You Stole My Work! And You Stole It Poorly!

I'd like to introduce my subject—the practical and theoretical complexities embedded in choreographic authorship and the legal regime of copyright—by way of an example. In 2015, choreographer Amanda Myers filed suit against her former employer, The Dance Shop (TDS), and its owner, Christina Harold, for copyright infringement of her choreography. Myers had worked at TDS as a dance teacher from 2011 to 2015, and part of her duties included choreographing dances for TDS's dance team. She left TDS in early 2015. Later, TDS used her choreography at an end-of-season recital and in summer performances without her permission. Myers objected to the unauthorized use of her artistic work by her former employer. Her complaint did not stop there, however; she was also unhappy with the way in which her choreography was executed. In her deposition, she complained that the allegedly infringing performances were not done according to her artistic wishes. She explained, “I just feel that since it wasn't mine and I didn't rehearse it, I didn't clean it, that it could have damaged my reputation as a choreographer” (Myers v. Harold et al. 2017; emphasis added). Here Myers makes two complaints that are actually in some tension with each other. On the one hand, she objects to the unauthorized use of her choreography by TDS. This contains an implied ownership claim—the dances were “hers,” and therefore, as copyright holder, she should control whether and when they are performed. On the other hand, she is also unhappy with the quality of the unauthorized performances. Because she had no oversight of the rehearsal and execution of her choreographic works, her artistic intentions have been misrepresented: “It wasn't mine,” she said, referring to the work performed. It is not difficult to understand her objections to the unauthorized use of her work, even as we also notice its ostensible contradiction: she felt that the dances performed without her oversight were not “hers,” even though of course she also is claiming that they were hers—so much so that she filed suit against TDS for copyright infringement.

This dual complaint, viz., “You stole my work! And you stole it poorly!” might seem paradoxical, like the old joke about the woman who complains that the restaurant's food tasted bad and that it came in such small portions. Yet anyone who understands the structure of a performed art, such as music or dance, in which the work consists of an abstract score that must be instantiated, can grasp the distinction that Myers is invoking here. She is basically asserting that there have been two separate harms done to her as an author of an artistic work: that her copyright was infringed through the unauthorized performances, and that those viewing the inept performances will not be able to see the difference between the work as it was performed and what it was supposed to have been. As a result, they might draw negative conclusions about her artistic judgment and capabilities based on their encounter with the infringing performance of the work.Footnote ² And these concerns about artistic reputation are not trivial. Although it is common in copyright law to distinguish between the economic and the noneconomic or reputational interests of artists in their work, these are of course intertwined: artistic reputation is a form of wealth, both metaphorically but also quite concretely, as a performance that negatively distorts a choreographer's artistic intentions may affect the number and quality of professional opportunities she is offered in the future.Footnote ³

Myers's seemingly self-contradictory complaint—that the work is both hers and not hers—can be resolved when we recognize that authorship involves more than just a factual claim about who is responsible for having created the work. It also involves authorization.Footnote ⁴ Philosopher Dennis Dutton has articulated this as the distinction between nominal authenticity and expressive authenticity: an author might acknowledge that they made the work, but nevertheless disavow it because it fails to express their artistic intentions (Reference Dutton and Levinson2003). The worry in Myers's case is that a “bad” performance of the choreography—that is, one which does not satisfy the artist's intentions for how it should be presented—will not just reflect poorly on the dancers but on the choreographer as well.Footnote ⁵

In the performing arts, the distinction between the work itself and a given performance of the work seems, on the surface, to be an obvious and essential one to make. When music critics compare performances of Beethoven sonatas by different pianists, we have no trouble understanding, at least conceptually, that they are differentiating between the work itself and the pianist's execution of that work. Philosophers sometimes articulate this distinction in terms of the constitutive features of the work, which are essential to its work identity or “type,” and the contingent ones, which vary with a given interpretation or performance, also called the “token.” Presumably, fine art or “concert” dance operates according to the same type-token relationship: the choreography, the sequence of movements, and the instructions for how those movement sequences should be executed, is metaphysically separate from any particular iteration thereof.

We accept this as commonplace, and at the same time, because the finer nuances of kinesthetic style and movement impulse cannot be notated, we are unable to fully extricate the work itself from the bodies that actuate it in a given performance. The distinction between the work itself and the performances thereof that we readily assent to in theory is not so easy to parse in practice. Insofar as one's access to the choreographic work is only through a given performance, it can be difficult if not impossible for a viewer to determine which aspects of the performance deviate from the work unless they have prior, specialized knowledge of the choreographer and the piece in question. How do we separate what counts as essential to the choreographic work from the adventitious features of a given performance, instantiated by a particular body or bodies, at a particular moment in time? These questions, articulated in different ways, are at the heart of most scholarly discussions surrounding choreography and copyright in the legal literature. The general consensus seems to be that choreography is by nature poorly suited to the conceptual framework provided by copyright, even as there is widespread agreement that works of dance deserve the legal protection and cultural endorsement that its inclusion represents.Footnote ⁶ In the next section, I give a brief overview of the history of choreography and copyright. Then I reexamine the factors that are often cited as barriers to choreography's suitability for copyright. I argue, contrary to many commentators on the subject, that choreography is better suited to the copyright regime than it appears, so long as we keep in mind the distinction that Myers invoked in her complaint against TDS: one may claim authorship over a movement sequence even if the execution in a given instance is artistically unsatisfactory.

Before I go on, however, I want first to address a concern that I anticipate will arise from some of my argument's fundamental premises. In this essay, I take as given the following: at least some choreographic works are authored; these works are regarded by both their authors and the community as their intellectual property; and at least part of our interest when we see performances of such works is whether they accurately represent the artistic intentions of the choreographer-author, such that we can say that the performance was an instantiation of the work in question. I make no general claims or assumptions about the nature of dance generally (I highly doubt there is one single adequate definition that could cover all of the practices we label as “dance”), and I do not claim that all dance artists view their work in this way.

However, I am aware that my approach in this essay may sound out of step with contemporary dance theory, and that I am perhaps naively advocating what Mark Franko calls “the old myth of repeatability” (Reference Franko1989, 73). There is a well-known and highly influential line of thought in contemporary dance theory and practice which challenges the notion that choreographic works are fundamentally tied to the intentions of their nominal authors, and which seeks to liberate the creative potential of a choreographic work from “the author's intentions as commanding authority over a work's afterlives” (Lepecki Reference Lepecki2010, 35). These theoretical and artistic attempts to destabilize the primacy of a certain understanding of authorship as a kind of ownership and control were articulated by Roland Barthes and Michel Foucault in their influential essays on the topic half a century ago, and I assume I do not need to rehearse those arguments here.Footnote ⁷ As I understand it, dance theorists such as Franko and André Lepecki are concerned that choreographic performances be understood as events that are active, living, and replete with creative possibility, rather than dutiful reenactments dictated by a need to be faithful to some absent original. Repetition and authorship are conceptually intertwined here: the governing expectation of repeatability implicates the choreographer's intentions and authority as the ultimate standard for what counts as correct.

In taking up the subject of choreography and copyright in this essay, I am choosing not to engage what for some may be the threshold question of whether there is, or should be, such a thing as choreographic authorship in the conventional sense, and whether choreographers are justified in making intellectual property claims over their creations. I understand that for some dance artists and theorists, these are essential and importantly political concepts that need to be resisted and challenged. Perhaps it will be objected that the practices of concert dance and choreographic authorship are particular to a certain Anglo or European artistic tradition, and that this history alone makes it politically suspect, given the oppressive hegemony of Western European culture and property regimes.Footnote ⁸ In this essay, however, I have a more modest aim, which sets aside some of the more radical political and metaphysical concerns one might have about the implications of owning a choreographic work. I take as my starting point a couple of observations, supported by my own years of experience as a dance student and audience member: that at least some contemporary choreographers—of all races and nationalities—do regard themselves as creative artists who create and present concert dances; they have proprietary and reputational interests in their creations; and they retain the right, enforced either by law or informal community norms, to authorize the dissemination and presentation of their artistic works because of those interests. In my experience, these norms surrounding authorship and authorization do not simply go away even when the artist in question is aligned with a post-structuralist or political tradition that explicitly challenges those same norms of authorship.Footnote ⁹

Rather than engage with post-structuralist critiques of authorship and authenticity, in this essay I examine a different, yet related, concern regarding choreography's suitability for a legal copyright regime. The basic claim, which I examine in detail in the second section, is that choreography is by nature incompatible with copyright because it is a uniquely embodied art form that cannot be distilled into a score; hence, the choreographic work cannot simply be copied or imitated by some unauthorized performer. Unlike music, choreography is not reducible to an abstract structure that can be instantiated by anyone with sufficient technical ability to bring the score to life in movement. Yet copyright as a legal concept assumes and requires, for both conceptual and practical reasons, that the artwork be the kind of thing that others can copy. There must be some original, authored work or score against which allegedly infringing imitations can be compared. But these basic requirements for choreographic copyright necessarily implicate an understanding of what the identity conditions of a choreographic work are, such that a judge would be able to find that an unauthorized performance counts as infringement. Some theorists hold, however, that the uniquely embodied nature of dance makes it virtually impossible to imitate without expert coaching. If the dance movements are not executed in just the right way, then it is in principle not the same dance. The perhaps unintended consequence of maintaining choreography's ontological uniqueness is that the theory basically endorses dance's marginal status in the US copyright regime.

In what follows, I argue that this problem can be overcome. It is a matter of recognizing two different standards for assessing whether a given dance performance is accurate enough to count as an instantiation of a given choreographic work. There is the standard of “substantial similarity,” which US courts use when assessing alleged instances of copyright infringement. I call this “de jure similarity.” But this should not be confused—though, as I argue below, it often is—with the much higher connoisseurship standard of similarity for assessing whether a given performance of a dance work got it “right” according to the creator's standards. I call this “de facto similarity.” Put another way, there can be a difference between the legal standard for determining the identity of a dance work when considering whether a given performance counts as infringing, and the artistic or cultural standard we might want to apply in a connoisseurship context.

In the final section, I offer a case study of choreographic copying that was adjudicated in the court of public and scholarly opinion, but not in a formal courtroom: the plagiarism of Anne Teresa De Keersmaeker's choreography in Beyoncé's “Countdown” video. As I see it, Beyoncé's so-called borrowings were exactly the kind of unauthorized imitation that copyright is designed to protect choreographers against. Although there are multiple factors leading to the lack of critical consensus on that controversy, I argue that one of the most salient issues concerns a conflation, not just by scholar-commentators but by the choreographer herself, of the legal and connoisseurship standards of similarity. In order for choreography to be meaningfully subsumed under the US copyright regime, one needs to be able to claim that a dance copycat has stolen one's movement sequence, while simultaneously withholding endorsement of the skill or artistic accuracy with which it was performed.

Copyright and the Ambiguity of the Choreographic Work

Since the 1976 extension of copyright protection in the United States to abstract, nonnarrative dance works, surprisingly few cases claiming infringement of choreography have appeared in the courts.Footnote ¹⁰ One might wonder why this is the case, and why it matters. The traditional justification for copyright protection in the United States is that it incentivizes cultural productivity by rewarding authors.Footnote ¹¹ There is also a symbolic significance to copyright: by giving choreography the legal protection that copyright confers, the legislature recognizes that dance is just as legitimate and valuable a fine art as painting, sculpture, literature, and film. But the fact that there have only been only a handful of cases in US federal court involving choreography and copyright in the forty years since nonnarrative dance was included, calls into question whether copyright actually has much to offer choreography other than a symbolic token of esteem.Footnote ¹²

Perhaps the law has been exceedingly effective in deterring would-be dance copycats from unauthorized imitations. It is more likely, however, that choreographers who object to the illicit reproduction of their material are reluctant to sue for certain practical reasons.Footnote ¹³ Among the probable factors that inhibit the decision to file suit are the prohibitive cost, especially for practitioners of an art form in which profits are scarce, and the cultural norm (perhaps born of necessity) that, in the dance community, ostensible copyright infringements will be informally policed by its own members as plagiarism. Plagiarism is the ethical, rather than legal, term of reproach for illicit copying.Footnote ¹⁴

But there is another factor that may contribute to the paucity of copyright suits involving dance: the lack of critical consensus, both in the realm of dance theory and in the legal scholarship, about the nature of a dance work and what its identity conditions are. The 2012 policy statement by the US Copyright Office seems relatively straightforward on this matter. It states that choreographic authorship is considered, for copyright purposes, to be “the composition and arrangement of a related series of dance movements and patterns organized into an integrated, coherent, and expressive whole.”Footnote ¹⁵ This definition is better suited to contemporary practice than the pre-1976 statute, insofar as it defines a choreographic work in terms of its form—as a composed sequence of movements—rather than its content. We will leave to the side the evident circularity, insofar as the definition suggests that a protected work of dance must be made up of dance movements. This is a form of question-begging, as the definition does not articulate just what counts as a dance movement and what does not. It also ignores decades of formal innovation in which choreographers have made dance works out of pedestrian movements and gestures.Footnote ¹⁶

In the United States, copyright understands works as writings that must be fixed in a tangible medium of expression in order to be protected.Footnote ¹⁷ Although the word “choreography” literally means “written dancing,” many commentators on choreography and copyright have pointed out that the dance work, unlike classical music or theater, cannot simply be written down, either with notation or video recording.Footnote ¹⁸ Notational systems for capturing dance are prohibitively expensive, labor intensive, and imprecise. Film and video, on the other hand, can distort the choreography by flattening it out and presenting it from a limited number of angles; it also captures a single instantiation of the choreography, so that it can be unclear which elements are part of the ideal work and which are unintended contributions from the performers. Much of the extant literature on dance and copyright focuses on the fixation condition because it poses such a problem for choreographers and for the philosophical question of what the choreographic score consists in.Footnote ¹⁹

One might argue that an original work of art, even a choreographic one, should not have to be fixed in a tangible medium in order to enjoy legal recognition and protection. But the fixation requirement serves an important practical purpose: it is necessary because judges need to have a paradigm against which to compare putative instances of copying should an allegation arise. In the United States, in order to make an infringement claim, the plaintiff must show three things: First, they must show that the defendant had access to the original work, as an identical work created independently is not infringing. Second, the allegedly infringing work must be “substantively similar” to the original. The standard of what counts as substantial similarity is notoriously difficult to define, even for other more “stable” art forms.Footnote ²⁰ Third, even if the secondary work is substantially similar, the court must assess whether it should count as an instance of “fair use,” either because it is used for educational or reporting purposes, or because the secondary use transforms the work in such a way that it effectively creates a new work, even though substantially similar to the original one.

One reason why choreographic works fit awkwardly within the copyright regime is because there is no adequate system of notation to capture, in abstract form, the essence of “the work itself.”Footnote ²¹ But this means that the question of what counts as “substantial similarity” to the original is even more vague and ill-defined. This uncertainty surrounding the standards for substantial similarity in unauthorized uses of dance choreography may contribute to a vicious cycle, whereby the paucity of litigation in the area of dance copyright yields very little case law to use as a guideline for what counts as infringement, which in turn may inhibit dance artists from pressing their cases in court because of the uncertainty that surrounds the likelihood of success.

The marginal status of dance with respect to copyright can be attributed to two main factors: history and ontology. From the historical point of view, the concept of a choreographic work is a relatively recent invention. Stephanie Jordan points to the “stabilization of the dance work as an independent conceptual entity during our century, with a relatively fixed form, devised by one acknowledged choreographer and usually one composer” and that this contrasts with “earlier practice where ballets were highly flexible conceptions that could be edited, added to, or remoulded for different occasions and casts” (Reference Jordan2002, 4). Renee Conroy echoes this idea, noting that “the idea of the dancework as a stable, enduring art object, is a distinctly twentieth-century notion, one that is plausibly the product of analogizing dance to other art forms in an attempt to improve its artistic status” (Reference Conroy, Adajian and Ribeiro2012, 162). The observation that dances have only recently come to be seen as fixed works of choreography connects to Lydia Goehr's argument that the concept of a musical work with a fixed score is itself a modern invention (Reference Goehr2007).Footnote ²² Our failure to historicize the work-concept in music, she argues, leads us into unnecessary and misleading philosophical puzzles about the ontology of musical works. The emergence of the choreographic work as a regulative ideal is even more recent, and many of the problems that Goehr points to with respect to music apply here as well: we assume that the choreographic work is a fixed and stable abstract entity, when in fact for both historical and practical reasons this has not been the case.

A corollary to the application of the work-concept to dance is that such a work must have a nominal creator, an author who owns the choreography as property. Francis Sparshott points out that the cultural shifts in the performance and consumption of dance works had implications for its subsequent status as copyrightable:

The legal regime of copyright presumes that the work has one or more author-creators who own the work and maintain control over its circulation. This, too, is a relatively recent and narrow conception of what dance is and how dances are made. Even if we, as dance theorists, want to resist the notion of choreographic authorship for some of the reasons I have pointed to above, we may have to embrace this legal fiction as a necessary and constitutive feature of our culture. Choreographic artists build their careers and reputations on the dances that circulate in their name (just as dance theorists do with their articles and books!); as audience members, we attach critical praise or blame to the authors of those works as well as to the performers who execute them. Whatever high-minded theoretical objections one might have to the notion of choreography as intellectual property, these objections can and do coexist with the pragmatic realities of the concert dance system.Footnote ²³

However, what is frequently cited as the main obstacle to subsuming choreography under the umbrella of the work-concept, which copyright presupposes, is not its historical status as a latecomer but rather its ontological peculiarity as situated in the human body. Copyright treats choreography as akin to a musical score, insofar as it represents a set of instructions for carrying out a performance. The problem is that dance is different from music in some decisive ways. Drawing on Nelson Goodman's distinction between allographic and autographic forms of art, Joseph Margolis argues that dance is actually an “autographic” art form more akin to unique paintings and sculptures than scored music because genuine instances of dance works cannot be generated from notational scores in the way that a musical work can be. The key variable for Margolis is the nature of the body and its centrality to the dance work: “The body is not an instrument for a dancer in the sense in which a violin is for a musician, or a brush for a painter. We cannot help but see the peculiarly intimate and sui generis sense in which the dancer and the mime use themselves in performance” (Reference Margolis1981, 421). A consequence of the argument that a dance performance is inextricably linked to the unique physicality of the dancer(s) who perform it is that the identity of the dance work is not simply transferrable from one set of performers to another.

Frédéric Pouillaude agrees with Margolis's observation that dance is uniquely situated in the individual performing body, and he argues that, for this reason, it belongs neither to the category of unique, singular works, nor to that of infinitely iterable works, such as those made up of text or a score:

On the one hand, Pouillaude suggests, the notion of a choreographic work implies it is an abstract structure, in principle repeatable. On the other hand, however, it is uniquely and perhaps inextricably tied to the dancers who embody the work in performance. For this reason, choreographic works occupy a liminal conceptual space between the autographic and allographic, to use Goodman's terminology, which distinguishes it from other kinds of artworks.

Yet, despite these admitted difficulties with the medium, in practice we recognize choreographers as artist-authors, who at least theoretically have the same status as creators that painters, sculptors, composers, and fiction writers do. We agree that choreographers create artworks, and that they deserve credit for those accomplishments. At the same time, however, many experts on the subject express doubt that dance can be accommodated to copyright's conceptual and practical requirements. The degrees of skepticism expressed fall along a spectrum. On the moderate end we find Julie Van Camp, for example, who claims that “the very notion of ‘similarity,’ regardless of degree, is problematic for choreographic works” and suggests that questions of what counts as the “same work” be left to the dance community to determine (Reference Van Camp, Breimer, Thorne and Viera1994, 72). The problem with this suggestion, as we saw in the Myers case at the beginning of this essay and will revisit at the end with Beyoncé, is that the dance community may endorse a very refined, connoisseurship standard for substantial similarity. This would effectively preclude any performance of a choreographic work that was deemed artistically inadequate by the community from a claim of copyright infringement.

Indeed, another abiding irony that one finds in the scholarship on choreography and copyright is that some of the same theorists who champion the ownership rights of choreographers in their works provide theories of work identity that make infringement practically impossible. At the extreme end of the scale are those who argue that a dance work is not reducible to a movement sequence at all, but it essentially consists of the precise way in which that sequence is executed. Adina Armelagos and Mary Sirridge are perhaps the most often cited exponents of this argument:

According to this view, the identity of the dance work consists not simply in what is done in terms of movement sequence, but how it is performed. In fact, we might say that, according to this view, the “what” essentially entails the “how.” In the case of a dance work, style is substance—so much so that the very identity of the work itself is at stake. Nearly thirty years after Armelagos and Sirridge, Joi Michelle Lakes echoes their view:

As anyone who has trained in dance knows, there certainly is a large gap between a bare grasp of the movement sequence and the ability to perform that sequence with the right kinesthetic quality. It is something akin to learning a foreign language: there is a difference, sometimes insurmountable, between making oneself understood and sounding like a native speaker. The problem with applying such a fine ontological distinction between style and sequence in dance, however, is that style cannot be notated, codified, or fixed in such a way that it can be included in the identity of a choreographic work for the purposes of copyright. Commentators such as Van Camp, Armelagos and Sirridge, and Lakes do not seem to recognize that the implication of their arguments would severely limit what could count as infringement of choreography. Of course, quality of execution is an important expressive element of a choreographic work. This is why many choreographers require that they or a trusted associate coach the dancer(s) in the correct execution of the choreography in order to authorize a performance. Although this emphasis on correct execution, overseen directly or indirectly by the source of the work, serves an important function culturally, it has disastrous implications for the copyrightability of choreography if taken too far. Taken to its logical endpoint, it would mean that a performance of a dance work that is not executed in precisely the right way, under the supervision of the artist or her agent, would not in fact count as an instance of the work at all. It would not be the “same” work, even if the movement sequence were recognizably the same to a casual viewer.Footnote ²⁵

It might seem as though this aspect of fine art choreography—the difficulty, if not impossibility, of reproducing it correctly in the absence of hands-on coaching—would result in a built-in protection against copyright infringement. After all, if the only way to produce a substantially similar copy of a dance work is to be specially trained in its execution, then choreographers are at very little risk of infringement. The only possible case would be if a dancer who had been personally trained by the choreographer went rogue and performed the dance when unauthorized to do so. Surely copyright protection for choreography must have a wider applicability than such unlikely cases as that. I propose that if we are to have meaningful copyright protection of choreographic works, we need to apply a pragmatic, working standard for substantial similarity. It need not be, and indeed cannot be, as high as the standard that the artist or connoisseur would consider essential to a successful performance. I am calling this a distinction between de jure and de facto similarity in dance works.

Beyoncé and De Keersmaeker: Plagiarism but Not Infringement?

As I see it, the critical confusion about the identity conditions of a choreographic work, along with the paucity of cases that make it to the courts, have led to a climate in which even straightforward cases of infringement become caught up in speculative debates about what would count as substantial similarity in a dance. Perhaps the most notorious case to illustrate this problem involves the pop star Beyoncé and the Belgian avant-garde choreographer Anne Teresa De Keersmaeker. In her 2011 music video, “Countdown,” Beyoncé appropriated movement sequences from two of De Keersmaeker's dance works, Rosas danst Rosas and Achterland. This instance of unauthorized choreographic borrowing was notable not only for the widespread attention it received, but for the blatant way in which it was carried out. Because Beyoncé's video reproduced not just some of the movement sequences but even the costuming and staging of Rosas (which from the perspective of copyright is covered separately from the dance sequences themselves), the debt to the source material was unmistakable.Footnote ²⁶ De Keersmaeker publicly objected to the video's borrowings as “stealing” and “plagiarism,” and commentators in both the popular media and the academic literature agreed, though the latter have been reluctant to take either a legal or a moralizing position against Beyoncé, as I discuss below.Footnote ²⁷ We can see the ambivalence about Beyoncé's “borrowings” in both De Keersmaeker's response—a version of “You stole my work! And you stole it poorly!”—and in the analyses by two dance scholars, Francis Yeoh and Anthea Kraut. Before I continue my discussion, I want to point out that I am focusing only on the aspects of this controversy that bear directly on my argument in this essay: the mismatch between the connoisseurship standard for choreographic performances regarding what counts as a proper instantiation of a work, and the legal standard for adjudicating alleged cases of infringement. There is much to say regarding race, class, gender, and power regarding the Beyoncé/De Keersmaeker case, which I do not address here because it is beyond the scope of my argument and because Anthea Kraut has done a masterful job in this regard already (Kraut Reference Kraut2016).

In her public statements about the appropriation of her choreography in “Countdown,” De Keersmaeker denies that she was angry at Beyoncé for using her work without asking. At the same time, however, she describes her choreography as not only having been “plundered,” but she accuses Beyoncé of distorting the original.Footnote ²⁸ Like Myers in the TDS case, De Keersmaeker makes two related but distinct complaints regarding the use of her artistic property. She claims ownership of the source material because she created it—to use Dutton's terminology, this is the nominal moment—but she distances herself from the performance because, in her view, it misrepresents her original choreographic intentions: in other words, it is expressively inauthentic.Footnote ²⁹ In Myers's case, it was not a matter of shifting context—the choreography was put to its intended use for the summer dance school recital—but that she had not authorized its use, and she had not been present to oversee its staging and rehearsal. For De Keersmaeker, on the other hand, the alteration of the work comes mainly from the radical shortening of its duration, but she also complains about the way in which the video's dancers execute the movements. Ultimately, however, she frames it as a problem of shifting the material from an avant-garde, fine art context to pop culture:

De Keersmaeker claims that the performance in the Beyoncé video treats her work superficially, both because of the shortened time frame in which it was presented and in its movement quality. “It's seductive in an entertaining consumerist way,” she says.Footnote ³¹ While accusing Beyoncé of reducing her work to a cliché, one might observe that De Keersmaeker invokes a cliché as well: she repeats the well-worn refrain, leveled by the avant-garde against popular culture, that the latter is concerned with offering the masses empty pleasures, as opposed to the higher yet hard-won gratifications of leading-edge art.

What is fascinating to me about De Keersmaeker's reaction to the “Countdown” case is that she objects not only to the unauthorized use of her work, but she complains about its execution, almost as if to emphasize the insult of her work having been plagiarized, and poorly. I want to maintain, however, that the question of whether the choreographer's copyright has been infringed is different from the question of whether the author is satisfied with its execution. De Keersmaeker is certainly entitled to comment on the performance of her choreography, authorized or not. We can sense in her complaint a sense of propriety over the work: not only was it taken without asking, but it is being put to a use for which it was never intended, and which distorts its meaning. At the same time, however, De Keersmaeker's objections to the way in which her work was presented threaten to undermine her claim of propriety, for, as we have seen, some theorists maintain that a dance work inexpertly executed, not to mention done without the choreographer's authorization, is not strictly speaking an instantiation of the work at all. If the dance sequences in “Countdown,” despite their unmistakable resemblance to Rosas danst Rosas and Achterland, cannot be said to be De Keersmaeker's creation, then there is no copyright violation. The fundamental ontological ambiguity of the choreographic work, as both abstract and embodied, threatens to foreclose the possibility of its being included within the scope of copyright protection.

Just as De Keersmaeker herself expressed, at times, conflicting and ambivalent attitudes toward the use of her work in “Countdown,” expert commentators on the case express similarly uncertain attitudes toward Beyoncé's appropriation, though for different reasons. For example, dance scholar and legal theorist Francis Yeoh argues that there should be increased intellectual property protections for choreography (Reference Yeoh2013b). Nevertheless, his analysis of the “Countdown” case reflects the general ambivalence and confusion in the scholarship surrounding what does or should count as infringement of choreographic copyright. Yeoh acknowledges that Beyoncé appropriated De Keersmaeker's choreography. He refers to it as “borrowings,” a term that eschews the moralizing connotations of “plagiarism,” but may swing too far in the other direction: after all, when one borrows something, it is usually with permission of the owner. While Yeoh acknowledges that Beyoncé sourced some of her choreographic material from De Keersmaeker, he expresses doubt that it would be found to be an infringement of her copyright. Although some of the dance sequences in the “Countdown” video clearly derive from Rosas danst Rosas and Achterland, he is unsure whether the differences in execution by the dancers in the video compared to the filmed original would prevent a finding of substantial similarity. He also wonders whether it would be substantially similar yet transformative. Either way, he argues, a court probably would not have found them to be a violation of De Keersmaeker's copyright:

Here we see that Yeoh offers the same argument for the irreproducibility of choreography that other dance connoisseurs such as Lakes and Armelagos and Sirridge have made: the artist's choreography is so “distinctive” that it is essentially inimitable, unless the dancer has been personally coached. Yeoh takes what he calls the “subtle differences” in the performance of De Keersmaeker's choreography by Beyoncé and her dancers as possibly decisive against a court's finding of infringement (Reference Yeoh2013b, 97).

Yet this seems to be the wrong standard of similarity to appeal to when adjudicating infringement in choreography cases. As the court saw in the Myers case, the unauthorized performance does not need to meet the choreographer's high aesthetic standards in order to count as infringement: that is not what the substantial similarity standard is meant to ascertain. Although the standard for what counts as substantial similarity is notoriously imprecise, it seems clear to me that, artistic nuance aside, the version of De Keersmaeker's work in the video was well within its limits. Indeed, the similarity was obvious and blatant: if one is wondering what the “ordinary observer” might think, one only has to look at the court of public opinion represented in the comments on social media, or the overwhelming consensus in the academic and popular media articles on the subject—that Beyoncé's appropriation amounted to plagiarism. Of course, plagiarism is a moral and academic term, not a legal one. But what stops the judgment of plagiarism in the “Countdown” case from leading to the conclusion that this is also, legally speaking, infringement? If dance scholars insist on an aesthetically impeccable rendering of plagiarized choreography in order for it to count as infringing, then they are letting ontological hair-splitting get in the way of what even Beyoncé's fans can recognize is unauthorized imitation.

However, even if we agree that the choreography in “Countdown” is substantially similar according to the legal standard in the United States, there is another possibility: by transposing the dance sequences to a new context—a pop music video—Beyoncé might have transformed it, and hence avoided a judgment of infringement if it is found to be an instance fair use. Yeoh also considers this possibility in the quote above. Yet Yeoh cannot have it both ways: if the choreography in the video is not substantially similar to De Keersmaeker's, then as a legal matter it is prima facie not transformative, because the latter depends on the former. In order for a court to reach a finding of transformative fair use, it would first have to agree that the movement sections in the “Countdown” video were substantially similar to De Keersmaeker's, and that the secondary use was fair because it made something new out of the appropriation.

This second line of argument—that the “borrowing” of De Keersmaeker's choreography resulted in a transformative, and hence fair, use—is perhaps more promising as a defense against an infringement claim, rather than that the use lacked substantial similarity. It was clearly the defensive strategy of Beyoncé's legal team. While acknowledging that she had consulted De Keersmaeker's work when designing her video, Beyoncé claimed in a statement, “I've always been fascinated by the way contemporary art uses different elements and references to produce something unique.”Footnote ³³ Indeed, the “Countdown” video makes abundant use of references to pop culture icons, such as Twiggy, Audrey Hepburn, and Andy Warhol. But there is an important difference between making visual references to pop cultural icons and reproducing (down to the costumes and staging in the case of Rosas) copyrighted dance works.

Moreover, another important difference in Beyoncé's appropriation of De Keersmaeker's choreography, compared to the poses and clothing of well-known celebrities, is that the Belgian avant-garde artist is emphatically not a pop culture icon, so any putative references to her work would go unnoticed by the vast majority of Beyoncé's audience—an important consideration for what counts as transformative in instances of fair use. Although we can say that Beyoncé transposed De Keersmaeker's choreography from the realm of avant-garde fine art dance to that of a pop culture music video, I would argue that the difference in context does not amount, in and of itself, to a transformation of the work into a new one that builds on the old. It seems to me that Beyoncé would need to meet a higher bar with her “borrowing” of De Keersmaeker's choreography than simply to transpose it into another different cultural field in order to count as transformation—otherwise, as with the arguments for appropriation art as transformative—it would open the door to what scholar Darren Hudson Hick has called “copyright anarchy,” as presumably almost any secondary use would be prima facie transformative (Reference Hick2013, 127).

Yeoh is not the only author in the academic commentary on the “Countdown” controversy to express uncertainty about the legal and aesthetic status of Beyoncé's plagiarism. Kraut expresses a similarly ambivalent stance. On the one hand, she acknowledges that Beyoncé took the choreography without asking, but she, too, stops short of calling it a copyright infringement, despite her expertise on the subject. If I understand her correctly, Kraut's hesitation stems from the fact that she finds the race and the cultural milieu of the two artists to be a mitigating factor. As Kraut says,

Kraut does a masterful job of parsing the contradictions embedded in De Keersmaeker's public statements about the “Countdown” video, and of connecting them to the larger cultural and historical context in which the controversy occurs. She diagnoses De Keersmaeker as anxious about the transposition of her work from an elite white milieu to one that is “subtly coded black” (Reference Kraut2016, 274). But Kraut's reading also betrays some anxiety of her own. One senses Kraut's hesitation to label Beyoncé's plagiarism for what it is. Her use of the double negative in the quote above—“it is not unfair to say that Beyoncé erred”—comes across as concessive but retreats from a simple declaration that what Beyoncé did was wrong, either morally or legally. Kraut's unwillingness to label the appropriation as either plagiarism or infringement seems to stem not from concerns about the adequacy of the imitation, as with Yeoh, but from a sensitivity to the historical and cultural context in which it has taken place. In her book, she points out that traditionally it has been white dancers and choreographers who have arrogated to themselves the work of Black artists without permission. In that sense, the “Countdown” case is a fascinating instance of a reversal of the usual exploitative dynamic, for here it is the Black artist who takes from the white one without asking. As Kraut puts it, Beyoncé's appropriations “flip the script” (Reference Kraut2016, 265). One thing that has not changed, however, is that the artist with the greater material wealth and resources—in this case, Sony's legal team—has an advantage in fending off such suits. This is a primary reason why copyright protection for choreography, which is so often poorly remunerated compared to other art forms, is important. There is no doubt, as Kraut has argued convincingly in her published work, that racism is woven into the history of choreography and copyright in America. My interest in returning to the Beyoncé/De Keersmaeker controversy is not to deny this, but to shed light on the ways in which the barriers to supporting copyright protection for choreography often comes from the most unexpected of places: the artists and dance theorists who are the art form's greatest exponents and advocates.

To insist that a given performance does not count as a true instantiation of a choreographic work unless it is done correctly, and in accord with the artist's intentions, is a fitting tribute to the rich complexity and potential for expressive nuance that the dancing body possesses. It may seem that this tight form of control is empowering for choreographers, but it may in fact have the opposite effect: it diminishes the choreographer's ability to claim that a given secondary use is infringing, because in principle the work cannot be correctly copied without the artist's help. This confusion between legal and connoisseurship standards of similarity, in addition to the practical and cultural barriers to copyright litigation for dance works, makes it more likely that choreography will remain on the margins of copyright's statutory scope, rather than finding a place in its heartland.Footnote ³⁴