saturday, April 13th at 2:07 PM

The Articles page (right hand column) is being updated, and links to each article are being added (when available on a public website).


Saturday, April 13th at 2:07 PM

Kirtsaeng and digital goods: Is parallel import and resale of ‘copies’ the real issue?

by Daniel Gervais on Friday, April 12th at 4:40 PM

Friday, April 12th at 4:40 PM