abstract. Copyright law’s experiment with transformative use is failing. So argue a growing number of scholars who contend that the standard conflicts with the goals of art. In their view, transformative use goes astray by conflating the accused work’s meaning with the defendant’s intent. However, they argue, art succeeds precisely because it rejects this “intentionalism.” By situating the viewer as the authority on the work’s meaning, art permits viewers to express values that would otherwise be suppressed. Consequently, opponents conclude, copyright law should replace transformative use with various anti-intentionalist strategies, similar to those deployed in art criticism.

Yet despite the prominence of anti-intentionalism in debates over transformative use, scholars have not critically examined anti-intentionalism’s application to copyright law. Instead, debates have proceeded in isolation from the broader dialectic between intentionalism and anti-intentionalism. This isolation is unfortunate. Outside of copyright law, scholars have responded to anti-intentionalism’s emergence by challenging anti-intentionalism’s premises and relaxing the assumptions that motivated anti-intentionalism’s development.

This Note argues that opponents of intentionalism are wrong to reject intentionalism outright. Transformative use’s inquiry into artistic intent protects the First Amendment principles embodied in fair use. Many First Amendment doctrines and theories require courts to consider the defendant’s intentions in determining her liability for the harmful consequences of her speech. At the same time, aesthetic and practical arguments against intentionalism are not compelling. None of the arguments offered by opponents of intentionalism would trouble a committed intentionalist, and appeals to uncertainty are exaggerated. Calls to reject transformative use based on intentionalism are therefore premature.

author. Yale Law School, J.D. 2016. Thanks to BJ Ard, Gregory Cui, Camilla Hrdy, Amy Kapczynski, Urja Mittal, Cobus van der Ven, Rebecca Wexler, and James Whitman for helpful comments and conversations. Thanks also to the editors of the Yale Law Journal for exceptional editing. All errors are mine.

“Considering the philosophic intelligence that has set out to discredit it, intentionalism in critical interpretation has shown an uncanny resilience.”

—Denis Dutton1

introduction

Do an artist’s intentions affect the meaning of her work? Should courts consider a defendant’s intentions when deciding whether an accused work is “transformative”? A growing number of copyright scholars argue that they should not. In their view, courts should abandon the transformative use standard, at least as currently conceived, because it relies on an “intentionalist” theory of interpretation that conflicts with the goals of contemporary art and copyright law. Under that standard, an otherwise infringing work is fair use if it “adds something new” to the copyrighted work, “with a further purpose or different character, altering [that work] with new expression, meaning, or message.”2 Since the Supreme Court adopted transformative use in Campbell v. Acuff-Rose Music, Inc., courts have fashioned the standard into a flexible instrument for protecting creative expression.3

Surprisingly, however, several scholars who otherwise support strong fair use protections have joined expansionists in calling for transformative use’s abandonment or modification. By equating transformativeness with “expression, meaning, or message,”4 they argue, transformative use assumes that there is an artist behind the work whose intentions determine the work’s artistic meaning. This assumption, opponents of intentionalism believe, is wrong. Following prominent artists and art critics,5 opponents argue that courts should embrace “anti-intentionalist” strategies, which focus on the viewer’s perceptions of the work, not the artist’s intentions.6 Although these arguments have their foundation in contemporary art theory, they apply broadly to all works of artistic expression.7

Arguments against intentionalism come in several forms. Strong versions of the anti-intentionalist critique hold that courts should never consider the defendant’s artistic intentions. For example, Amy Adler argues that courts should abandon the transformative use test and adopt a market-based analysis, drawing on critical theory for support.8 As she explains, “To search for meaning by relying on the author’s intent would be to ask precisely the wrong question, to miss that the author is dead and that the work is now living its own life.”9 Transformative use, she concludes, denies that life by encouraging courts to use the defendant’s intent “as a straightforward way to get at the question of transformativeness.”10

Weaker versions of the anti-intentionalist critique hold that courts should not consider the defendant’s intentions when viewers of the accused work would not do so. In particular, Laura Heymann argues that courts should “reconceptualize” the transformative use standard to shift the focus of the fair use inquiry to the viewer’s response.11 Because all art is “representational” in the sense that it is a “copy” of something else, she contends, fair use should not be concerned with “what an author does when she creates—whether the second author changes the first author’s expression in some ascertainable or substantial way—but rather whether the reader perceives an interpretive distance between one copy and another.”12 Adler and Heymann are not alone, and several other scholars have criticized transformative use on anti-intentionalist grounds.13

Despite growing criticism of intentionalism, courts have not adequately defended fair use doctrine’s reliance on artists’ intentions. Following Campbell, courts frequently ask whether the defendant intended to alter the copyrighted work’s “expression, meaning, or message.”14 If they find that the defendant’s intentions are transformative, courts generally hold that the accused work is fair use, even if that work does not actually alter the copyrighted work’s content.15 However, in reaching these conclusions, courts have not justified their reliance on intentionalism, instead appealing to Campbell’s formula for transformativeness. Indeed, even when courts have found transformativeness based on the defendant’s intentions alone, they have done so unreflectively, without considering the stakes of the debate between intentionalism and anti-intentionalism.16

Furthermore, recent decisions suggest that intentionalism’s hold on transformative use may be slipping. In Cariou v. Prince,17 the Second Circuit held that courts may properly ignore the defendant’s intentions if the content of the accused work is sufficiently transformative. In the court’s view, whether a work is transformative depends on “how the work in question appears to the reasonable observer, not simply what an artist might say about a particular piece or body of work.”18 Applying this rule, the court found that most of the accused works were fair use because of observable differences between the copyrighted and accused works.19

This Note makes two contributions to the debate over intentionalism in copyright law. First, this Note defends transformative use’s intentionalism against critics. So far, scholars and jurists have allowed opponents of intentionalism to advance their conclusions largely uncontested.20 Accordingly, this Note attempts to move copyright scholarship to a more balanced discussion of intentionalism’s strengths and weaknesses. Second, this Note describes several strategies that courts could adopt in deciding whether a particular work was transformative given the unsettled nature of the debate between intentionalism and anti-intentionalism. Although superficially similar to existing approaches, both strategies provide an organizing framework for making transformative use determinations that preserves reference to the defendant’s intentions while avoiding the full force of anti-intentionalist arguments.

This Note proceeds in three parts. Part I discusses modern fair use doctrine and the emergence of anti-intentionalism in copyright law. This discussion illustrates the motivations that animate anti-intentionalist proposals and provides a doctrinal platform for the analysis that follows. As this Part explains, the Second Circuit’s trilogy of appropriation art cases drives much of the opposition to transformative use’s intentionalism.21 Although the debate between intentionalism and anti-intentionalism transcends appropriation art, and even contemporary art, these cases remain influential, both theoretically and doctrinally.22

Part II advances three arguments in support of intentionalism. First, intentionalism is arguably more consistent with fair use’s status as a First Amendment accommodation. Many First Amendment doctrines require courts to consider the defendant’s intentions in determining her liability for the harmful effects of her speech. Likewise, many leading First Amendment theories, which view the First Amendment in terms of the speaker’s interests, support the use of intentionalist standards. Although competing theories provide weaker support for intentionalism, they have difficulty explaining the First Amendment’s application to art.

Second, aesthetic theory provides greater support for anti-intentionalism than opponents of intentionalism assume. Although opponents frequently contend that anti-intentionalism has won the debate, the reality is more complex. Since anti-intentionalism’s emergence, philosophers of aesthetics and scholars in related disciplines have responded by challenging anti-intentionalism’s premises. However, neither proponents nor opponents of intentionalism have critically examined these arguments. Instead, the debate over transformative use has proceeded in isolation from the broader dialectic between intentionalism and anti-intentionalism. Consequently, opponents of intentionalism treat unsophisticated anti-intentionalist arguments as decisive, when their status is much more uncertain.23

Third, intentionalism generates no more unpredictability than anti-intentionalist alternatives. Indeed, the proposals suggested by opponents of intentionalism, such as greater reliance on art markets and art critics to determine a work’s significance, are highly unreliable, and their use would likely not produce greater consistency in outcomes. At the same time, intentionalism may not be the true cause of unpredictability in transformative use analysis. The multifactor, all-or-nothing structure of the fair use defense provides a plausible alternative explanation for apparent inconsistences in transformative use’s application.

To be clear, the purpose of this discussion is not to establish that anti-intentionalism is invalid as an aesthetic theory or that fair use should exclusively focus on the defendant’s intentions. Indeed, as the strategies proposed in Part III admit, intentionalism and anti-intentionalism remain controversial within the philosophy of aesthetics and related disciplines. Rather, this Note argues that there are legally persuasive reasons for taking the defendant’s intentions into account in many cases and that these justifications should be integrated into fair use law.

Accordingly, Part III describes two strategies for preserving transformative use’s reference to the artist’s intentions while accommodating anti-intentionalist concerns, in accordance with the unsettled nature of the debate between intentionalism and anti-intentionalism. Under a “dual standards” strategy, courts would consider both whether the defendant intended to transform the copyrighted work and whether viewers perceived the accused work to be transformative. Under a “categorial intentionalist” strategy, courts would consider only the defendant’s intentions regarding how the accused work should be classified or approached, not her intentions regarding that work’s aesthetic meaning. While these strategies would produce different balances between artists’ and viewers’ interests, both would be superior to the alternatives offered by opponents of intentionalism.

To some, responding to anti-intentionalism’s aesthetic challenge to transformative use (as distinct from its legal and practical challenges) may seem unnecessary. On one popular view, law and art belong to separate domains, and their principles are not cross-transferable.24 If this view is correct, lawyers and artists have little to offer each other, both for the debate over transformative use and in general. Yet, although opponents of intentionalism misrepresent the strength of the anti-intentionalist position, they correctly recognize that law can learn from art.

As the case law demonstrates,25 judges who make fair use determinations consider the same factual questions as their non-legal counterparts. They inquire into the defendant’s intentions and, in some cases, make judgments about the aesthetic meaning of the accused work (as opposed to some distinct, legal meaning). Insisting that art and law remain separate, as some commentators suggest, perversely encourages judges to make aesthetic judgments in a concealed and unprincipled fashion.26 Further, the willingness of artists and viewers to comply with legal norms relies, in part, on their sense that the law is legitimate. Ignoring aesthetic concerns alienates those upon whom law depends.27 Thus, the view that law and art belong to separate domains is both inaccurate and perverse.

i. Transformative Use

The anti-intentionalist backlash against transformative use responds to recent trends in fair use doctrine. In particular, transformative use’s collision with contemporary art, especially appropriation art, drives much of the opposition. Because many contemporary artists maintain that their intentions are irrelevant to the meaning of their works, several scholars have argued that courts should not consider the defendant’s intentions in determining fair use.

This Part describes the development of the transformative use standard, and anti-intentionalism’s emergence within copyright law. Section I.A explains developments in the justification and structure of fair use. Section I.B describes courts’ current, unreflective approaches to transformative use and intentionalism. Section I.C summarizes the Second Circuit’s influential trilogy of appropriation art cases, which motivate many anti-intentionalist arguments.

A. Modern Fair Use Doctrine

Fair use is an affirmative defense to copyright infringement that permits limited use of a copyrighted work without permission from the owner. Section 107 of the Copyright Act establishes four factors for determining whether the defendant’s use qualifies for the defense: (1) “the purpose and character of the use,” (2) “the nature of the copyrighted work,” (3) “the amount and substantiality of the portion used in relation to the copyrighted work as a whole,” and (4) “the effect of the use upon the potential market for or value of the copyrighted work.”28 However, the Copyright Act provides little guidance for determining how these factors should be applied, leaving the task of operationalizing the test to courts.29

In recent decades, the Supreme Court has clarified several aspects of fair use doctrine. Two developments are particularly relevant to the debate over intentionalism. First, in Eldred v. Ashcroft, the Court reframed fair use as one of copyright’s First Amendment internal “accommodations.”30 In Eldred, the Court confronted a First Amendment challenge to the Copyright Term Extension Act,31 which increased the length of the copyright term from the life of the author plus fifty years to the life of the author plus seventy years. The plaintiffs argued that the Act should be subject to heightened scrutiny as a content-neutral speech restriction.32 Rejecting this argument, the Court explained that “copyright law contains built-in First Amendment accommodations,” which diminish the need for heightened review.33 Among these accommodations, the fair use defense “allows the public to use not only facts and ideas contained in a copyrighted work, but also expression itself in certain circumstances.”34 Accordingly, fair use “affords considerable ‘latitude for scholarship and comment’” and other expressive uses of copyrighted works.35

Although Eldred purported to follow precedent, the Court effectively revised fair use’s relationship to the First Amendment and free expression. Before Eldred, the Supreme Court’s decisions described fair use in predominantly economic terms. For example, Harper & Row, Publishers, Inc. v. Nation Enterprises,36 which introduced the internal-accommodations approach to reconciling the Copyright Act with the First Amendment, described fair use as a market-corrective mechanism.37 According to the Court, fair use was a “necessary incident of the constitutional policy of promoting the progress of science and the useful arts, since a prohibition of such use would inhibit subsequent writers from attempting to improve upon prior works and thus . . . frustrate the very ends sought to be attained.”38 Although before Harper & Row, some scholars had speculated that fair use might perform a First Amendment function,39 the Court primarily attributed this role to the idea-expression dichotomy.40 Although Eldred purported to follow Harper & Row’s understanding of fair use,41 its description of the doctrine as a First Amendment accommodation was an innovative departure from existing law.42

Second, the Supreme Court embraced “transformative use,” which emphasizes the fair use test’s first factor, as the predominant standard for determining fair use. The transformative use standard, proposed by Judge Pierre Leval43 and adopted by the Supreme Court in Campbell,44 asks whether the accused work is “transformative” in the sense that it does not “merely ‘supersede[] the objects’” of the original work, but instead “adds something new, with a further purpose or different character, altering the first with new expression, meaning, or message.”45 Although transformative use is “not absolutely necessary for a finding of fair use,” it lies “at the heart of the fair use doctrine’s guarantee of breathing space within the confines of copyright.”46 Consequently, “the more transformative the new work, the less will be the significance of other factors, like commercialism, that may weigh against a finding of fair use.”47

Since its adoption, the transformative use standard has come to dominate fair use analysis. Indeed, fair use determinations almost always turn on whether the defendant’s use was transformative. In a survey of all copyright decisions reported between 2006 and 2010, Neil Netanel finds that 85.5 percent of all district court opinions and 93.75 percent of all appellate opinions applied the transformative use standard in deciding whether the accused work was fair use.48 The only appellate-court decision not to apply the transformative use standard was an unpublished opinion in which the defendant “relied exclusively on the fourth factor in its brief.”49 Netanel’s findings update earlier studies, which document a significant but smaller shift toward transformative use before 2006.50

Transformative use’s dominance is readily apparent from courts’ willingness to overlook the other fair use factors once they have held that a work is transformative. For example, the Second and Fourth Circuits have explained that the second factor, which considers the nature of the copyrighted work, “may be of limited usefulness where the creative work of art is being used for a transformative purpose.”51 Likewise, in Author’s Guild, Inc. v. HathiTrust,52 the Second Circuit held that the significance of the third factor, like the second factor, depends on the extent to which the accused work is transformative. “For some purposes, it may be necessary to copy the entire copyrighted work, in which case [the amount and substantiality of the portion used] does not weigh against a finding of fair use.”53 And in Perfect 10, Inc. v. Amazon.com, Inc.,54 the Ninth Circuit found, in addressing the fourth factor, which considers the accused work’s effect on the market for the copyrighted work, that “Google’s use of thumbnails for search engine purposes is highly transformative, and so market harm cannot be presumed.”55

Attitudes over transformative use have shifted dramatically with its development. Over time, many commentators have recognized that the standard provides a flexible, predictable, and robust defense, protecting even exact copies.56 Recently, copyright expansionists have begun to criticize the transformative use standard as uncertain, even though they once relied on the standard’s purported robustness to argue in favor of further strengthening copyright protections.57 Copyright restrictionists, for their part, have executed a similar about-face.58 Given these developments, the recent opposition to transformative use is all the more surprising.

B. Fair Use and Intentionalism

Despite transformative use’s success, neither the Supreme Court nor lower courts have offered a defense of transformative use’s intentionalism. “Transformativeness” is an ambiguous concept, capable of being understood in one of two ways. Courts may determine “transformativeness” either by measuring differences between the defendant’s and copyright holder’s intentions or by measuring differences in the accused and copyrighted works’ content.

Generally, transformative use determinations turn on intentions, rather than content. For example, Anthony Reese concludes that courts are much more likely to find the defendant’s use transformative based on her intentions rather than the accused work’s content.59 This conclusion holds across jurisdictions and categories of art.60 Indeed, Reese finds, “in all of the cases where transformativeness was found based on the defendant’s transformative purpose, the opinion’s ultimate conclusion was that the use was, or was likely to be, fair.”61

Significantly, courts generally hold that the defendant’s use is transformative if she intends to transform the copyrighted work, even if she does not alter that work’s content.62 For example, in Castle Rock Entertainment, Inc. v. Carol Publishing Group, Inc.,63 the Second Circuit concluded that a defendant’s work “need not necessarily transform the original work’s expression to have a transformative purpose,”64 so long as the defendant intends to transform the copyrighted work.65

Nevertheless, courts have not attempted to defend transformative use’s elevation of the defendant’s intentions over the accused work’s content. In Castle Rock Entertainment and similar cases, courts have treated the relative importance of the defendant’s intentions as the necessary consequence of fair use principles.66 Yet, that assumption is not obvious.67 In theory, either the defendant’s intentions or the accused work’s content could supply the basis for a finding of transformativeness. As Parts II and III demonstrate, anti-intentionalist arguments against transformative use challenge that assumption. Furthermore, recent developments suggest that as these arguments gain traction, intentionalism’s hold on transformative use may be slipping.68

C. Transformative Use Meets Appropriation Art

Recently, a growing number of scholars who otherwise oppose copyright’s expansion, have joined alongside expansionists in calling for transformative use’s abandonment or modification.69 Transformative use’s collision with contemporary art, particularly appropriation art, drives much of this opposition. “Appropriation” describes the process whereby artists copy elements from existing works, such as images from popular culture, and assimilate those elements into their own works.70 In doing so, appropriation artists follow a longstanding tradition of artistic borrowing. As Baudelaire noted, artistic traditions often progress through a sequence of copying. Delacroix copied Rubens; Manet copied Delacroix; and Van Gogh, Millet.71 Baudelaire, for his part, copied Delacroix and Hugo.72

Because appropriation, by definition, involves copying, appropriation artists often violate copyright law’s reproduction and derivative works rights, which grant the copyright holder the exclusive right to copy or adapt the copyrighted work.73 More generally, appropriation art challenges copyright law’s protection of creativity and originality. Indeed, given copyright law’s “insistence” on these values, one scholar has argued that appropriation art presents “the most radical challenge to the copyright laws to date.”74

The Second Circuit’s influential trilogy of appropriation art cases—Rogers v. Koons, Blanch v. Koons, and Cariou v. Prince—motivates much of the debate over transformative use’s intentionalism.75 The first two cases involved Jeff Koons, an American artist whose appropriation of lesser-known artists’ photographs caused repeated controversy.

In Rogers, the Second Circuit held that Koons’s sculpture String of Puppies was not fair use.76 Koons modeled his sculpture on Art Rogers’s photograph Puppies, which depicted a married couple and their eight German Shepherds.77 Rogers sued after learning of the infringement.78 At trial, Koons testified that the sculpture was “a satire or parody of society,” which qualified as fair use.79 In support, Koons alleged that he belonged to a school of artists who believed that “the mass production of commodities and media images has caused a deterioration in the quality of society.”80 The Second Circuit rejected Koons’s fair use defense. To receive protection, the court held, a parody or satire must comment, at least in part, on the copyrighted work; otherwise, “there would be no real limitation on the copier’s use of another’s copyrighted work to make a statement on some aspect of society at large.”81

Despite Koons’s defeat in Rogers and two similar cases,82 he continued to incorporate other artists’ photographs into his works. In Blanch, the Second Circuit considered Koons’s painting Niagara, which depicted “four pairs of women’s feet and lower legs dangling prominently over images of confections.”83 To create the painting, Koons had copied one of the pairs of legs from Andrea Blanch’s photograph, Silk Sandals by Gucci, which was featured in an issue of Allure, a women’s beauty magazine.84

This time, the Second Circuit held that Koons’s painting was fair use.85 Applying the transformative use test, adopted after Rogers,86 the Second Circuit explained that Koons’s “purposes in using Blanch’s image [we]re sharply different from Blanch’s goals in creating it.”87 Whereas Koons testified that he wanted the viewer “to think about his/her personal experience” with the objects present in the painting and how they “affect our lives,” Blanch “wanted to show some sort of erotic sense” and “sexuality.”88 Although Koons’s critical aims transcended the appropriated work, the court held that “Koons had a genuine creative rationale for borrowing Blanch’s image,” because he considered their inclusion “necessary” to “comment on the culture and attitudes embodied in Allure Magazine.”89

The third case involved a different defendant, Richard Prince, but similar questions. In Cariou, the Second Circuit held that most of the paintings and collages in Prince’s collection Canal Zone were fair use.90 Of the thirty-one works, thirty incorporated photographs from Patrick Cariou’s book, Yes Rasta, which contained photographs of Cariou’s experiences with the Rastafarians in Jamaica.91 Unlike Koons, Prince did not testify that the works were intended as a parody or satire of Cariou’s photographs or the values that they embodied.92 Instead, Prince testified that he had “no interest in the original meaning of the photographs he uses” and did not “really have a message” that he intended to communicate in making art.”93 Accordingly, the district court concluded, Prince’s “own testimony show[ed] that his intent was not transformative.”94

The Second Circuit reversed, holding that the absence of transformative intentions did not foreclose Prince’s fair-use defense. Instead, the court explained that whether fair use applies depends on “how the work in question appears to the reasonable observer, not simply what an artist might say.”95 Applying this standard, the court found that “observation of Prince’s artworks themselves” demonstrated that twenty-five of Prince’s works were transformative.96 In the court’s view, those twenty-five works “manifest an entirely different aesthetic from Cariou’s photographs. Where Cariou’s serene and deliberately composed portraits and landscape photographs depict the natural beauty of Rastafarians and their surrounding environs, Prince’s crude and jarring works, on the other hand, are hectic and provocative.”97

Scholarly commentary on these cases is mixed. While many scholars celebrate courts’ increasing sympathy toward appropriation art, exemplified by the Second Circuit’s decisions in Blanch and Cariou,98 others argue that the trilogy illustrates the problems inherent in transformative use’s reliance on the defendant’s intentions.99 According to this vocal group, transformative use’s intentionalism is perverse. In their view, the defendant’s intentions are irrelevant to the accused work’s meaning100 and a court’s reliance on her testimony yields highly uncertain outcomes.101 Instead, they argue, copyright law, like art criticism, should abandon transformative use and refocus the fair use inquiry on viewers’ responses.102 The remainder of this Note critically examines these arguments.

ii. The Anti-Intentionalist Critique

Copyright scholars are wrong to reject transformative use’s intentionalism outright, at least given the arguments presented thus far. Outside of copyright law, intentionalism’s status remains controversial. Scholars have responded to anti-intentionalism’s emergence by challenging anti-intentionalism’s premises and relaxing the assumptions that motivated its development.

This Part describes the debate between intentionalism and anti-intentionalism outside of copyright law and evaluates the arguments for and against intentionalism in fair use. Section II.A provides a more accurate account of the debate between intentionalism and anti-intentionalism than those offered by opponents. Section II.B defends transformative use’s intentionalism.

A. The Intentionalism/Anti-Intentionalism Debate

Scholars outside of copyright law sharply disagree over the relevance and significance of an artist’s intentions to artistic meaning. Broadly, scholars adopt one of two positions. Intentionalists argue that a work’s meaning is identical with or constrained by the artist’s intent; anti-intentionalists deny this conclusion.103 Within the Anglo-American tradition, New Criticism provided an early endorsement of the anti-intentionalist thesis. In 1946, William Wimsatt and Monroe Beardsley provided an influential description of anti-intentionalism in literary studies. Criticizing what they termed “the intentional fallacy,” Wimsatt and Beardsley argued that the “design or intention of the author is neither available nor desirable as a standard for judging the success of a work of literary art.”104 Because the author’s intentions were unknowable to the reader, they contended, criticism should proceed by concentrating on the literary work itself, through close reading and similar techniques.105

During the 1960s and 1970s, reader response theory replaced New Criticism as the dominant critical approach. Like the New Critics, reader response theorists argued that the author’s intent was irrelevant to the work’s meaning. However, unlike the New Critics, they argued that interpretation should proceed by analyzing the reader’s affective responses.106

Simultaneously, European structuralist and post-structuralist theorists developed their own critiques of intentionalism and authorship. In 1967, Roland Barthes announced the “death of the Author” and the “birth of the reader.”107 According to him, modern literature had transcended, or soon would transcend, the limitations of authorship by shifting the focus of interpretation to a depersonalized and dehistoricized reader.108 Two years later, Michel Foucault published his own analysis of authorship, arguing that the author was not a person but a “function” that emerged during the history of literature.109 In the same vein, Jacques Derrida and Paul de Man offered deconstructive theories of meaning, which rejected inquiry into the actual author’s intent. Instead, they believed, the radical indeterminacy of language prevented works from communicating anything other than the impossibility of communication.110 Collectively, these approaches continue to exert significant influence on contemporary criticism.

But anti-intentionalism never replaced intentionalism. Indeed, in the philosophy of aesthetics, intentionalism rebounded. In 1968, E.D. Hirsch offered an early rejoinder to the New Critics and their European counterparts. According to Hirsch, the impossibility of ascertaining the author’s intent with absolute certainty did not prevent readers from ascertaining the author’s intent with sufficient certainty to ground a coherent interpretive program.111 Further, Hirsch argued, anti-intentionalist theories were incoherent because without reference to the author’s intent, readers would have no reason to prefer one textually plausible interpretation of the author’s work to another.112

Since Hirsch’s response, both intentionalists and anti-intentionalists have retreated from extreme versions of their respective theories. Modern intentionalists embrace a wide range of positions. Following Hirsch, “actual intentionalists” maintain that the actual author’s intentions and the work’s meaning are logically equivalent; but many accept that those intentions may fail.113 “Hypothetical intentionalists” hold that the work’s meaning is the best hypothesis about the author’s intentions.114 And “fictional intentionalists” propose that the work’s meaning is the meaning intended by a fictional, artistically relevant author.115 At the same time, many modern anti-intentionalists accept that authors’ intentions may be relevant under some circumstances.116

Outside of philosophy, feminists and critical theorists have defended authorship against structuralist and post-structuralist critics. Nancy Miller, for example, argues that Barthes’s “decision that the Author is dead, and subjective agency along with him, does not necessarily work for women and prematurely forecloses the question of identity for them.”117 According to her, many instances of feminist literature succeed precisely because they speak from a position of authority and subjectivity.118 While Miller’s view is by no means universally held,119 authorship continues to hold a place within feminist and critical theory.

B. Judging Anti-Intentionalism

Opponents of intentionalism in fair use ignore the complexities of modern intentionalist and anti-intentionalist debate.120 Opponents offer political, aesthetic, and practical arguments. But across each dimension, intentionalism is at least as viable as anti-intentionalism, and in some cases, arguably superior. Thus, jurists and scholars should not abandon transformative use’s intentionalism, at least on the basis of the arguments raised thus far.

1. Free Expression

Despite Eldred and Golan’s reframing of fair use as a First Amendment accommodation, copyright scholars have devoted little attention to how First Amendment doctrine and theory ought to influence debates internal to copyright law. Instead, scholarship has concentrated on reconciling the perceived tensions between the First Amendment’s free speech guarantee and the Copyright Act’s exclusive rights provision.121

Regardless of whether the First Amendment and Copyright Act can ultimately be reconciled, fair use’s relationship to the First Amendment suggests that, as a legal matter, copyright law ought to err on the side of granting protection based on the defendant’s intentions. Intentionalism resonates strongly with many First Amendment doctrines, which require courts to consider the defendant’s intentions in determining her liability for the harmful effects of her speech. It also resonates with many leading First Amendment theories, which view the First Amendment’s purpose as furthering individual autonomy or democratic participation or culture, broadly understood. Although competing theories provide weaker support for intentionalism, they have difficulty explaining the First Amendment’s application to art.

Opponents of intentionalism often assume that anti-intentionalism provides greater protection for defendants. For example, Laura Heymann argues that whereas “a focus on the defendant’s purpose yields a conclusion that the copyrighted work has not been ‘transformed’ in the physical or legal sense, . . . a focus on reader response may well yield the opposite conclusion.”122 But Heymann ignores the opposite case, where a focus on the defendant’s intentions would result in a finding of transformativeness, but a finding on the viewer’s response would not. In those circumstances, anti-intentionalist standards create potential First Amendment problems because they favor punishing the defendant for the harmful effects of her speech, even though she intends for her speech to have quite different results.

To see this, consider the following example. Jeff, an appropriation artist, copies a political advertisement, intending to parody the featured candidate. Unfortunately for him, viewers do not apprehend Jeff’s intentions and treat the parody as a reproduction. At trial for copyright infringement, Jeff raises a fair use defense, arguing that his work is transformative. As evidence, he attempts to introduce his artist’s notes, which document his reasons for copying the advertisement. Should the court consider this evidence? Should Jeff’s defense fail as a matter of law?123

Next, consider a similar example. This time, viewers know that Jeff intends for his work to function as a parody, but disregard Jeff’s intent. Perhaps they believe Jeff is a lousy artist. Or perhaps they care more about Jeff’s work than his political beliefs.124 Should Jeff’s fair use defense fail in this case too?

First Amendment doctrine suggests that a defendant’s intentions ought to determine her liability, at least much of the time. However, the extent to which this is the case turns on the conventional distinction between coverage and protection.125 Coverage doctrines govern whether an activity counts as “speech,” to which the First Amendment applies, or else counts as “non-speech.” In contrast, protection doctrines govern whether the First Amendment prohibits a particular government regulation. Given an alleged First Amendment violation, courts first determine whether the First Amendment applies, using coverage doctrines. If the First Amendment is found to apply, courts proceed to evaluate the government regulation at issue, using protection doctrines.126

In distinguishing between particular categories of “speech” and “non-speech,” coverage doctrines place particular importance on the defendant’s intentions. For example, the First Amendment prevents the government from penalizing incitement—that is, activities that tend to promote violence or lawlessness—unless the speaker actually intends to incite “imminent lawless action and is likely to incite or produce such action.”127 Similarly, the government may not penalize defamation against public officials unless the speaker knew that her statements were false or acted in reckless disregard of the truth.128 And on one plausible reading of the Supreme Court’s “true threats” jurisprudence, the government may not penalize threats unless the speaker intended to intimidate.129

Similar rules apply to the boundary between “speech” and non-speech “conduct.” Whether a non-verbal activity, such as displaying a symbol, is sufficiently “communicative” to qualify for First Amendment coverage depends on whether the speaker intended “to convey a particularized message” and whether “the likelihood was great that the message would be understood by those who viewed it.”130 To meet this standard, the “message” need not be “narrow” or “succinctly articulable,” and may include intentional but non-representational messages, such as those embodied in abstract paintings or music.131

Furthermore, the Supreme Court has narrowed the reach of coverage doctrines that do not consider a speaker’s intentions. For example, the definition of fighting words—“personally abusive epithets which, when addressed to the ordinary citizen, are, as a matter of common knowledge, inherently likely to provoke violent reaction”—contains no reference to the speaker’s intentions.132 However, over time, the Court has narrowed that definition by characterizing certain actions as incitement, rather than fighting words,133 and by excluding speech that is “not ‘directed to the person of the hearer.’”134 The Court has also invalidated numerous fighting words statutes through aggressive enforcement of the overbreadth doctrine, which prohibits the government from enacting or enforcing regulations that regulate speech and non-speech.135

In similar fashion, the Supreme Court has relied upon the First Amendment to effectively integrate mental-state requirements into obscenity law. Although the definition of obscenity, like that of fighting words, does not refer to the speaker’s intentions,136 the Court has invalidated statutes that criminalize the distribution of obscene materials but do not require the defendant to have acted willfully or knowingly.137 It has applied similar rules to statutes that criminalize the distribution of child pornography.138

More generally, the Supreme Court’s vagueness doctrine infuses a constructive knowledge requirement across First Amendment law by ensuring that the defendant has fair notice of the government regulation at issue.139 As the Court has explained, the doctrine rests upon the assumption that individuals are “free to steer between lawful and unlawful conduct.”140 Unless “laws give the person of ordinary intelligence a reasonable opportunity to know what is prohibited,” they may “trap the innocent,” who do not know, or at least have no reason to know, what the laws require of them.141

In contrast with coverage doctrines, protection doctrines are essentially concerned with the strength of the government’s justification and its connection to the regulated activity. Content-based regulations generally must satisfy strict scrutiny,142 whereas content-neutral regulations need only satisfy some variation of intermediate scrutiny.143 For example, the government may reasonably regulate the time, place, and manner of protected speech, provided that such regulations “are justified without reference to the content of the regulated speech, that they are narrowly tailored to serve a significant governmental interest, and that they leave open ample alternative channels for communication of the information.”144 Regardless of the level of scrutiny, the government’s means and ends, rather than the speaker’s intentions, are the focus of analysis.

Fair use involves questions of both coverage and protection. In determining whether the accused work is fair use, the court decides both whether the defendant’s use might merit protection and, assuming that it does, whether the government’s justification for imposing copyright liability, that is, furthering the constitutional policy of promoting the progress of science and the useful arts,145 is strong enough to justify subjecting the particular defendant to liability. However, different fair use factors capture different aspects of this analysis.

The first factor—“the purpose and character of the use”146—most strongly resembles coverage doctrines. Like other coverage doctrines, the first factor attempts to categorize the defendant’s activity relying, in part, on her intentions. Under that factor, the court attempts to determine whether the defendant’s use belongs to the category of transformative or non-transformative uses.147 If the court finds that the defendant’s use is transformative, the court considers whether the other three fair use factors weigh against imposing liability.148

Those factors—“the nature of the copyrighted work,” “the amount and substantiality of the portion used in relation to the copyrighted work as a whole,” and “the effect of the use upon the potential market for or value of the copyrighted work”149—bear closer resemblance to protection doctrines. Unlike the first factor, the latter three factors are more concerned with the government’s justification than the defendant’s intentions. The more commercially valuable the copyrighted work, the greater the portion used, and the greater the effect of that use on the market for the copyrighted work, the greater that interest.150

This framework, which separates the fair use factors into coverage and protection doctrines, suggests that a defendant’s intentions ought to be relevant in determining whether the purpose and character of her work weighs against imposing liability. Like other coverage doctrines, the transformative use standard asks whether the defendant intended to cause the harmful effects that copyright law attempts to guard against. To the extent that the defendant instead intended to transform the copyright holder’s work, that intention is lacking.151

Leading First Amendment theories also support the conclusion that the defendant’s intentions should be relevant in determining her liability. Modern First Amendment theories, including autonomy, legitimation, and cultural theories, share a common concern for speakers rather than speech. That is, they emphasize the role of the speaker’s interests in determining whether the First Amendment ought to protect her actions against government regulation.152

Autonomy theories view the First Amendment’s value in terms of self-expression or self-realization.153 As Leslie Kendrick has argued, such theories have a strong affinity with intentionalism because imposing liability on speakers based on listeners’ perceptions undermines speakers’ autonomy interests in free and open communication.154

Put another way, objective standards penalize “speakers who do not intend harm and who are reasonably unaware of the harmful aspects of their speech. Such speakers have no relation to the aspect of their speech for which they are being penalized. They also have personal communicative aims that . . . are deserving of respect.”155 Accordingly, speakers and listeners “generally have a claim against purposeful interference” with speech, unless “interference is otherwise justified” by an overriding public interest.156 Conversely, “speakers who intend to cause harm forfeit a claim against purposeful interference” by virtue of that intention.157 But “where purposeful interference is otherwise justified, speakers still have a claim not to be held strictly liable for their messages.”158

The autonomy-based account offers a normative foundation for the doctrinal principles described above. It explains why, for example, governments may not penalize inflammatory speech, unless that speech is actually likely to cause incitement,159 or libel, unless the speaker acts intentionally or recklessly with regard to the truth.160 It also explains why transformative use ought to respect the defendant’s intentions. Imposing liability on the defendant based on viewers’ perceptions alone would violate the defendant’s autonomy to determine the significance of her actions. Conversely, imposing liability on a defendant who intends to create a competing product, instead of transforming the original work, would not violate the defendant’s autonomy. Such a defendant “forfeits” her claim against purposeful interference with her speech. Further, if the defendant intends to create an infringing work for purely commercial reasons, the need to protect personal communicative aims is much reduced.

Kendrick’s account also illustrates intentionalism’s flexibility. For example, intentionalism need not be limited to instances in which a defendant acts “intentionally,” as that term is used in criminal law.161 “It is possible . . . that where speech poses a real and serious harm, and the speaker’s mental state encompasses that risk, the conditions for purposeful interference have been met.”162 Thus, intentionalism can embrace instances in which the defendant acts knowingly or recklessly as to the consequences of her actions. Accordingly, intentionalism might permit liability for repeat offenders, such as Koons, who have been notified that their form of expression constitutes infringement. By contrast, instances in which the defendant merely acts negligently would not be subject to liability because notice would not be present.

Of course, the defendant’s autonomy interest in avoiding government interference with her speech must be weighed against competing autonomy interests, such as a viewer’s interest in independently interpreting the accused work163 or the copyright owner’s interest in preserving the integrity of her work from unauthorized use.164 However, to the extent that the defendant’s autonomy interests conflict with others’, fair use arguably ought to treat the former as legally decisive. In copyright infringement cases, defendants, viewers, and copyright holders do not occupy equivalent positions. If the court decides that the accused work is not fair use, the defendant may be subject to coercive punishment; and the accused work, destroyed.165 But if the court reaches the opposite conclusion, viewers remain free to attribute to the accused work any meaning or significance they choose, unconstrained by the court’s judgment. Likewise, the existence of the copyrighted work remains unaffected.166 Moreover, imposing copyright liability on the defendant may discourage other artists from creating works in the future. Thus, treating defendants’ autonomy interests as decisive may, at least in some cases, further viewers’ long-term autonomy interests, even if it seems to undermine their short-term ones.

Legitimation theories support intentionalism for similar reasons. Unlike autonomy theories, legitimation theories hold that the First Amendment protects speakers’ rights to “experience the value of self-government” by “participating in the formation of public opinion” within the political sphere.167 Yet, like autonomy theories, legitimation theories emphasize the speakers’ interests in speaking, rather than the effects of her speech. As Robert Post explains, democracy requires “a certain relationship between persons and their government. Democracy is achieved when those who are subject to law believe that they are also potential authors of law.”168 Because citizens must “have access to the public sphere so that they can participate in the formation of public opinion,” legitimation theories treat speakers “as autonomous in many of the ways that autonomy theory would predict.”169

Consequently, legitimation theories suggest that fair use ought to consider the defendant’s intentions to the extent that she attempts to contribute to public discourse. Much of copyright’s subject matter would meet that requirement, even if it were not narrowly political in the traditional sense. As Post explains, “Art and other forms of no[n]cognitive, nonpolitical speech fit comfortably within the scope of public discourse” because they belong to a sociological category that we recognize as strong contributors to public opinion.170 Given this, fair use ought to consider the defendant’s intentions in most cases. Banishing the defendant’s intentions from the fair use inquiry would undermine the legitimating function that the First Amendment serves, preventing the defendant from experiencing what she perceives to be democratic participation, even if that perception were not shared by viewers.171

Cultural theories yield similar conclusions. According to Jack Balkin, the First Amendment protects and promotes a democratic culture “in which individuals have a fair opportunity to participate in the forms of meaning-making that constitute them as individuals.”172 Such meaning-making includes popular art and other forms of expression traditionally considered outside of the domain of politics.173 As with autonomy and legitimation theories, cultural theories provide strong support for intentionalism. Anti-intentionalist standards deny defendants the opportunity to engage in meaning-making by insisting that they accept the meanings proposed by copyright owners and viewers.174

Not all First Amendment theories support intentionalism. Following Alexander Meiklejohn, self-governance theories view the First Amendment as embodying the principle that citizens “shall govern themselves” by ensuring that “citizens shall, so far as possible, understand the issues which bear on common life.”175 Because self-governance theories view speech instrumentally, as a means for promoting traditional democratic institutions, they emphasize the public effects of speech, rather than speakers’ reasons for speaking. Similarly, marketplace theories view the First Amendment as an epistemological device for establishing the truth and falsity of competing ideas.176 Like self-governance theories, marketplace theories provide little inherent support for intentionalism because they emphasize the effects of speech, potentially at the expense of speakers.

However, self-governance and marketplace theories face severe explanatory difficulties when applied to fair use and copyrightable subject matter. In particular, they are difficult to square with the Supreme Court’s conclusion that the First Amendment “unquestionably” covers art.177 In declaring that the First Amendment covers even art without a “succinctly articulable message,”178 the Court has apparently placed little importance on whether a particular work promotes democracy or truth.179

Anticipating these difficulties, self-governance theorists have attempted to ground First Amendment coverage of art within traditional democratic principles. For example, Meiklejohn argued that self-governance theories were consistent with First Amendment coverage for art because, he believed, literature and other art forms encouraged voters to develop a “sensitive and informed appreciation and response to the values out of which the riches of the general welfare are created.”180 However, attempts to reconcile self-governance theories with the First Amendment’s coverage of art have failed to persuade many modern scholars.

In particular, Meiklejohn’s account runs into two related obstacles. First, the premise that art, as a category, encourages sensitivity to democratic values is highly dubious. As Richard Posner has forcefully argued, art and politics serve different masters. Some works teach democratic values; others, the opposite. Still others have no relationship to democratic values at all. Hence, examples of illiberal aesthetes and boorish liberals abound.181 Second, and relatedly, art’s encouragement of democratic values does not appear to track our intuitions about which works ought to be protected. Some works are profoundly undemocratic, yet neither the Supreme Court182 nor commentators183 have been willing to exclude them from First Amendment coverage on that basis.

Marketplace theories fail for similar reasons. Many works, especially those produced by contemporary artists, do not attempt to express truth, at least in the sense envisioned by marketplace theorists.184 And to the extent that works do express truth, their entitlement to protection does not seem to turn on that fact. Thus, although self-governance and marketplace theories oppose intentionalist standards because of their emphasis on speech, they are less compelling than the intentionalist-friendly theories discussed above, at least in the context of copyright.

2. Aesthetics

Aside from their free expression concerns, opponents of intentionalism also argue that inquiry into the defendant’s intentions undermines the aesthetic goals of contemporary art. For example, opponents argue that an artist’s intentions are irrelevant to interpretation because many contemporary artworks have no artist or many artists. That is, many contemporary artworks emerge from the independent creative activities of a multiplicity of artists, critics, viewers, and others, none of whom has an exclusive claim to the meaning of those works.185 Moreover, even when individual artists do exercise significant control over the work, they may be “unaware” of their intentions or create works by “accident” or while “unconscious.”186 Given this, they conclude, an artist’s intentions provide only an incomplete, and fallible, basis for interpretation.187 However, each of these arguments is dubious, especially when applied to transformative use.

The problem of “multiple authorship” is not unique to fair use. Courts commonly confront situations in which multiple actors participate in the joint commission of an offense, even though some actors may not perform all the elements of that offense.188 In the same vein, legal and philosophical accounts of “joint authorship,” which describe situations in which multiple actors contribute to the creation of individual works, might permit courts to impose liability on artists based on their contributions to the collective creative process.189 Although individual actors may have incomplete or conflicting intentions, the process of collaboration requires them to harmonize their actions in some fashion. From this harmonization, interpreters could infer a common purpose and impute that purpose to individual actors.190

Within copyright law, inferring individual intentions from collective actions is a familiar exercise. For example, as Shyamkrishna Balganesh has noted, courts determine whether parties are coauthors by analyzing their observable behavior.191 Specifically, to determine coauthorship in the absence of an express agreement, “courts began looking to the very process of collaboration and the parties’ behavior therein (as it unfolds during the collaboration) in order to treat the parties as coauthors of the work, based entirely on their actions.”192 Based on their observations, courts may treat parties as coauthors even though no formal agreement governs their collaboration.193 In a similar fashion, courts could make intentionalist judgments about transformativeness despite the collaborative, and even spontaneous, character of contemporary art.194

Additionally, appeals to the incompleteness or fallibility of artists’ intentions do not adequately differentiate between different kinds of intentions. As Jerold Levinson’s theory of “semantic” and “categorial” intentions illustrates, some intentions are much more likely to be present, and much less likely to fail, than others. Semantic intentions are the artist’s intentions “to mean something in or by” a particular work, while categorial intentions are her intentions for the work “to be classified or taken in some specific or general way.”195 That is, categorial intentions “involve the maker’s conception of what he has produced and what it is for, on a rather basic level; they govern not what a work is to mean but how it is to be fundamentally conceived or approached.”196

Unlike semantic intentions, categorial intentions are almost always present and almost never fail. For example, an artist may not intend for viewers to perceive a particular message in a given work, or if she does, may fail to achieve that intention. But she will almost certainly intend for the work to be perceived as an example of a particular genre or, at least, as art, and that intention will almost certainly succeed.197 Given this, Levinson concludes, interpreters can rely on categorial intentions to “determine how a text is to be conceptualized and approached on a fundamental level and thus indirectly affect what it will resultingly say or express.”198

Accordingly, even if a defendant has no semantic intentions with respect to the accused work—that is, she does not intend for viewers to perceive a particular message in that work—she will likely have categorial intentions relevant to its transformativeness. For example, the defendant may intend for the work to be perceived as belonging a particular genre or to be displayed in a particular location or to a particular community. Thus, even if opponents of intentionalism are correct in arguing that artists’ semantic intentions are incomplete or fallible, this would not prevent courts from judging the categorial intentions that are present and realized.

Furthermore, even if artists’ intentions were systematically fallible, courts could still coherently analyze them in determining transformativeness. Opponents of intentionalism assume that the standard requires courts to judge only those intentions that are realized in the accused work. But that assumption is questionable. By its own terms, the standard focuses on “uses” rather than “works.”199 That is, courts sometimes terminate the fair use inquiry after finding that the defendant’s intended meaning was transformative, without further determining whether the accused work actually has that meaning.200 Thus, the fallibility of artists’ intentions does not prevent their use in determining transformativeness.201

Notwithstanding this, it is unlikely that artists actually are completely unaware of their intentions. Anti-intentionalists commonly propose Surrealism’s “automatic writing” as an example of an utterly unintentional creative activity.202 However, as Paisley Livingston notes, “it is rather far-fetched to assume that these persons were entirely successful in eliminating all intentions from the process of writing, somehow scribbling for hours and days on end in a somnabulistic and totally unreflective trance.”203 Indeed, automatic writers purposefully experimented with different techniques, and many of the resulting works were heavily edited before publication.204 The same is likely true of artists. Notwithstanding dramatic claims about the unconscious nature of “the creative act,”205 it is highly questionable that artists genuinely have no attitudes relevant to the reception of their works at the time of creation.

To the extent that opponents of intentionalism maintain that intentionalism is invalid because interpreters cannot know artists’ intentions with absolute certainty, they place an extraordinarily, and perhaps unreasonably, high epistemological burden on interpretation.206 In many contexts, including ordinary communication, interpreters make judgments about agents’ intentions even though they cannot access their mental states directly. Yet anti-intentionalists do not raise the same skepticism about these activities.207 Indeed, many anti-intentionalist arguments—such as the claim that artists are unable to know their unconscious attitudes—apply to viewers as much as artists. Opponents of intentionalism do not convincingly explain why shifting the focus of interpretation from artists to viewers would escape these problems.208

Finally, some opponents of fair use concede that the debate between intentionalism and anti-intentionalism is unsettled, but argue that courts should reject intentionalism for precisely this reason. In their view, treating artists’ testimony in an intentionalist fashion amounts to taking sides.209 However, that argument cuts in both directions. Refusing to treat artists’ testimony in an intentionalist fashion, as opponents of intentionalism urge, requires the court to take sides against intentionalism.210 Why should courts take sides when doing so favors anti-intentionalism, but not when the opposite is true?

To be clear, the purpose of this discussion is not to argue that anti-intentionalists are wrong and that intentionalists are right. Indeed, anti-intentionalists have other arguments at their disposal.211 But so do intentionalists. Rather than fairly present the complexities of this ongoing debate, opponents of intentionalism simply declare anti-intentionalism’s victory. That assessment might be appropriate in the future, but it does not accurately describe the debate between intentionalism and anti-intentionalism as it exists today.212

3. Uncertainty

Appeals to uncertainty are also not compelling. Opponents of intentionalism argue that it creates inconsistent outcomes because it forces artists to testify about their intentions. In their view, artists are ineffective, unwilling, or unreliable witnesses.213 Instead, they contend, courts should rely on art markets and art experts to determine the significance of the defendant’s use in relation to the copyrighted work.214 However, as before, each of these arguments is dubious.

Arguments that intentionalism creates inconsistent outcomes are based more on feelings than on evidence. Empirical studies suggest that transformative use has increased consistency generally. For example, Neil Natanel finds that transformative use has imposed predictability on fair use determinations by unifying analysis around a single standard.215 Similarly, scholars have shown that courts make fair use determinations using consistent “patterns” or “clusters.”216 Although it is possible that contemporary art cases systematically depart from these general trends, transformative use’s general success in achieving consistency weighs against its elimination.

Furthermore, the structure of the fair use test provides a plausible alternative explanation for inconsistencies. As with other multifactor balancing tests, courts have strong strategic incentives to skew their analysis to ensure that every fair use factor supports the outcome reached.217 The more factors the court finds in favor of its desired outcome, the less likely the court’s decision will be reversed on appeal. Accordingly, two courts facing works on either side of the fair use line are likely to reach different results with respect to all four factors, even though the actual difference in status between the works may be small.218

Similarly, because fair use provides a complete defense to infringement liability, courts may split the difference by finding liability for one work, but not another, even though a straightforward application of the fair use test would not support this outcome. Just as courts engage in difference splitting by finding for copyright owners on infringement, but denying an injunction,219 courts may attempt to moderate their decisions through mixed liability findings.

As with stampeding, courts have strong strategic incentives to engage in difference splitting. For example, courts may believe that denying the plaintiff any remedy would be fair in cases like Cariou, where the defendant profited significantly from the plaintiff’s work.220 Alternately, difference splitting may enable courts to maintain their legitimacy with both parties and encourage compliance with their decisions.221 As before, these motivations create alternative sources of inconsistency, unrelated to intentionalism.

Opponents of intentionalism also do not convincingly explain why reliance on artist testimony is particularly likely to cause uncertainty. In many cases, parties must testify about their intentions even though they are unlikely to be able to do so effectively. Some commentators believe that witness testimony is systematically unreliable in adversarial frameworks because witnesses have strategic incentives to lie and lawyers have strategic incentives to distort their testimony.222 But this problem is endemic. There is no compelling reason why the law should excuse artists from testifying when it does not afford the same privilege to other parties, who may be equally ineffective or unreliable on the stand.223

Whatever the disadvantages of intentionalism, adopting the alternative strategies proposed by opponents would also produce uncertainty. Art markets, suggests Amy Adler,224 are notoriously unstable. Behavioral irrationalities, information asymmetries, and poor liquidity all make art markets bad mechanisms for determining the price of a defendant’s work.225 For example, many art purchases are made by private collectors, who tend to overvalue what they own and not consider the opportunity costs of investing their assets in art over alternatives.226 Further, because top artworks are traded very infrequently, and their prices are greatly affected by government interventions, such as export restrictions, sophisticated traders cannot cancel out the irrational trades of their unsophisticated counterparts.227 Thus, an artwork’s price may reflect nothing more than the idiosyncratic preferences of an individual purchaser.

Art experts fare no better.228 Even within anti-intentionalist schools, art critics do not agree on basic methods and principles of analysis, much less the meaning of individual works.229 Indeed, Walter Gallie described the very category of art as a quintessential “essentially contested concept”230—that is, a concept subject to endless disputes that, “although not resolvable by argument of any kind, are nevertheless sustained by perfectly respectable arguments and evidence.”231 Even if these disputes did not foreclose art critic testimony outright,232 plaintiffs could exploit them to introduce favorable testimony on fair use.233 In short, moving from artist testimony to any of the proposed alternatives would substitute one imperfect system for another.

iii. Whither Transformative Use?

So far, this Note has argued that intentionalism protects important First Amendment principles and that none of the arguments raised by opponents of intentionalism are compelling. How should copyright law proceed? Sections III.A and III.B present two strategies for preserving intentionalism within the fair use inquiry, while acknowledging the concerns raised by opponents. While superficially similar to existing approaches, both strategies provide an organizing framework for making transformative use determinations that preserves reference to the defendant’s intentions while avoiding the full force of anti-intentionalist arguments. Both strategies also have comparative advantages, which courts will need to evaluate in developing transformative use doctrine.

Section III.C anticipates objections that the proposed strategies would undermine the economic principles underlying copyright law. In particular, proponents of strong intellectual property rights protections might worry that granting defendants greater protections based on their intentions would undercut copyright owners’ incentives to create. In fact, neither strategy is likely to reduce creative output. And, depending on assumptions, granting artists greater freedoms to explore the objects of their intentions may increase the amount or quality of art produced.

A. Dual Standards

One strategy that courts could pursue would be to adopt expressly both intentionalist and anti-intentionalist standards for determining transformative use. Under this dual standards strategy, courts would ask both whether the defendant intended to transform the copyrighted work and whether the accused work was actually transformative. Lay or expert testimony (to the extent that it could be deemed reliable) could assist the latter inquiry, according to the strategies proposed by opponents of intentionalism.234 Currently, courts examine both the defendant’s intentions and the accused work’s content in determining transformativeness, but they do so haphazardly, without any organizing framework to guide their analysis. Under a dual standards strategy, findings of transformativeness under either standard would weigh in favor of fair use.

This strategy would have several advantages over alternatives proposed by opponents of intentionalism. First, such a strategy would accommodate the concerns of both intentionalists and anti-intentionalists. Combining intentionalist and anti-intentionalist standards would recognize the need to avoid interfering with the defendant’s communicative aims, which are present in the defendant’s intentions, while at the same time recognizing that the aesthetic debate between intentionalism and anti-intentionalism remains unsettled and that viewers might have residual interests in determining the meaning or significance of the defendant’s work, independent of what the defendant might intend.235 Accordingly, a dual standards strategy would give both artists and viewers an opportunity to influence the outcome of the fair use analysis in particular cases.236

Second, a dual standards strategy would provide superior ex ante predictability to potential defendants than pure anti-intentionalist approaches would. An artist deciding whether to create a potentially infringing work would know that she could rely on her intention to transform the original work as a defense to liability. Absent this, the same artist would need to predict the critical or commercial reception of the work, placing her in a much more precarious position.237

Third, a dual standards strategy would provide a more organized and defensible approach than those currently in use by the courts. As Sections I.B and I.C explain, individual courts have held the defendant’s intentions or the accused work’s content to be sufficient evidence of transformative use, even where one is present and the other is not.238 However, they have done so unreflectively, without explaining when or why courts should find these factors decisive. A dual standards strategy would provide a general framework for organizing these decisions, reducing apparent inconsistencies.239 At the same time, a dual standards strategy could acknowledge the arguments that motivate intentionalism and anti-intentionalism without resorting to the extreme positions present in some decisions.240

Of course, such a strategy would have detractors. Plaintiffs might argue that the use of both intentionalist and anti-intentionalist standards would provide too much protection for defendants, discouraging creators from producing original works. The strength of this argument depends, ultimately, on the extent to which copyright law (and competing principles, such as free speech) should protect first- and second-generation creators.241 But in any event, courts could moderate the harshness of the proposed strategy on plaintiffs by reducing the weight of a finding of transformativeness on the ultimate fair use determination. In other words, a finding of transformativeness would not necessarily be dispositive under either standard.242

B. Categorical Intentionalism

A second strategy that courts could pursue would be to adopt a “categorial intentionalist” standard for determining transformativeness. Unlike the dual standards strategy described above, categorial intentionalism would impose a single standard for determining transformativeness. Such a strategy would adopt an intermediate approach between absolute intentionalism and absolute anti-intentionalism, preserving the benefits of both extremes. As a form of intentionalism, categorial intentionalism maintains that the actual artist’s intentions are relevant to the work’s meaning. However, unlike absolute intentionalism, categorial intentionalism would limit the kinds of intentions that interpreters may consider.

Recall from Part II that, on Levinson’s influential account, categorial intentions describe the author’s intentions regarding how the work is to be “classified, taken, [or] approached,” including the author’s intentions that the work be considered as an example of a particular genre.243 According to Levinson, a work’s meaning depends on the intentions that an ideal audience would attribute to the work’s author.244 That ideal audience, in turn, is defined by the actual author’s categorial intentions regarding the work. Thus, interpreters may permissibly rely on the author’s categorial intentions because they “virtually cannot fail.”245 For example, an artist who creates a painting and intends that painting to be a work of abstract art almost always succeeds in doing so.

However, Levinson breaks from absolute intentionalists by prohibiting reference to the artist’s semantic intentions. Unlike categorial intentions, semantic intentions—the artist’s intentions “to mean something in or by” a particular work—are unreliable.246 An artist may intend for her painting to represent that “the mass production of commodities and media images has caused a deterioration in the quality of society.”247 However, she may be utterly unsuccessful in doing so if, for example, she chooses to situate that painting in a context that evinces tenderheartedness. Given this asymmetry in fallibility, semantic intentions “do not determine meaning, but categorial intentions, such as concern a literature maker’s basic conception of what is made, do in general determine how a text is to be conceptualized and approached on a fundamental level and thus indirectly affect what it will resultingly say or express.”248

Consequently, categorial intentionalism would require transformative use determinations to follow a two-stage process. First, parties would introduce evidence of the defendant’s categorial intentions, such as those regarding genre and setting. Evidence of the defendant’s semantic intentions would be excluded. Second, parties would contest whether the accused work was transformative given those intentions. Over time, courts could develop rules for determining which kinds of intentions satisfied this requirement and which kinds of evidence could be introduced to establish those intentions.

As with the dual standards strategy, this approach would provide a more organized and defensible approach than courts’ current approaches, helping to eliminate apparent inconsistencies in case outcomes. Although, as Blanch demonstrates, courts sometimes rely on evidence of the defendant’s categorial intentions, such as the defendant’s intention to display the accused work in a particular gallery,249 they do so unsystematically, and without distinguishing those intentions from other, more abstract intentions, such as the defendant’s intention to parody the copyrighted work. Adopting a categorial intentionalist strategy would give courts firm instructions about which intentions to consider and which to ignore, while providing a theoretical foundation for these choices.

The advantages and disadvantages of categorial intentionalism would be similar to those of the dual standards strategy. Like the dual standards strategy, categorial intentionalism would respect the defendant’s communicative aims, even where those aims conflicted with viewers’ responses to the accused work. Similarly, categorial intentionalism would provide greater ex ante predictability to artists by allowing them to insulate their activities from liability by taking steps to establish works as belonging to a particular genre or by associating those work with others that express a particular, critical message. Also like the dual standards strategy, categorial intentionalism might overprotect defendants at plaintiffs’ expense. But again, courts could moderate the potential harshness of this strategy by reducing the weight of a finding of transformativeness within the fair use inquiry.

However, categorial intentionalism would result in a different balance between artists’ and viewers’ interests than would a dual standards strategy. Whereas the dual standards strategy would respect artists’ categorial and semantic intentions, categorial intentionalism would respect only the former. Given this, the First Amendment principles discussed in Section II.B.1 may make this approach impermissible to the extent that the First Amendment protects the artist’s communicative aims, which would include her semantic aims. Still, respecting the artist’s categorial intentions might be sufficient because it provides at least some protection. And even if categorial intentionalism’s failure to respect artists’ semantic intentions might raise autonomy concerns, this might not be troubling in the case of artists who, like Prince, disavow having a transformative purpose.

Which of these considerations dominate is a question that courts will need to answer in attempting to implement these strategies. For now, it is enough to state that the choice between intentionalism and anti-intentionalism is not an all-or-nothing proposition. To the extent that anti-intentionalist arguments raise concerns, courts can choose to adopt both intentionalist and anti-intentionalist standards or restrict the range of intentions upon which defendants may rely.

C. Incentives To Create

According to the standard economic theory of intellectual property, copyright systems exist to encourage creative production. Because creative works, like other information goods, have high fixed costs and low marginal costs—they are expensive to produce and cheap to copy—creators selling in competitive markets may not be able to recover their investments.250 Copyright systems attempt to solve this perceived problem by allowing creators to exclude competitors for fixed periods of time, allowing those creators to charge higher prices than would otherwise be possible.251

Against this backdrop, some might worry that, by increasing the scope of protections afforded to defendants, the strategies proposed in Sections III.A and III.B would reduce prospective creators’ expected profits and, therefore, their incentives to create. However, even if intentionalist standards were systematically more protective than anti-intentionalist ones, there are several reasons to doubt that either proposal would significantly retard creative production.

Significantly, the standard theory assumes that prospective creators are rational wealth-maximizing individuals, whose primary concern is to maximize the economic returns on their labor investments. But that assumption is questionable. Drawing on economics, psychology, and other disciplines, a growing number of legal scholars argue that artists and other creators are intrinsically motivated to create artistic works, without the need for copyright systems and other external incentives.252 Relying on external incentives may even reduce creative output. For example, Bruno Frey theorizes that extrinsic interventions, such as copyright systems, may crowd out intrinsically motivated creators, who would otherwise flourish.253 Practical barriers make Frey’s theory difficult to evaluate.254 But if he is correct, granting artists greater freedom to pursue the objects of their intentions may result in more or higher quality art.

Furthermore, even if creators are perfectly rational wealth-maximizers, as the standard theory assumes, granting greater fair use protections will often have no effect on prospective creators’ expected returns. This is because demand for the copyright owner’s work and the defendant’s work are frequently unrelated.255 A buyer interested in Rogers’s $200 photograph is unlikely to be interested in Koons’s $367,000 painting, and vice versa. Although some investments may be unprofitable absent revenues from derivative works, it is unclear whether this is true for the majority of cases, or even a substantial minority.256

Relatedly, the effect of greater fair use protections on overall creative production, the ultimate goal of the copyright system under the standard theory, would be at the very least ambivalent. To the extent that the proposals would decrease first-generation creators’ incentives, they would also increase second-generation creators’ incentives to transform those works in potentially artistically valuable ways.257 This tradeoff is especially acute for areas like appropriation art, where first-generation creators are poorly positioned to create second-generation works because of their destructive or critical character.

Whether, in the final analysis, greater fair use protections would produce more or higher quality art is a difficult empirical and normative question. It is empirically difficult because the optimal balance between first- and second-generation creators’ incentives is hard to measure and normatively difficult because the relative artistic value of first- and second-generation works is contested.258 Thus, objecting to greater fair use protections because they would decrease copyright owners’ incentives is premature without first establishing their role within the broader process of creative production.

Finally, even if stronger fair use protections did, on balance, reduce copyright owners’ incentives, many prospective creators would still retain powerful incentives to continue production even in the face of copying. The prospect of obtaining a monopoly is not the only, or even most powerful, economic incentive available to prospective creators. Copyright owners would still benefit from lead-time advantages, and many would derive economic benefits, such as goodwill, that are only weakly affected by copying.259 As with inter-generational effects, the significance of these non-intellectual property incentives is not fully understood, but, as before, they weaken the force of the economic worry.

conclusion

Commentators who oppose transformative use’s “intentionalism” are wrong to argue that anti-intentionalism provides unequivocally greater protection for art. If the defendant intends to transform the copyrighted work, but viewers do not perceive the accused work as transformed, only an intentionalist standard will protect her against liability. Moreover, in those circumstances, anti-intentionalist standards create potential First Amendment problems. Both doctrine and theory suggest that an artist’s intentions should be at least relevant to the question of whether she is liable for the harmful effects of her speech.

Moreover, opponents of intentionalism are wrong to declare anti-intentionalism’s victory as an aesthetic theory. In fact, the debate between intentionalism and anti-intentionalism outside of copyright law is much more complex. Whether anti-intentionalism ultimately triumphs is a question best answered by philosophy and the critical disciplines. At the moment, anti-intentionalism’s victory over intentionalism is not nearly as decisive as opponents suggest. And, as Dennis Dutton explains in the epigraph to this Note, the answer to that question is unlikely to emerge in the near future.260

Opponents of intentionalism often assert that copyright law should broadly allow artistic copying.261 Perhaps they are right. Perhaps fair use should provide greater protection for art or even exempt all artistic works from liability. After all, copying is rampant in the art world, and some artists have produced their most celebrated works by borrowing from their predecessors.262 But these arguments transcend the debate over intentionalism, and should be considered on their own merits, without one-sided reference to aesthetic theory.