Restructuring Copyright Infringement

Article - - Issue 4

Copyright law employs a one-size-fits-all strict liability regime against all unauthorized users of copyrighted works. The current regime takes no account of the blameworthiness of the unauthorized user or of the information costs she faces. Nor does it consider ways in which the rightsholders may have contributed to potential infringements, or ways in which they could have cheaply avoided them. A nonconsensual use of a copyrighted work entitles copyright owners to the full panoply of remedies available under the Copyright Act, including supra-compensatory damage awards, disgorgement of profits, and injunctive relief. This liability regime is unjust, as it largely fails to differentiate among willful infringers, good-faith users who accidently infringe, and ordinary users who fail to understand the intricacies of copyright-protection infringement. No less importantly, one-size-fits-all liability sets back the goals of the copyright system by excessively deterring newly created works and encouraging copyright owners to be less than forthcoming about their expectations from users.

In this Article, we propose a radically different liability regime predicated on the degree of the infringer’s blameworthiness. Concretely, we call for the establishment of three distinct liability categories—inadvertent, standard, and willful infringements—and tailor a specific menu of remedies for each of the categories. Under our proposal, inadvertent infringements would be remedied by compensatory damages only. Standard infringements would entitle copyright owners to a broader variety of monetary damages, including disgorgement of profits and limited statutory damages, but typically no injunctive relief. Willful infringements would be addressed by the full variety of remedies available under the Copyright Act, including punitive statutory damages and injunctions.

We demonstrate that our proposed system of graduated liability yields a far more nuanced and just liability system, one that would also enhance use of existing works and promote future creativity. Limiting the potential liability of users and creators would facilitate the use of copyrighted content and the production of future works. It would also result in a fairer copyright system, the benefits of which would inure to society at large. At the same time, by organizing infringement into our proposed categories, our system would reform infringement law without imposing excessive new fact-finding burdens on courts and litigants.

Introduction

Earlier in 2018, the music world found itself engulfed in yet another copyright controversy, when Universal Music Studios—which owns the copyright in The White Stripes’ “Seven Nation Army”—sent a presuit letter to the creators of the song that won the 2018 Eurovision song contest. According to Universal Music Studios, the winning Eurovision song, “Toy,” performed by the singer Netta, was impermissibly substantially similar to “Seven Nation Army.” Despite the obvious differences in lyrics and melody between the two songs, Universal’s attorneys claimed that they share the same rhythm and harmony.[1]

Such controversies are all too common in the copyright world. In Williams v. Bridgeport Music, Inc.,[2] Robin Thicke acknowledged that Marvin Gaye’s music was an inspiration for the Thicke and Pharrell Williams collaboration “Blurred Lines.”[3] However, while “Blurred Lines” shared some of the “feel” of Gaye’s “Got to Give It Up,” it copied none of the melody or lyrics.[4] The actual sound recording of “Got to Give It Up,” meanwhile, was unprotected due to a copyright technicality,[5] meaning that Thicke and Pharrell could copy Gaye’s recording (but not composition) with impunity. Nevertheless, a jury decided that the appropriation of the mood and feel of Gaye’s song was enough to ground a decision of infringement.[6] From there, it was a short distance to the order of $5.6 million in damages (an amount that was subsequently reduced to about $3.5 million) to the estate of Marvin Gaye, as well as 50% of future proceeds from the song.[7] The appeals and rehearings eventually upheld the result.[8]

These and all too many other cases highlight a central flaw in the extant copyright system of enforcement. Under copyright law’s standard of uniform liability, a defendant who subconsciously or inadvertently borrows protected elements of a copyrighted work is just as liable as one who willfully pirates and distributes an entire work. A trivial infringement—such as a too-long quotation from a book[9] or the unconscious borrowing of elements of a pop tune[10]—is treated just like a willful and malicious pirating and distributing of a copyrighted work.[11] Infringers are liable for damages that bear no proportion to the moral culpability of the defendant or, for that matter, to the damages that their behavior actually causes. The compensation amount in copyright-infringement cases depends only on amounts specified by statute, the harm inflicted on the plaintiff, or the profit of the defendant.[12] Blatant infringers who profit little may end up paying modest damages, while inadvertent infringers who create popular works may end up paying enormous damages.[13] The degree of liability is completely divorced from the degree of culpability of the defendant. It is likewise divorced from the actual harm to society caused by the infringement.

Compare, for instance, the cases of Dr. Seuss Enterprises, L.P. v. Penguin Books USA, Inc.[14] and Campbell v. Acuff-Rose Music, Inc.[15] Campbell concerned a claim of infringement against 2 Live Crew. The band’s song “Pretty Woman” had copied a bass riff and several lyrics from Roy Orbison’s “Oh, Pretty Woman.”[16] The Supreme Court ruled that the copying was within the bounds of fair use and thus not actionable.[17] A short time later, Alan Katz and Chris Wrinn wrote an account of the O.J. Simpson trial in the style of a Dr. Seuss book, entitled The Cat NOT in the Hat! A Parody by Dr. Juice.[18] Katz and Wrinn clearly thought their copying of the Dr. Seuss style was permissible fair use, as in Campbell.[19] The courts disagreed. Once they determined that The Cat NOT in the Hat! used materials in a style not favored by the fair use doctrine, the courts fell back upon the ordinary doctrines of copyright infringement.[20] The trial court therefore enjoined further distribution of the book.[21] It is extremely difficult to locate copies of the Katz–Wrinn work today.

In this Article, we propose a radical reform in the way copyright law assigns liability. Extant copyright law treats virtually all copyright infringements similarly for purposes of liability.[22] As a general rule, copyright law does not distinguish among different types of infringement. Nor is it open, in its current design, to the idea that liability should be graded based on the degree of wrongdoing by the infringer.[23] Consequently, all infringers are exposed to the full cascade of monetary remedies and injunctive relief.[24] Liability is currently an objective inquiry in which courts decide whether the alleged defendant copied protected expressive elements from the plaintiff.[25] Once the basic elements of an infringement suit—copying and improper appropriation[26]—have been established, liability attaches to the defendant, and the plaintiff is entitled to a menu of potential remedies: statutory damages, recovery of her losses or the defendant’s profits, and, in addition, seizure of infringing copies and even injunctions barring the defendant from using or distributing her own work.[27]

Infringement remedies under current law can be quite harsh, even where the infringement is questionable. The celebrated cases of Bright Tunes Music Corp. v. Harrisongs Music, Ltd.[28] (regarding alleged infringement by the George Harrison song “My Sweet Lord”)[29] and Williams v. Bridgeport Music, Inc. (regarding alleged infringement by the Thicke–T.I.–Pharrell song “Blurred Lines”)[30] provide a powerful illustration. In Harrisongs, George Harrison was found to have subconsciously infringed the copyright in The Chiffons’ song “He’s So Fine” by writing and releasing to the public his own hit song “My Sweet Lord.”[31] The court believed Harrison’s testimony about the creative process that yielded “My Sweet Lord,” including the key fact that Harrison composed his song without consciously copying anything from The Chiffons’ earlier song.[32] But little did it help Harrison. The court ordered the payment of $1.6 million of Harrison’s earnings from “My Sweet Lord” to the music publisher of The Chiffons.[33]

As these and many other cases illustrate, copyright law is deliberately draconian. The lines separating infringing “copying” from noninfringing inspiration can be difficult to discern. As well, the boundaries of permitted behavior—under the doctrine of fair use, or as a result of owner-granted or statutory licenses—can be fuzzy.[34] But court decisions are dichotomous. Once the line is crossed and an infringement is established, the rules of relief are uniform and harsh.

There is one outstanding exception to the rule of uniform damages: statutory damages. The statutory provision that provides for statutory damages embodies a three-part design.[35] Section 504(c) differentiates among standard infringements, willful infringements, and innocent infringements.[36] It then gives courts discretion to award damages ranging from $750 to $30,000 per infringement in cases of standard infringements, while empowering them to step up the amount to $150,000 in cases of willful infringements and reduce it to merely $200 in cases of innocent infringements.[37] We posit that the statutory-damages framework constitutes a useful blueprint for reforming infringement analysis in all cases. Yet, we also argue that the other forms of available relief, as well as the structure of statutory damages, should be altered to create a harmonious legal regime.

Our proposed reform, thus, consists of two changes. First, drawing inspiration from the extant rules of statutory damages, we argue that infringements should be divided into three different categories, with varying legal consequences. The first category would be called “inadvertent infringement” and would encompass two prototypical infringement cases: (a) subconscious infringements, as in the case of Bright Tunes Music v. Harrisongs Music;[38] and (b) infringements involving nothing more than appropriation of “concept and feel,” à la Williams v. Bridgeport Music[39] or other cases where the infringement is so small or questionable that a court could conclude that the infringer reasonably believed in good faith that nothing protectable was copied.

At the other end of culpability, the third category would be labelled “willful infringements.” This category would include blatant and inexcusable infringement, such as wholesale, repeated copying of works in their entirety, without plausible claims of license or statutory excuse. The actions of copyright pirates would ordinarily fall within this category.

Finally, a middle, second category would include all other infringements and be labelled “standard infringements.” This category would encompass cases where the defendant is not blameless but is clearly not as culpable as the willful infringer. This would include cases such as colorable cases of fair use in which the defendant mistakenly “went too far,” as happened in Dr. Seuss Enterprises, L.P. v. Penguin Books USA, Inc.[40] and cases in which the infringers accidently exceeded the terms of their licenses. This last type of case is best illustrated by the litigation surrounding claims that the group The Verve infringed upon the copyright in the Rolling Stones’ song “The Last Time” when The Verve sampled too-large a portion of the Andrew Oldham Orchestra’s symphonic recording of the Rolling Stones’ song.[41] According to former Rolling Stones manager Allen Klein, who owned the copyright in the Rolling Stones composition, The Verve had negotiated a license to sample five notes from the Andrew Oldham Orchestra recording, but the portion it actually used slightly exceeded the five-note limit.[42] Rather than face costly and uncertain litigation, The Verve settled out of court and yielded all the revenues from “Bitter Sweet Symphony,” the group’s biggest hit.[43]

The second prong of our reform is to tailor the remedies for each of our proposed three categories to the degree of infringer culpability. Under our proposal, for cases of inadvertent infringements, courts would be limited to awarding purely compensatory damages and, in very rare cases, also injunctive relief to stop unauthorized distribution of unauthorized copies. For cases of standard infringements, courts would be able to expand the remedial menu and, in addition to compensatory damages, order disgorgement of profits or, in the alternative, limited statutory damages. Here, too, injunctive relief would typically be withheld. Finally, in cases of willful infringement, courts would be able to utilize the full panoply of copyright remedies, including all of the above, as well as high statutory damages, criminal sanctions, and injunctive relief. Importantly, the new liability regime would limit courts both in lawsuits for standard remedies and for statutory damages, thus bringing uniformity to copyright remedies while differentiating liability according to culpability.

Proper allocation of the burdens of proof would be necessary for our reform to be workable. In our scheme, all infringements would be presumptively standard, unless either the defendant could prove her innocence or the plaintiff could prove willfulness or malice. In other words, the burden of proving that a case should be classified as an inadvertent infringement case would rest with the user–infringer, while the burden of proof that an infringement was willful would lie with the plaintiff copyright owner. Yet, we would supplement the aforementioned rules concerning the burden of proof with one additional possibility: we would allow users to initiate suits seeking declaratory judgments that they were entitled to the status of an innocent infringer.

Calibrating liability (and remedies) to culpability holds obvious intuitive moral appeal. But it is also advantageous from a utilitarian perspective. Copyright protection imposes substantial information costs on third parties, among them, other authors. The high level of risk of legal liability when creating new works is the result of many factors in combination: the vast universe of protected works, the difficulty of distinguishing between elements that are copyrighted (that cannot be used without permission) and uncopyrighted (that can be freely used), the broad scope of liability that makes even unconscious copying actionable, and the existence of malleable defenses to liability. The problem is exacerbated by the absence of a readily available registry of all existing works. This makes it nearly impossible to know not only what works and which aspects of them are protected but also who the relevant rightsholders are. In this informationally treacherous environment, the best way to curb the risks faced by aspiring authors is to limit the liability to which they will be exposed if they unknowingly commit a copyright infringement. This not only reduces unnecessary barriers to entry by new authors, it also incentivizes authors to make their claims known as widely as possible, so as to foreclose the possibility of innocent infringements.

Efficiency supports bringing copyright punishments closer to existing moral intuitions for another reason as well. Extant literature points to a significant gap between popular understandings of social norms of copyright and the law of copyright, as well as between the demands of the law and popular understanding of those demands.[44] These gaps increase the costs of enforcing copyright. Potential defendants will not conform their behavior to a law of which they are unaware, and given both the difficulty of detecting infringement and the difficulty of understanding the law, it may be rational to remain ignorant. At the same time, these gaps may lead judges and courts to depart unilaterally from the legal norms when they believe the law is unnecessarily harsh. By reducing these gaps, lawmakers can help copyright enforcement become cheaper and more uniform. Even if penalties are reduced in some cases, improvements in the rate and accuracy of enforcement can lead to a better incentive structure of the law.[45]

Concerns for efficiency also lay behind the scope of our potential reform. Given the wide variety of potential types of infringements, one might imagine an infinite range of remedies, each tailored to the precise degree of culpability. However, administering such infinitely tailored remedies would prove enormously costly for two reasons: learning and administering a new (and unbounded) legal system would place a heavy burden on courts and litigants, as would gathering and exposing evidence relevant to the infinitely tailored system.[46] By building on the existing categories of statutory damages, and by limiting the potential classes of liability to three, we minimize the potential administrative and litigation costs associated with our proposed overhaul of the law of liability. In particular, given the fact that most copyright-infringement cases already involve claims for statutory damages,[47] our proposal will not add substantially to the existing evidentiary burden.

We lay out our argument in three parts. In Part I, we review the extant law of infringements, demonstrating the vast array of circumstances and harms encompassed by the term “copyright infringement.” In Part II, we examine the remedies available under copyright law, while emphasizing, in particular, their Procrustean nature. Additionally, we demonstrate why the narrow range of available remedies is socially undesirable. In Part III, we offer our proposed reform of copyright remedies. Finally, in Part IV, we explain why, despite the advantage of staggered remedies, we offer only three categories based on culpability.

I. The Wide World of Infringements

In this Part, we highlight just some of the many ways in which copyright law might be infringed. The categories of infringement we offer here are not those of the law. In fact, the law conceives of all infringements as identical. As we will show in the next Part, no matter what kind of infringement, copyright law offers the same remedy.

We begin with some general observations about infringements, before offering a tentative classification of different kinds of infringements based on culpability. As we show in the conclusion of this Part, mistakes are the source of many—and probably most—infringements.

A. The Vagueness of Infringement Law

As a preliminary matter, it is important to observe that the variability of infringements is deeply rooted in copyright law. Copyright law is, of necessity, vague and often maddeningly so. Copyright law grants authors[48] monopoly rights in their expressions, but defining the nature or even the protected expressions of that monopoly is no easy matter. Perhaps the most famous expression of this vagueness is Justice Story’s observation that “copyrights approach, nearer than any other class of cases . . . to what may be called the metaphysics of the law, where the distinctions are, or at least may be, very subtle and refined, and, sometimes, almost evanescent.”[49] No less famous is Judge Learned Hand’s explanation of copyright law’s idea–expression dichotomy:

The test for infringement of a copyright is of necessity vague. In the case of verbal “works” it is well settled that although the “proprietor’s” monopoly extends beyond an exact reproduction of the words, there can be no copyright in the “ideas” disclosed but only in their “expression.” Obviously, no principle can be stated as to when an imitator has gone beyond copying the “idea,” and has borrowed its “expression.” Decisions must therefore inevitably be ad hoc.[50]

Copyright law protects all original works of authorship fixed in a tangible medium of expression for more than a fleeting moment.[51] Works do not need to be published to be protected.[52] While there is a copyright registry, works need not be marked or registered to qualify for protection.[53] It is difficult to know what qualifies as a work of authorship; in recent decades, protection has been claimed for works as diverse as computer operating systems,[54] tattoos,[55] and a wildflower garden.[56]

The uncertain legal standards combine with the law’s formalities to create yet more doubt. For instance, while copyright protection is extremely long-lasting (generally lasting for more than a century),[57] the terms of protection have changed over time, along with the formal requirements for protection,[58] and even the definitions of protected elements of a work.[59] Thus, it is that as recently as 2013, the courts found themselves ruling on an ownership dispute over the copyright in different elements of the Superman character,[60] who first appeared in comic books in 1938.[61] In some cases, courts have split up the rights in works in ways that would strike a layperson as bizarre, such as by distinguishing ownership of a character’s catchphrases from the character’s characteristic wardrobe.[62] The result is that users of expressive works can never be sure which works are under legal protection and which have fallen into the public domain and at what point the legal protection, if it exists, expires. The informational haze shrouding the copyright terrain has even spawned the odd legal phenomenon of “orphan works”—works that are subject to copyright protection but whose rightsholders are unknown.[63]

Together, legal vagueness, unclear information in rights, and a number of market practices combine to create an atmosphere in which copyright rights can be infringed in numerous ways that are not willful and perhaps not even readily detectable.

It is against this background that we turn to a description of several different kinds of infringement, with a focus on the ways rights may be infringed without willfulness. We will argue infra that these several kinds of infringements can be roughly grouped into three categories—willful infringements, inadvertent infringements, and “standard” infringements.[64]

B. The Different Types of Infringements

1. Willful and Blatant Infringements.—We begin with the infringements that the law must combat most strongly: willful and blatant infringements of copyrights. Notwithstanding all the ambiguity surrounding copyright law, there are many cases that are easy. Counterfeit DVDs, for instance, are produced wholesale by pirates who know that they are copying a copyrighted work (a motion picture) and then distributing it to the public without any authorization.[65] The counterfeiters are not confused about the illegality of their actions. They are simply relying on not getting caught.

Every year there are numerous enforcement actions against such willful and blatant infringements.[66] Consider the enforcement action that was brought against Danny Ferrer.[67] Ferrer, together with others, operated the piracy website www.buyusa.com, where he sold illicit copies of copyrighted software products, such as Adobe and Autodesk (among others).[68] Ferrer’s infringing activities lasted several years, causing the rightsholders harm that totaled nearly $20 million.[69] Ultimately, Ferrer was sentenced to a prison term of six years and was ordered to pay $4.1 million in restitution.[70] In addition, he was forced to forfeit a whole variety of luxury goods he had acquired from the proceeds of his illegal activities.[71]

Although Ferrer’s case is extreme in terms of the punishment meted out to him, it is highly representative of infringing activities concerning copyrighted software, movies, and music files on the Internet. Infringements of copyrights in such works often consist of a knowing and willful mass production of illegal copies of protected works, and, thereafter, the knowing and willful offering of those copies to the public at a heavily discounted price.

As we explicate further in Part III, we classify these kinds of infringements “willful infringements.”[72]

2. Unknowing Infringements.—At the other end of the spectrum are inadvertent infringements: instances where the infringement is due to a reasonable and good-faith interpretation of the law and facts that leads the infringer to believe she is doing nothing wrong. As we will see, there are many different instances where someone may use a copyrighted work in a forbidden manner, yet lack culpability. Roughly speaking, the innocence of the infringer may stem from two different kinds of mistakes. The infringer may make a mistake of law, or incorrect predictions about the way courts are likely to interpret the law that lead her to believe that the law permits her activity. Alternatively, the infringer may make a mistake of fact or believe certain things about the quality of a work or the available licenses that lead her to believe that her activity is not infringing.

The most extreme type of inadvertent infringement is one where even after committing an act that the court will eventually find to have been infringing, the infringer is unlikely to know that she did anything wrong.

In theory, such unknowing infringements should not be possible. Although the Copyright Act does not explicitly require it, case law is quite clear in demanding that any case of infringement be supported by a finding of “copying in fact.”[73] That is, according to the courts, an infringement does not take place unless “the defendant, in creating its own rival work, used the plaintiff’s material as a model, template, or even inspiration.”[74] Yet, despite this clear element of the case law, courts have signaled their willingness to find an infringement where the defendant was not aware of the plaintiff’s work, and did not consciously use it even as inspiration.

The extreme example of this kind of infringement is provided by the famous case of Bright Tunes Music v. Harrisongs. Bright Tunes involved a suit brought against George Harrison, the former Beatle, alleging that his hit song “My Sweet Lord” infringed Bright Tunes’ copyright in The Chiffons’ “He’s So Fine.”[75] In response to the suit, Harrison attested that he did not copy The Chiffons’ song or borrow from it.[76] The court explained that while it believed Harrison, he had nevertheless infringed Bright Tunes’ copyright:

Seeking the wellsprings of musical composition—why a composer chooses the succession of notes and the harmonies he does—whether it be George Harrison or Richard Wagner—is a fascinating inquiry. It is apparent from the extensive colloquy between the court and Harrison covering forty pages in the transcript that . . . Harrison [was not] conscious of the fact that [he was] utilizing the He’s So Fine theme. . . .

What happened? I conclude that the composer, in seeking musical materials to clothe his thoughts, was working with various possibilities. As he tried this possibility and that, there came to the surface of his mind a particular combination that pleased him as being one he felt would be appealing to a prospective listener; in other words, that this combination of sounds would work. Why? Because his subconscious knew it already had worked in a song his conscious mind did not remember. . . . Did Harrison deliberately use the music of He’s So Fine? I do not believe he did so deliberately.[77]

The court reasoned that Harrison was probably exposed to The Chiffons’ song that was released in 1963 and was subconsciously influenced by it when he wrote his song that was released to the public in November of 1970. That Harrison was unaware of this fact was of no help to him.[78]

Subconscious copying also played a key role in the case of Three Boys Music Corp. v. Bolton.[79] This case proceeded on a claim that Michael Bolton’s song entitled “Love is a Wonderful Thing” plagiarized a song with the same title by the Isley Brothers.[80] Naturally, the identical title of the two songs could (and did) raise suspicion of illicit behavior, but it turns out that 129 songs bearing the same title were registered at the time of the suit.[81] That Bolton’s song was released more than 30 years after the original did not save him.[82] Nor did it help that the Isley Brothers’ song did not attain meaningful commercial success.[83] In fact, the court used Bolton’s admission of his admiration of the Isley Brothers to corroborate the jury’s finding that Bolton copied the Isley Brothers’ song.[84]

To be sure, the court did not rule that Bolton had infringed the Isley Brothers’ copyright; it only upheld a jury finding that Bolton had infringed. Nonetheless, it is difficult to read the court’s opinion and remain unimpressed with how a borderline case of plausible “copying” can suffice to uphold a ruling of infringement. According to the Court:

The appellants contend that the Isley Brothers’ theory of access amounts to a “twenty-five-years-after-the-fact-subconscious copying claim.” Indeed, this is a more attenuated case of reasonable access and subconscious copying than [Bright Tunes Music]. In this case, the appellants never admitted hearing the Isley Brothers’ “Love is a Wonderful Thing.” That song never topped the Billboard charts or even made the top 100 for a single week. The song was not released on an album or compact disc until 1991, a year after Bolton and Goldmark wrote their song. Nor did the Isley Brothers ever claim that Bolton’s and Goldmark’s song is so “strikingly similar” to the Isley Brothers’ that proof of access is presumed and need not be proven. . . .

. . . .

Teenagers are generally avid music listeners. It is entirely plausible that two Connecticut teenagers obsessed with rhythm and blues music could remember an Isley Brothers’ song that was played on the radio and television for a few weeks, and subconsciously copy it twenty years later.[85]

Needless to say, there is a vast gulf between blatant pirating of copyrighted works and subconscious copying of the same.

As we explicate further in Part III, we classify these kinds of infringements as “inadvertent infringements.”[86]

3. Infringements Related to Questions About the Work’s Protection.—Infringements vary not only due to the behavior of the infringer. They also vary due to the nature of the copyrighted work. Sometimes, the application of copyright law to a particular work is obvious. Copyright law, for instance, clearly protects the text of a novel.[87] But the law’s protection does not end there. The novel’s characters[88] and style[89] may be protected, as may other “nonliteral” aspects of the work. Indeed, every protected work contains both protected and unprotected parts. The text of a history book is protected but not its historical facts.[90] The plot of a movie is protected but not insofar as the plot is unoriginal or consists of scènes à faire.[91] Someone may copy elements of a work she thought were unprotected, and find out later to her chagrin that the court considers her copying an infringement.

In some cases, the statute explicitly divides the work into protected and unprotected parts. For example, the copyright in architectural works is limited to “overall form as well as the arrangement and composition of spaces and elements in the design [of a building], but . . . not . . . individual standard features.”[92] A more controversial example is provided by the law of copyright protection of works of industrial design—“pictorial, graphic, and sculptural works” that are “useful articles” in the jargon of the Copyright Act.[93] Such works are protected only in their “form” but not their “mechanical or utilitarian aspects,” and only to the degree that their design “incorporates pictorial, graphic, or sculptural features that can be identified separately from, and are capable of existing independently of, the utilitarian aspects of the article.”[94] An espresso machine may thus enjoy copyright protection for part of the way it looks: for its form, but not its mechanical aspects, and even then, only to the degree that the artistic aspects of the form are identified separately from the utilitarian aspects. Years of litigation about such items as belt buckles,[95] bicycle racks,[96] streetlamps,[97] mannequin heads,[98] and cheerleading uniforms[99] have failed to draw a clear line.[100]

An additional complication is created by the fact that copyrights are time limited, but the protected aspects of works are not created instantaneously at a single moment. Consider the litigation surrounding the copyrighted character Sherlock Holmes. Sir Arthur Conan Doyle was a writer in the late nineteenth and early-twentieth century whose most famous works concerned the adventures of the fictional detective Sherlock Holmes. The first Sherlock Holmes story appeared in 1887 and the last in 1927.[101] It is evident that the majority of the Sherlock Holmes stories have long been in the public domain. However, the Doyle estate has successfully upheld the copyright in ten short stories published from 1923 and onward.[102] The result is that any marginal accretions to the Sherlock Holmes character that appear in those final short stories are still under copyright protection. Warner Brothers could lawfully use Sherlock Holmes as the central character in its 2009[103] and 2011[104] films, without a license from the Doyle estate, so long as it avoided any references to Holmes’s development in a character in the final ten short stories. Likewise, a different studio could today make its own original Sherlock Holmes films, as long as it avoided using the last ten short stories, or any aspects of the character that were original to the Warner Brothers’ films (or other still-protected depictions in films and television programs). In other words, when considering the Sherlock Holmes character, like any other copyrighted work created over time, one must view the work as layered; although the work may seem a seamless whole, in fact it is comprised of different pieces built upon one another, some of which enjoy copyright protection, and others of which do not.

Figuring out whether one has copied a protected part of a work is a perennial difficulty in the law of copyright. As Judge Learned Hand famously observed in the 1930 case Nichols v. Universal Pictures Corp.,[105] “[n]obody has ever been able to fix that boundary [between protected elements and unprotected “abstractions”], and nobody ever can.”[106] Indeed, in Nichols, which concerned the claim that the plot of the play The Cohens and the Kellys infringed upon the plot of the earlier play Abie’s Irish Rose, Judge Hand was forced to recite the plots of both plays at length, before dismissing the case with such imprecise observations as:

[W]e think both as to incident and character, the defendant took no more—assuming that it took anything at all—than the law allowed. The stories are quite different[:]

. . . [T]he theme was too generalized an abstraction from what she wrote . . . .

 . . . [And] as to her characters[, they are mere] . . . stock figures.[107]

The difficulty in applying the law is perhaps best illustrated by comparing two cases involving stories about the continuing adventures of well-known characters. The first, Warner Bros. Pictures Inc. v. Columbia Broadcasting System,[108] involved the fictional detective Sam Spade and other more minor fictional characters who appeared in the Sam Spade adventure The Maltese Falcon.[109] Sam Spade was the main character in Dashiell Hammett’s novel The Maltese Falcon, published in serial form from 1929–1930.[110] Later in 1930, Hammett (and his publisher Alfred A. Knopf, Inc.) sold all television, radio, and movie rights to Warner Brothers; Warner Brothers adapted the novel into several different film versions, of which the most successful by far was the 1941 film The Maltese Falcon, which starred Humphrey Bogart.[111] The controversy developed in 1946, when Hammett granted the Columbia Broadcasting System the right to use the character Sam Spade (and several other characters) in a radio drama entitled “Adventures of Sam Spade” that was later broadcast weekly in half-hour episodes until 1950.[112] Warner Brothers sued, claiming that the radio shows infringed upon the rights in the character Sam Spade it had acquired when it purchased the radio rights to The Maltese Falcon.[113] However, the district court and then the Ninth Circuit Court of Appeals ruled in favor of Hammett and Columbia Broadcasting System on the grounds that the character Sam Spade was not a copyrighted work included in The Maltese Falcon; indeed, according to the court, the character Sam Spade could not be copyrighted at all since he was merely the “vehicle[] for the story told, and the vehicle[] did not go with the sale of the story.”[114]

Nearly the opposite approach was exhibited by a district court in California in Anderson v. Stallone[115]—litigation concerning the character Rocky Balboa, and other, more minor, fictional characters that appeared in the Rocky film series.[116] Timothy Anderson, a fan of the Rocky film series, was inspired by Sylvester Stallone’s brief comments during a promotional tour of Rocky III regarding the potential plot of Rocky IV.[117] Building on Stallone’s vague idea of a fight in Moscow between Rocky Balboa and a Soviet fighter, Anderson wrote a 31-page summary of the plot of a film (a “treatment,” in the jargon of the film industry) involving a fight between Rocky and a highly trained Soviet fighter, which he then delivered to studio executives.[118] According to Anderson, Stallone’s final script for Rocky IV copied the plot of Anderson’s treatment.[119] The court ruled for Stallone.[120] While purporting to apply the same standard as the Warner Brothers decisions,[121] the court ruled that “the Rocky characters were so highly developed and central to the three movies made before Anderson’s treatment that they ‘constituted the story being told’” and were, therefore, “[e]ntitled [t]o [c]opyright [p]rotection [a]s [a] [m]atter [o]f [l]aw.”[122] Consequently, ruled the court, Anderson could not enjoy copyright protection for any of his treatment; Anderson’s treatment was nothing more than an unauthorized derivative work which unlawfully used Stallone’s copyrighted characters.[123] It was Anderson, rather than Stallone, who had infringed.

Against this background of legal uncertainty, one might easily infringe, yet fail to identify the legal consequences until the court’s ruling. Does a television commercial infringe the copyrights in the James Bond films when it shows a tuxedoed American male spy engaged in flirtatious banter with an attractive female companion while driving a Honda del Sol to escape a villain in a helicopter, all to the sounds of “loud, exciting horn music”?[124] The district court in Metro-Goldwyn-Mayer v. American Honda Motor Co. found the question sufficiently vexing that it rejected Honda’s motion for summary judgment and left the question of infringement to the jury.[125] Certainly, it is possible to rule that Honda infringed upon the copyright in the James Bond character as depicted in the films owned by MGM (though not non-MGM films and not the Ian Fleming novels about James Bond). However, it is also possible to conceive of a court ruling that Honda did not copy anything protected by copyright.

Perhaps the most controversial nonliteral infringements are those that infringe the “look and feel” of the copyrighted work. A “look and feel” infringement, unlike an infringement of nonliteral aspects like the plot of a film, is the most ethereal type of infringement. In cases that find the defendant to have infringed upon the “look and feel” or “concept and feel” of a copyrighted work, the courts suffice with the decision that the infringing work “feels” excessively like the original.[126] Consider, for instance, the case of Roth Greeting Cards v. United Card Co.[127] The case involved two companies that produced greeting cards with simple illustrations and banal greeting card sentiments. Roth complained that United had copied seven greeting cards.[128] The court originally expressed some doubt that Roth’s cards were sufficiently original to warrant copyright—the cards were, after all, greeting cards expressing greeting card sentiments—but the court ultimately ruled that the cards were protected by copyright.[129] When it came to the question of infringement, Roth could not point to anything specific that had been copied, but the court nevertheless reversed a judgment in favor of United and found that United had violated Roth’s copyright by copying the “total concept and feel” of Roth’s cards.[130] The court thought that the combinations of simple pictures and trite messages in United’s cards were so similar to the combinations of simple pictures and trite messages in Roth’s cards that an “ordinary observer” would view the Roth cards as having been impermissibly copied.[131]

Perhaps the most famous “concept and feel” case is Sid & Marty Krofft Television Productions, Inc. v. McDonald’s Corp.,[132] which considered whether McDonald’s had infringed the Kroffts’ copyright in their H.R. Pufnstuf television series when the fast food company produced a series of commercials showing the fantasy “McDonaldland.”[133] While acknowledging that the McDonald’s commercials did not literally copy anything in the H.R. Pufnstuf series, and that they did not copy nonliteral elements such as plot or characters, the Ninth Circuit Court of Appeals nevertheless upheld a ruling in favor of the Kroffts due to the McDonald’s commercial’s similarity to the “total concept and feel” of the H.R. Pufnstuf series.[134] According to the court, it was essential that the target audience was children because this meant the court had to consider “the impact of the respective works upon the minds and imaginations of young people.”[135] The court then determined that notwithstanding the “subjective and unpredictable nature of children’s responses,”[136] an audience of children would necessarily view the commercials as having copied the “total concept and feel” of H.R. Pufnstuf.[137]

Elements of the maddening vagueness of “total concept and feel” cases threaten to leak into other infringement cases due to the standards used by courts to determine impermissible similarity. Consider, for instance, the standard developed by Judge Learned Hand in the 1960 case Peter Pan Fabrics, Inc. v. Martin Weiner Corp.[138] and used ever since in the Second Circuit. Hand explained that the way to determine whether a fabric pattern had infringed on the copyrighted pattern of a different fabric was by asking whether “the ordinary observer . . . would be disposed to overlook [the disparities], and regard [the patterns’] aesthetic appeal as the same.”[139]

Clearly, classifying infringements of nonliteral elements of copyrighted works is not an easy task. Certainly, infringements of this sort are not as culpable as willful and knowing piracy. Yet, often, one cannot view the infringements as inadvertent or innocent. As we explicate further in Part III, we include most of these kinds of infringements within a category we call “standard infringements.”[140] However, in light of the vagueness of the doctrine, there may be extreme cases where the infringement involves nothing more than adoption of the “look and feel” of the copyrighted work. In those cases, we suggest that the infringer should be seen as no more culpable than an inadvertent infringer.

4. Infringements Related to Questions About the Scope of Legal Protection.—Uncertainty in the law lies at the heart of other types of infringements as well. Some infringements are distinguished by questions about the scope of the law’s protection rather than the scope of the work. It is clear, for example, that the actual text of a novel is part of the copyrighted work protected by the law, even if there may be disputes about the parts of the plot that are protected. Nonetheless, not every use of the text constitutes infringement. The fair use doctrine, for instance, gives critics the rights to quote a small portion of the text in a book review without a license from the author.[141] Unfortunately, as with so many other aspects of copyright law, limitations on the scope of copyright are far from obvious, and users of copyrighted works frequently make mistakes.

Consider the extreme case of “appropriation art.” Appropriation art consists of artistic works that use preexisting works with little or no transformation;[142] in the terminology of the Tate Gallery, appropriation art is “the more or less direct taking over . . . [of] a real object or even an existing work of art” “into a [new] work of art.”[143] Appropriation artist Sherrie Levine, for instance, photographs the preexisting photographs of other photographers.[144] Her work After Walker Evans, created and displayed in 1981, consisted of Levine’s photographs of a set of famous Walker Evans photographs of poor rural Americans during the Great Depression.[145] The Evans estate naturally accused Levine of copyright infringement, though the matter was ultimately settled without litigation.[146]

Appropriation artists have often claimed that their works do not infringe on the grounds that the works are “fair uses” of the original. The courts have sometimes accepted and sometimes rejected such claims. Jeff Koons has been repeatedly sued for copyright infringement, and while his fair use claims prevailed in Blanch v. Koons,[147] they failed in Rogers v. Koons,[148] United Feature Syndicate v. Koons,[149] and Campbell v. Koons.[150] The most striking recent ruling on appropriation art was issued by the Second Circuit Court of Appeals in the case of Cariou v. Prince.[151] Richard Prince, the defendant, was an appropriation artist who copied thirty of Cariou’s photographs of Jamaicans and transformed them into Prince’s style “by among other things painting ‘lozenges’ over their subjects’ facial features and using only portions of some of the images.”[152] As one might expect of an appropriation artist, Prince denied any intent for his works to be transformative, saying that “he ‘do[es]n’t really have a message,’ that he was not ‘trying to create anything with a new meaning or a new message.’”[153] Nonetheless, the court found fair use with respect to twenty-five of Cariou’s images used by Prince,[154] while “express[ing] no view as to” Prince’s claim of fair use for the other five images.[155]

Fair use claims are not limited to appropriation art, of course. Fair use claims have been made regarding multimillion dollar parody films,[156] exact copying of software in order to reverse engineer it,[157] advertising campaigns,[158] the compilation of photocopied materials for instructional courses,[159] the copying of television broadcast for later viewing at home,[160] the recording of pieces of music in order to listen to them on different devices,[161] the mass digitization of books,[162] systematic miniaturization and transmission of digital images,[163] the photocopying and archiving of academic articles for future research,[164] the display of a copyrighted image in the background of a television show set for a total of twenty-seven seconds,[165] and the quotation of fewer than approximately 300 words from a 2,000-word book in a news story.[166] Of course, only some of these fair use claims have been successful, but for those unfamiliar with the line of case law, the outcomes are difficult to predict: who would guess, for example, that copying the entire text of five million books would be found fair use,[167] while the quotation of a mere 300–400 words from a 2,000-word book would not?[168]

Fair use is not the only legal tool that may grant users the ability to copy protected works in a manner that would otherwise be seen as infringing. The Copyright Act is replete with mandatory licenses and exceptions granting users the rights to use works in such diverse situations as face-to-face classroom instruction,[169] advertising,[170] and background music in retail outlets.[171] Some of the statutory licenses are straightforward, but many others are hideously complex and have led to repeated litigation over such apparently trivial questions as the square footage of the retail space in which the musical work was being played.[172]

The uncertain boundaries of the many statutory licenses, and especially that of the notoriously vague fair use provision, lie at the heart of many infringement cases. Many users of copyrighted works reasonably believe that their use is permitted by the statutory exceptions to copyright protection. And, indeed, often the courts vindicate these beliefs with a finding of fair use or other statutory privilege.

The culpability of the user in cases where the user improperly relies on statutory exceptions is mixed. On the one hand, the user often knows that the acts she is committing are prima facie within the monopoly rights of the copyright owner, or, at the very least, that they skate very close to the owner’s rights. Google, for instance, clearly knew that when it scanned millions of books for its Google Books project, it was copying many copyrighted works without permission from the copyright owners.[173] Likewise, Aereo, in copying over-the-air transmissions of copyrighted programs and then streaming them to customers, knew that its activities could be characterized as forbidden copying and public performance.[174] Yet, while the users know they are potentially infringing, they also have good reason to believe they are not violating the law. Google had a more than plausible claim that its activity was within the ambit of fair use, while Aereo had a strong argument that its copying was permitted by fair use and its transmissions were not “public” or “volitional.” As it happened, Google’s gamble was successful: its claim of fair use was vindicated first by the trial court[175] and then by the Second Circuit Court of Appeals.[176] Aereo’s gamble, on the other hand, was not. After Aereo prevailed in the First and Second Circuit Courts of Appeals, the Supreme Court ultimately ruled that Aereo’s model was too similar to that of cable television and therefore a forbidden public performance.[177]

In short, the culpability of infringers is best seen as somewhere between the innocence of “subconscious” infringers and the willfulness of pirates. As we explicate further in Part III, we classify most of these kinds of infringements as “standard infringements.”[178]

5. Infringements Related to Questions About the Identity of the Owner.—One of the most unusual sources of infringements is user confusion regarding the ownership of a copyrighted work. Even where one is certain that a work is subject to copyright, and likely still within the protection of the copyright law––as with a piece of music that is clearly of recent origin––one may still not know to whom the rights belong.

Modern changes in copyright law are partly to blame for this situation. Prior to 1978, copyright law required notice of copyright ownership to accompany general distributions of copies of the work to the public, which meant that often one could discover the initial owner of copyright in a work by examining a copy.[179] However, even in the era before 1978, knowledge of copyright ownership was limited. For instance, copyright ownership could be transferred after distribution of the work, but the copies would not reflect this fact. In any event, changes in the law in recent decades have exacerbated, rather than ameliorated, the problem. After January 1, 1978 (the effective date of the Copyright Act of 1976), and especially after the Berne Convention Implementation Act of 1988, the notice requirement was gradually eliminated.[180] Today, there is no notice requirement at all, meaning that a copy of a protected work can bear no indication whatsoever of the identity of the copyright owner, or even the fact that the work is protected by copyright law, yet still enjoy the full protection of copyright law.[181] Additionally, while there are circumstances in which the owner must register copyright ownership, failure to do so is never fatal. A search through the copyright registry may thus provide no indication of ownership, even though the work is protected.

The difficulty of identifying ownership of copyrighted works has become known as the “orphan works” problem. A 2015 report of the Copyright Office defined orphan works based on its 2006 report as “original work[s] of authorship for which a good faith prospective user cannot readily identify and/‌or locate the copyright owner(s) in a situation where permission from the copyright owner(s) is necessary as a matter of law.”[182] While the Copyright Office described the scope of the problem as “elusive,” it acknowledged that the “orphan works problem is real.”[183] The orphan works problem occurs when potential users have encountered a work of authorship of which they wish to make use but cannot figure out to whom it belongs. The situation may arise because no identifying information appears in the copy of the work examined by the potential user or because the information is out of date (due to transfers, etc.). While in some cases the potential user will throw up her hands and avoid using the work, in others, she will take a chance and use the work, risking a future infringement suit.

Much speculation surrounds the topic of orphan works. According to one published estimate, up to 90% of all works currently protected by copyright law are “orphaned.”[184] As technology changes, and the number of digital copies of copyrighted works moves into the trillions,[185] the number of copies without sufficient identifying information will doubtless grow.

Given the hype surrounding the problem of orphan works,[186] it may be surprising to learn that the problem is even worse than usually described. The reason for this is the layered nature of copyrighted works. Almost inevitably, every copyrighted work contains within itself other copyrighted works. A novel, for instance, contains a plot and characters, which can be separately copyrighted and protected. This means that even if a novel is sufficiently well identified to allow potential users to obtain licenses from the owner of the novel as a whole, it will not be entirely clear who owns the component pieces of the novel that the user wishes to license. Thus, for instance, if Alice wishes to utilize a portion of a film whose copyright owner is known to her, she will still have to worry about the question of whether the film incorporates plot elements or characters from a work owned by someone else.[187]

Beyond these basic problems in identifying owners, several copyright doctrines—and in particular, the termination right—further exacerbate the problem of tracing ownership. The 1976 Copyright Act granted authors who transfer their copyright (and, in limited cases, other copyright owners)[188] the right to cancel the transfer several decades later and reclaim ownership in the work.[189] The 1976 Act’s termination right applies retrospectively as well as prospectively[190]: transfers that took place before the effective date of the 1976 Act can be terminated as well. Additionally, while the 1976 Act’s termination right was new, even under older copyright law, there were instances in which a transferee might lose ownership of a copyright that had been lawfully purchased. Under the 1909 Act, for instance, copyright terms were compound; that is to say, creators of a copyrighted work would get two distinct terms of protection—an initial 28-year term and an additional 28-year term if the renewal paperwork was properly filed.[191] A peculiarity of case law under the 1909 Act provided that a creator could sell all her rights in a copyrighted work (for the full 56 years), but if she died before the onset of the renewal term, the transfer of the second 28-year term would fail, and ownership would revert to the heirs of the original owner.[192] In other words, under the old law, as well as the new, even where someone “owns” all of a copyright, there might still be someone else lurking about who has a residual right to take it away.

As a result of all these factors, copyright “clearance” is extremely difficult and sometimes impossible. Despite the best efforts of a user, it may not be possible to figure out from whom to license a work.

In general, there is a large category of cases in which users wish to use works without having any clear idea of the identity of the owner or how to go about getting permission. If the frustrated user takes the chance of using the work without a proper license, she is clearly not an inadvertent infringer. However, just as clearly, it does not make sense to treat her as culpable to the same degree as a willful infringer. We suggest that these cases belong under the category of “standard infringements.”

6. Infringements Related to Questions About the Scope of a License.—It might seem odd to suggest that the scope of licenses can constitute a fertile source of mistaken infringements. One might suspect that the licensing of copyrighted works is the quintessential sign that no infringement has taken place, or that if a work has been infringed, it is due to malice. After all, if the right to the copyrighted work has been licensed, it would seem necessary for the user and the owner to have reached some meeting of the minds as to how the copyrighted work will be used. Yet, in practice, it turns out that even where users hold a license, infringements due to mistakes about the scope of licenses are legion.

Mistakes about the scope of licenses are more frequent than one might expect. For example, in 2014, Getty Images released 35 million copyrighted images from its catalog to the Internet, making them available, free of charge, for noncommercial use.[193] In exchange, Getty required users to embed the images with Getty’s software embedding tool in order to ensure that proper credit is given and to monitor users and show them ads.[194] Users, however, are often unaware of the existence of limitations and restrictions on the use of content or simply miscomprehend them. Furthermore, they are sometimes implicated in infringement activities by trusted third parties, such as web designers.[195]

A recent empirical study by Hong Luo and Julie Holland Mortimer of infringements involving digital images reveals that “infringement of digital images is often unintentional. Unintentional infringement may arise either through misinformation about licensing obligations, or because third parties infringe on behalf of a user (e.g., a web designer includes an image on a firm’s website).”[196]

Specifically, Luo and Mortimer report that 85% of infringers claimed that they were unaware of the infringement or that it arose as a consequence of the actions of hired third parties.[197] The percentage among small firms was even higher and reached 88%.[198] Interestingly, they also note that the price requested by copyright owners for the use of the content, i.e., the license fee, had no effect on the level of infringement.[199]

The ambiguity of licenses is compounded by the long-lasting nature of copyrights. In the course of the century and more that copyrights can last, technology can change, rendering the meaning of earlier copyright agreements uncertain. Does a distribution license for a film include the possibility of distribution by video cassette or streaming by internet?[200] Do publication rights for a book include the right to publish it electronically?[201] Such questions have proved a fertile source of litigation.

A further degree of confusion is added by the case law permitting unwritten implied licenses. The Copyright Act requires an attestation in writing for all “transfer[s]” of copyright rights,[202] but the writing requirement does not apply to the granting of nonexclusive licenses.[203] In such cases, the behavior of the parties may create a license whose terms are unknown and unknowable. Consider, for example, the case of Effects Associates, Inc. v. Cohen.[204] Cohen, “a low-budget horror movie mogul,” asked Effects Associates, a special effects studio, to produce special effects footage for the movie The Stuff.[205] While the studio produced the footage as promised, and Cohen used the footage in the movie, Cohen refused to pay the agreed-upon consideration.[206] According to the Copyright Act, transfers of copyright ownership require a written agreement.[207] Cohen had obtained no such written agreement.[208] Effects Associates sued for copyright infringement and fraud.[209] In a surprising decision, Judge Kozinski ruled that there was no copyright infringement since Cohen had an implied nonexclusive license to use the disputed footage in his movie.[210] While Effects Associates still had the right to payment according to the contract, it could not object to the use of its copyrighted footage.

Not only may licenses be created by behavior; they may also be rescinded by the same. An interesting illustration is provided by the recent case of Garcia v. Google Inc.,[211] in which an actress named Cindy Lee Garcia sought to enjoin the distribution on YouTube of the controversial film Innocence of Muslims on the grounds that she owned a copyright in her acting performance in the relatively small part in which she appeared in the film.[212] Garcia had signed a contract before acting in the film, but the contract had not explicitly resolved the issue of copyright in Garcia’s performance and she argued that any license she might have granted was invalid, as it had been obtained by fraud (she had been misled about the nature of the film and the way her performance was to be used).[213] Ultimately, the Ninth Circuit Court of Appeals ruled en banc that Garcia enjoyed no copyright in her performance, rendering the question of license moot.[214] However, the different court opinions entertained various theories of implied license, including the possibility that Garcia had granted an implied license to the filmmakers and that the filmmakers had granted Garcia an implied license to perform the copyrighted screenplay.[215]

The ambiguity of written licenses together with the possibility of implied licenses creates a high likelihood of mistaken infringement due to misinterpretation of licenses. The user who violates the license after misreading or misunderstanding it clearly bears some responsibility for her infringement, but it does not make sense to treat her as culpable to the same degree as a willful infringer. We suggest that these cases belong under the category of “standard infringements.”

II. Copyright’s Liability Regime

As we have seen, copyright infringements involve different degrees of blameworthiness and impose varying degrees and types of harms on copyright owners. Nonetheless, the Copyright Act has a single standard of liability and a single standard compensatory scheme.

Liability under the Copyright Act is predicated on a strict liability standard. Neither the state of mind of the infringer nor her cost of avoidance matter for liability determinations. The roadmap for finding infringement, according to the Copyright Act, is simple and strict. Section 106 of the Act reserves to the owner of a copyright the exclusive right to perform certain actions, such as copying the work.[216] Section 501 of the Act, in turn, defines infringement as “violat[ing] any of the exclusive rights of the copyright owner.”[217] For courts, this language means that a person becomes an infringer if he or she performs one of the actions regarding the copyrighted work (such as copying) that the statute reserves for the owner, without a proper license from the copyright owner or from the statute itself.[218] Other than to be sure that the infringer actually knew of and somehow used (in a manner protected by § 106) the protected work, courts do not interest themselves in the state of mind of the infringer. Indeed, as Nimmer writes, “[r]educed to most fundamental terms, there are only two elements necessary to the plaintiff’s case in an infringement action: ownership of the copyright by the plaintiff and copying by the defendant.”[219]

Once infringement is established, the Copyright Act imposes a single standard compensatory scheme. Under § 504, all infringements that give rise to liability under § 501 entitle a copyright owner to recover her actual harm or the infringer’s profits—at her choice.[220] Likewise, all infringements entitle the plaintiff to seek injunctive relief.[221]

The only exception is statutory damages. If a copyright owner wishes to receive statutory damages in lieu of actual ones, courts must consider the blameworthiness of the defendant in setting statutory-damages awards.[222] Specifically, the Copyright Act sets forth three different categories of infringements—innocent, standard, and willful—and establishes a different recovery range for each category by varying the floor and the cap of the damages for each of the categories.[223]

In this Part, we explore this cookie-cutter liability regime by looking first at the strict liability standard for establishing whether an infringement has occurred; second, at the standardized remedies that apply to all infringements; and finally at the harm caused by standardized remedies. We also explore the unusual cases in which extant law permits tailoring remedies to the defendant’s culpability.

A. The Liability Standard

Liability under the Copyright Act arises whenever a person performs one of the exclusive acts granted to authors—namely, reproduces a work, adapts it, distributes it to the public, publicly performs or displays it or digitally performs it (in the case of sound recording), or authorizes such an act without permission from the rightsholder.[224] It is well established that the Act adopts a strict liability regime.[225] Acts are the touchstone of liability under the Act. The state of mind of the infringer is irrelevant.

It was not always like this. As Anthony Reese points out, historically, the Copyright Acts exempted innocent or unknowing infringements from liability.[226] He explains that “[t]he copyright system originally made most types of innocent infringement easily avoidable, and where innocent infringement was difficult to avoid the imposition of liability in fact depended on a defendant’s culpable mental state.”[227]

The 1976 Copyright Act, however, omits reference to the mental state of the infringer, and it does not reference the infringer’s degree of culpability. The Act’s definition of infringements is laconic; it simply states that any violation of an author’s exclusive right triggers liability.[228] The courts’ interpretation of the statutory language, however, premises liability on the purely objective standard of “copying in fact.”[229] This is not an accident. As Shyamkrishna Balganesh points out, “[d]ating back to its origins, copyright law has operated principally by granting its holder the exclusive right to copy a creative work of authorship, and actions for copyright infringement have ever since revolved entirely around a showing of copying.”[230]

Practically, as construed by the courts, liability does not even require positive proof of copying. Courts will suffice with circumstantial evidence of copying, relying on proof of “access” combined with “similarity.”[231] Concretely, a copyright owner who sues for infringement needs to show that the defendant had access to the putatively infringing work, in the sense that she was exposed to it or was likely to be exposed to it,[232] and that the defendant’s work bears similarity to that of the plaintiff.[233] Access and similarity, in other words, suffice as a replacement proof of actual copying. Once these twin elements have been established by the plaintiff, the defendant can defeat the presumption of copying by raising and proving the defense of “independent creation.”[234] And, while it is true that if a defendant can show that she did not copy her work from the plaintiff she should prevail in court,[235] a mere proof of access and similarity by a copyright holder suffices to shift the burden of proof to the defendant, who must show lack of copying. Failure on the part of the plaintiff to lift this burden will result in a finding of copying.

In fact, courts have construed liability under the Copyright Act so broadly as to cover cases where the court was satisfied that the infringer never consciously copied anything, as in the infamous case of Bright Tunes Music v. Harrisongs, mentioned previously.[236] In that case, as we noted, the court believed George Harrison’s testimony that he had not consciously copied anything from The Chiffons’ “He’s So Fine,” but the court still found the similarity so overwhelming as to support a finding of infringement.[237]

While casting the net of infringement widely, the statute recognizes only one type of infringement. The statute does not provide for degrees of infringement or for other categorizations of infringements. All infringements—whatever the circumstances giving rise to the violation of copyright rights—are seen by the statute as the same.

B. Remedies for Infringement

The civil remedies for infringement are laid out by § 504 of the Copyright Act, and for the most part, they are not differentiated by culpability. Monetary damages for an infringement come in three categories: the copyright owner’s “actual damages,” any “additional profits” of the infringer, and “statutory damages.”[238] In addition to the monetary damages, courts may award injunctive relief as “reasonable to prevent or restrain infringement,”[239] and it may order impoundment and destruction of infringing copies as well as articles by means of which the infringing copies could be produced.[240] In one very narrowly tailored set of circumstances the court can award a punitive “additional award of two times the amount of the license fee that the proprietor of the establishment concerned should have paid the plaintiff for such use during the preceding period of up to 3 years.”[241] Finally, the court may award costs and attorneys’ fees.[242]

The statute fails to provide clear guidelines for measuring the central building blocks of monetary damages—actual damages and lost profits. Actual damages are defined simply and unhelpfully as “the actual damages suffered by [the copyright owner] as a result of the infringement.”[243] Lost profits are defined, only slightly more helpfully, as “any profits of the infringer that are attributable to the infringement and are not taken into account in computing the actual damages.”[244] Section 504(b) adds the useful evidentiary rule that in calculating profits, gross revenues, as proved by the plaintiff, are presumptively equal to profits, and it is up to the defendant to attempt to reduce this amount by proving that the revenues were the result of something other than infringement, or that there were relevant deductible expenses.[245] Both heads of damage—actual damages and lost profits—are available in all cases of infringement, irrespective of culpability, unless the plaintiff elects to take statutory damages in their place.

It is only in a handful of instances that the Copyright Act looks in the direction of culpability.

First, there is a narrow punitive damages provision that relates to § 110(5) of the Act.[246] This exceedingly narrow provision is not easily explained. Section 110(5) of the Act, in relevant part, states that businesses may perform (“communicat[e]” to the public, in the language of the statute) nondramatic musical works subject to a variety of conditions.[247] Essentially, § 110(5) allows retail businesses, as well as bars and restaurants, to play the radio or television for their customers without getting a separate license from the owners of the copyrighted works being shown or played. Section 110(5) imposes very detailed and strict conditions on the businesses: they can only use the exemption if the business falls within certain size limitations (for instance, for an establishment other than a food service or drinking establishment, 2,000 gross square feet of space not including customer parking)[248] and uses certain equipment (for example, an establishment other than a food service or drinking establishment may use not more than four audiovisual devices, no more than one per room, and none with a diagonal screen size greater than fifty-five inches).[249] Section 504(d) states that where a defendant business owner claims that the infringement was justified by § 110(5), but the court concludes there were no “reasonable grounds” for the owner’s believing that the § 110(5) exemption applied, the court should add an additional, punitive damage award to any other damage award given by the court.[250] The punitive damage award is set by a fixed formula: “two times the amount of the license fee that the proprietor of the establishment concerned should have paid the plaintiff for [the nonexempt performances] during the preceding period of up to 3 years.”[251] It is difficult to know what to make of this narrow punitive damages provision or to understand why this one case warrants such exceptional treatment.

The second instance in which the Copyright Act adjusts remedies according to culpability is broader but limited solely to statutory damages. Statutory damages are available to plaintiffs only in lieu of actual damages and lost profits.[252] However, statutory damages may also be awarded punitively by the court in addition to actual damages and lost profits in cases where the “infringement was committed willfully.”[253] The Copyright Act generally provides that statutory damages may be elected by a prevailing copyright owner in the amount of $750–$30,000 (as the court deems just) in lieu of the actual damages and profits related to infringements of any work.[254] However, if the owner proves the infringement was committed willfully, the statutory damages may be increased to $150,000.[255]

This is the clearest use of culpability in the civil section of the Copyright Act,[256] but almost no guidelines are provided to courts in deciding what is willful and how to choose damages. In fact, while the Act provides that certain provisions of false information can create a rebuttable presumption of willfulness, it explicitly notes that this presumption should not be read more broadly to “limit[] what may be considered willful infringement.”[257] This lack of clarity has led scholars to criticize the statutory-damages provision as “frequently arbitrary, inconsistent, unprincipled, and sometimes grossly excessive” as applied by courts.[258]

The third and final instance in which civil remedies are adjusted for culpability similarly relates to statutory damages but operates in the opposite direction. Where courts find an infringer was innocent (i.e., the “infringer was not aware and had no reason to believe that his or her acts constituted an infringement of copyright”), the statutory damages may be reduced to as little as $200.[259] In some very narrow cases of infringement (an infringer who believed the use fell under the category of fair use, who was acting on behalf of a nonprofit educational institution, library, or public-broadcasting entity, and who also meets other restrictive conditions), the court is instructed to remit statutory damages.[260] The narrow remission provision is difficult to explain, but the broader structure of innocent infringement also appears without statutory guidance.

C. What’s Wrong with Uniform Standards of Liability

As we demonstrated, copyright doctrine contains a high degree of uncertainty. It is easy for potential users of a work to mistakenly infringe due to misunderstandings about the legal protections afforded works or the facts surrounding the work, such as ownership. At the same time, outside the arena of statutory damages, copyright remedies apply uniformly across the many types of infringements.

The uncertainty inherent in copyright law undermines its utilitarian goal of promoting expressive creativity. Ideally, the copyright system should be designed to maximize total expressive output at the lowest possible cost. This means that while copyright law should grant copyright owners the legal tools to force potential users to pay for licenses to use the works, the law should also aim to reduce costs of transacting about rights and enforcing them, as well as to reduce the effect of copyright-infringement actions in unnecessarily deterring follow-on creation. It is particularly important, therefore, that the Copyright Act adopt legal directives that minimize information costs to potential users of copyrighted works and, in particular, ameliorate the potential for overdeterrence that accompanies legal uncertainty.

As several leading law and economics scholars have pointed out, ambiguous legal doctrines invariably result in overdeterrence, which, in turn, prompts defendants to invest excessively in precautions.[261] The reason is simple. When facing a vague standard, actors cannot know for certain how much to invest in precautions taken in order to avoid liability. But they know that there is no symmetry between under- and over-investment. Underinvestment in precautions exposes one to the full brunt of civil (and criminal) liability. Overinvestment, by contrast, buys one immunity. Hence, under conditions of uncertainty, a rational actor would always choose to err on the side of safety and overinvest in precautions. Accordingly, the ambiguity of copyright law works as a one-way ratchet that will in many cases lead to the under-use of copyrighted works.

Users of copyrighted expression can respond to the vagueness of copyright law by adopting two types of precautions. When information and transaction costs are sufficiently low,[262] and users are aware of the copyrighted nature of the material they wish to use, users may attempt to secure a license from the copyright owner.[263] When information and transaction costs are high,[264] users will copy less protected expression than they are legally entitled to or refrain from using expression altogether. Importantly, the latter will lead users to forgo the use of not only legally protected copyrighted expression but also unprotected expression. This is because users will not always know whether expression is protected, but the cautious strategy is to respond to the lack of knowledge by avoiding use.

Both responses are socially wasteful.[265] It is critical to understand that under extant law, the second response probably dominates the first. Given the volume of copyrighted content and the lack of a cost-effective way to readily identify protected content and its proprietors, it makes more sense from a purely utilitarian standpoint to abstain from creating. Given the informational haze that engulfs the world of copyright law, licensing only gets one very limited insurance against liability. A potential creator may scour the domain of copyrighted content and identify several rightsholders, whose copyrights may be infringed by her creation. Securing licenses from the relevant rightsholders will not bar lawsuits from other rightsholders of whom our potential author was not aware, or even from rightsholders identified by the potential author but whom the potential author neglected to contact because she mistakenly resolved that her work does not infringe upon the copyright in theirs. Moreover, obtaining a license is not a guarantee that the license actually covers the relevant use or that the user has properly identified the copyrighted materials.

The staggered remedies we propose can reduce these harmful effects in two ways. First, reducing remedies due to the user’s lack of familiarity with information at the disposal of the owner creates a welcome dynamic in which owners are incentivized to push information about their copyrighted works and licensing schemes to the public. Second, reducing remedies due to the user’s misunderstandings, whether they stem from uncertainty about copyright law or about knowledge the owner has about the work, reduces the deterrent effect of copyright-infringement remedies upon use of copyrighted works. This benefits society by encouraging socially beneficial use of copyrighted works and by reducing the costs of enforcing copyright law.

III. A New Liability Regime

As we established in Part I, copyright infringements come in many different varieties. Yet, as we have shown in Part II, the law treats all of them in the same way for purposes of assigning liability and fashioning relief, with one exception: suits for statutory damages. More importantly, we established that the uniform treatment of all infringements, other than for statutory-damages purposes, is at odds with the goal of promoting expressive creativity. In this Part we offer a blueprint for legal design that takes account of the differences among various types of infringement and is therefore better suited for advancing the goals of the copyright system.

Our reform plan consists of two interlocking steps. First, we aim to extend copyright law’s classification of infringements for purposes of statutory damages to all relief in the realm of copyright. Concretely, we call for the classification of all infringement cases into three categories: inadvertent infringements, standard infringements, and willful infringements. These three categories of infringements should serve as the basis for all remedial action ordered by courts in infringement cases. Second, we specify a menu of remedies that should apply to each infringement category. Under our proposal, the classification of the infringement as inadvertent, standard, or willful will determine the monetary and injunctive relief available to copyright owners. Naturally, the remedial options in cases of willful infringements will be broader than in cases of standard infringements. The same principle holds true for the relationship between standard infringements and inadvertent infringements. Also, our proposed regime, while harnessing the classifications used for statutory damages, places some limitations on the availability of statutory damages. Specifically, while seeking to retain the availability of statutory damages in order to save the administrative cost of proving actual loss, we limit the ability of statutory damages to serve any other function (such as serving as punitive damages) in any case other than a willful infringement.

A. Classifying Infringements

The first step in our proposal is to distinguish among three prototypes of infringements: inadvertent, standard, and willful.

The first category of “inadvertent infringements” covers all those cases in which the infringer was unaware and could not have reasonably become aware of the infringing nature of her activity. Infringements falling into this category involve no culpability on the infringer’s part and, furthermore, they could not have been avoided cost-effectively. Accordingly, the law should not seek to deter such infringers. Quite the opposite, the concern that should guide lawmakers should be to limit overdeterrence, which may result in the dampening of incentives to create. Inadvertent infringers, by our definition, are actors who commit copyright infringements in complete good faith while displaying respect to the copyright system. They are themselves “victims” of the informational uncertainty that engulfs copyright law. Hence, in cases of inadvertent infringement, the law should aim to avoid unduly deterring the violator.

The second category of “standard infringements” consists of all infringement cases in which the infringer had a reasonable basis to believe that her activity may qualify as noninfringing or that she had a good-faith defense to liability. This category is designed to address those instances in which an actor willingly assumes a calculated risk in deciding to use the protected work of another. The user in such cases is aware that if her use is challenged in court, it may be ultimately ruled an infringement. Yet, she also harbors a reasonable belief that her use should be classified as noninfringing or fair. In other words, infringers falling in this category engage in a reasonable construction of the copyright doctrine, as applied to the facts of their use. If their reasonable construction is accepted by the courts, they escape without liability. However, if the judge (or jury) disagrees with them and finds them guilty of a copyright infringement, the classification of infringement must account for the risk taken by the actor that ultimately did not pay off. The reasonableness standard we proffer is similar to the reasonableness standard used in the law of torts, in that it encompasses a subjective and an objective component. Therefore, it is not enough that the infringer reasonably believed that her activity should not be considered infringing; she also has to demonstrate that a reasonable person in her situation could harbor such a belief.

Standard infringers, unlike inadvertent infringers, bear a certain degree of culpability: after all, by definition, they have taken a calculated risk, knowing full well that a court may find them guilty of a copyright infringement. Furthermore, they could have secured permission from the relevant rightsholder to use the copyrighted content they utilized but elected not to do so. Yet, they did not act in willful disregard of the rights of the owner of the copyrighted work. Rather, they chose to cope with the uncertainty inherent in copyright law in a reasonable way. What should be the legal attitude to such infringers? In light of the danger of overdeterrence that inheres in vague legal standards, we think it reasonable to ensure that whatever damages are paid for the copyright infringement do not take on a punitive character. Recall that the goal of copyright law is to foster expressive creativity. If copyright laws were crystal clear, there would be no reasonable miscalculations. Alas, this is not the case. The meting out of excessively harsh sanctions for standard infringement would likely deter future creativity by excessively blocking unlicensed (but arguably lawful) uses of existing copyrighted works.

The third, and final, category of “willful infringements” covers all those cases involving blatant disregard of copyright rules. Infringements coming under this heading consist of knowing infringements in which the infringer had no reasonable basis to believe that she was not infringing or was sheltered by a defense. Willful infringements present a very different challenge to lawmakers than do inadvertent and standard infringements. It is not only that willful infringements display the highest degree of culpability. Just as importantly, there is no reason to worry about excessive deterrence of willful infringements. The copyright system incentivizes expressive creativity by securing to authors exclusive rights in their works for a limited time. The period of exclusive rights is intended to incentivize authors by giving them an extended opportunity to profit from their works. The statutory mechanism depends on the formation of markets for copyrighted works in which authors can transact with buyers. Willful infringers bypass the market and thereby deprive authors (or copyright owners) of the ability to profit from the infringing uses. By so doing, they not only deprive the author of revenues to which she is entitled by law but also destabilize the very premise of the copyright system by eating away at the authors’ incentives to create.[266]

Our proposed categorization of infringements for liability purposes is summarized in Table 1 below:

 

B. Tailoring Remedies to the Class of Infringement

The second step in our scheme is to match a remedial menu to each infringement category. This is the more radical component of our proposal. Yet, it is a necessary complement to the categorization we proposed above.[267]

1. Inadvertent Infringements.—The remedial menu for inadvertent infringement will consist almost exclusively of compensatory damages for actual losses suffered by copyright owners. Under our proposal, successful plaintiffs in inadvertent infringement cases would not be able to receive statutory damages or recover the inadvertent infringer’s profits. Nor would they be able to impound infringing copies or the equipment used to produce them. In very rare cases, however, we would allow the monetary relief to be supplemented by injunctive relief.

As we explained, the main policy goal in cases of inadvertent infringement is to avoid overdeterrence. Hence, it is critical to tailor the remedial menu such that it would afford copyright owners an opportunity to collect full compensation for the losses they suffered as a result of the infringement, without possibility for windfalls. For this reason, we leave open the possibility of standard compensatory damages but take the options of statutory damages that do not depend on proof of harm and even recovery of profits off the table. In our view, cases like Harrisongs and Williams would have been decided differently as far as the remedial aspect is concerned. Both The Chiffons (or more accurately, their record label) and the Gaye Estate would have had to suffice with compensation for the actual loss they suffered. Neither Harrison nor Thicke and Pharrell would have been required to forsake their profits.

The only other remedy copyright owners would be able to obtain against inadvertent infringers would be injunctive relief, but only in rare cases. In order to avoid abuse of injunctive relief, we would divide inadvertent infringements between two kinds of cases. Where the inadvertent infringement consists of using the protected work in a manner that does not involve any further creative work—for instance, the distribution of a book that turns out still to be under copyright protection—injunctions should be readily obtained. Courts should be able to block further distributions of the copyrighted work since no new content is implicated. By contrast, where the infringement is incorporated into a new work, injunctions should never be granted. Importantly, this new work need not reach the level of a derivative work, such as the incorporation of a protected book into the screenplay of a new film. Rather, any new content should be enough to block an injunction. Given the fact that the user will have to prove that the use of the copyrighted work was inadvertent, it will be difficult to add new content strategically simply in order to avoid injunctions.

2. Standard Infringements.—The category of standard infringements calls for a more expansive remedial menu. As we explained, in this case, overdeterrence is still a concern, but it is a weaker concern relative to the case of inadvertent infringements. Recall that standard infringers, by our definition, made a calculated decision to open themselves up to a liability finding. As we made clear, allowing this choice to serve as a bar to recovery would eviscerate copyright protection. And it must be borne in mind that the calculated decision made by standard infringers imposes a cost on copyright owners, who played no part in the decision. In light of this fact, copyright owners who prevail in an infringement suit against a standard infringement should be able to take advantage of a richer remedial menu. At the same time, standard infringements still place a heavy burden on users, requiring them to know information that may be uniquely at the disposal of owners, or requiring them to take excessive precautionary steps to avoid overstepping legal bounds. Thus, the remedies should not reflect the full extent of the copyright remedy’s punitive potential.

Concretely, we propose that successful plaintiffs in copyright infringements should be entitled to seek compensatory damages for their actual losses and restitution for lost profits. As far as restitution of profits is concerned, courts should not rush to award successful plaintiffs the full amount of the infringer’s profits, and should try to fashion, instead, fair profit-sharing arrangements that would reflect the respective contributions of the parties and maintain their mutual interest in the successful distribution of the infringing work in those cases where the infringing work contains original authorship.

Our scheme would also allow successful plaintiffs in standard infringement cases to substitute statutory damages for actual damages. However, statutory-damages awards in standard infringement cases would be available only for compensatory and not punitive purposes. Thus, defendants against whom statutory damages are sought would be entitled to prove that statutory damages exceeded actual losses suffered by the rightsholder and reduce the statutory damages accordingly.

Finally, our proposal would allow for the grant of injunctive relief to successful plaintiffs in standard infringement cases. The reason injunctive relief should be granted in some cases is that ongoing infringements can cause great harm to the financial interest of copyright owners, and ex post compensation cannot adequately address it. To illustrate this risk, assume that a user erroneously believes she has a license from the copyright owner to distribute her work on the Internet for free. To make the matter more concrete, assume that the work at issue is the next season of Game of Thrones. On these facts, the grant of injunctive relief to the copyright owner is warranted. The continuous, unauthorized distribution of the series causes great harm to the copyright owner and compensation after the fact may not be a viable option. Hence, courts ought to have the option of awarding injunctive relief in appropriate cases but must be circumspect in their use of this option.

However, we argue that injunctive relief should be used sparsely and not be granted as a matter of course by courts. Rather, as in the case of inadvertent infringements, courts should reserve injunctions to those cases in which monetary damages do not adequately compensate the plaintiff. It is important not to use injunctive relief excessively since its use not only affects the defendant but also deprives the public of the ability to use the disputed work.

3. Willful Infringements.—As for the third category, that of willful infringements, here, overdeterrence is not a paramount concern, at least not one that should lead to the removal of any of the remedial options that are currently available under the Copyright Act. It should be remembered that willful infringers not only harm the incentive to create of the individual authors whose work they copy but also threaten to undermine, via their behavior, the very foundation of the copyright system. In the case of willful infringers, therefore, rightsholders should be entitled to the full panoply of civil remedies available under the Copyright Act, as well as criminal sanctions.

Specifically, in cases of willful infringements, courts should have the discretion to award compensatory damages, restitution of lost profits, high awards of statutory damages, as well as to grant successful plaintiffs injunctive relief and impoundment orders. By our lights, criminal sanctions are also appropriate in this case. This is not to say that willful infringers should bear the full brunt of the law in each and every case. The particular relief granted in individual cases ought to be tailored to the particular circumstances of the case, including the actual loss of the copyright holder, the profit of the infringer, and the level of culpability of the defendant. Yet, in fashioning the remedial options for the category of willful infringements, we see no reason to scale back the list that exists today by removing certain categories of remedies.

Our proposed remedial menus are summarized in Table 2, following:

IV. Comparing Reforms

A. Administrative Costs of Reform

To this point, our discussion has featured two central elements. First, we have analyzed the vast array of types of infringement.[268] Second, we have divided these many infringements into three categories based on the degree of culpability of the infringer,[269] and we have proposed a remedial menu tailored to each of the proposed categories.[270]

At this point a question should naturally present itself: If there are so many different kinds of infringement, why do we limit our proposed reform in remedies to merely three new categories? Why not five, or a dozen? Or for that matter, why not have infinitely tailorable remedies, depending on the type and degree of culpability, as judges or juries may find them in any given case?

To answer this question, it is important to return to the premises of our analysis. Our analysis is frankly and openly pragmatic. While we recognize that liability schemes based on culpability are intuitively appealing for reasons of fairness, we do not presume that there is an ontological “wrong” to using rights that copyright law assigns to another. We do not suppose that we have the tools to measure the precise degree of moral wrong (if any) committed by scanning and reprinting without permission a pamphlet first published by a since-deceased author in 1980 and the difference in moral culpability of the same action if the pamphlet were first published by a since-deceased author in 1930. Our focus on culpability is not tied to a broader claim about the moral rightness or wrongness of any particular violation of the legal rules of copyright.

Rather, our focus on blameworthiness and remedies as the two key variables of our proposed reform stems from practical reasons. Specifically, by focusing on these two elements, our scheme provides a ready means to reduce overdeterrence and to incentivize the provision of further information without incurring excessive administrative, litigation, and other transaction costs. Our proposal can be readily implemented without the need for new legislation and without radically transforming the evidentiary or judicial burdens already produced by copyright-infringement cases.

The reason for our confidence that our proposal can be implemented easily is the importance of the blameworthiness of the alleged infringer to copyright litigation as it stands today. While culpability is not a part of the ordinary infringement claim, plaintiffs in copyright cases routinely seek statutory damages; indeed, in 89% of cases the plaintiff requests statutory damages.[271] Since the law dictates that statutory damages be keyed to culpability, in the overwhelming majority of copyright cases, plaintiffs must already bring evidence of the defendants’ culpability; defendants seeking to reduce statutory damages must bring their own evidence; and the courts—judges and juries alike—must already pay careful attention to questions of blameworthiness.[272] Courts, in other words, are no stranger to determinations about culpability; on the contrary, they are well versed in making them. This is not to say that courts always analyze and decide blameworthiness correctly. Professor Pamela Samuelson has pilloried the inconsistencies in courts’ statutory-damages rulings.[273] We do not dispute her findings. Whenever the law divides liability into categories, it runs the risk of errant classifications. Laws nevertheless often eschew uniform liability in favor of categorizations in order to calibrate deterrence and other aims of the law. We suggest no more.

The second prong of our proposal—to attach a predetermined remedial menu to each form of infringement—can also be readily effected. We do not call on courts to adopt new remedial forms. All we ask them to do is to limit the remedies awarded to successful plaintiffs in cases of inadvertent and standard infringements—willful infringers will continue to bear the full brunt of all the remedies available under the Copyright Act, as they do today.[274] As we explained, this aspect of our proposal is appealing not only for practical reasons but also for normative ones. The remedies that copyright owners can receive determine not only the extent of their compensation but also the level of deterrence experienced by users, who often cannot ascertain their use privileges with respect to copyrighted content.[275] Capping the potential risk of liability to which users are exposed is likely to increase the use of expressive content and ameliorate the problem of overdeterrence.

Our adoption of three classifications of culpability and our assignment of the question to the remedy stage thus closely matches existing patterns of litigation. Thanks to the extant structure of statutory damages, courts and litigants are already accustomed to arguing for the assignment of culpability to one of three categories and to making culpability arguments and determinations. Against the baseline of extant law, our proposal is a low-cost way of introducing greater calibration of liability and thereby better tailored deterrence against copyright infringement.

B. Alternative Reforms

To be sure, copyright law must be aimed at more than merely reducing the costs of administering the law. Our proposal is a rough attempt at calibrating deterrence and incentivizing the production of information without significantly raising the costs of administering copyright law. It is not the only possible balance, and we do not claim any empirical proof that ours is the optimal balance. We do not claim to have found the data that permit us to choose, from the vast universe of alternative possibilities, the right balance for calibrating incentives without increased administrative costs swallowing the gains. Thus, in the remainder of this part, we discuss potential lines of research that might shed some light on the optimal balance of deterrence, administrative costs, and incentivizing production of information. We do not claim that our discussion exhausts the full range of reform possibilities. Our aim, rather, is to chart paths for future research.

1. More Categories.—A natural extension of our scheme is to create additional categories of liability. The number of categories need not be limited to three. The more categories one adds, the more justice will be done to the complexity that has become the hallmark of the digital copyright era. For example, our category of standard defendants can be divided into the well-known categories from the law of torts—namely negligent, grossly negligent, and reckless defendants.[276] This would create a legal field of five categories of infringement. One need not stop there, of course. It is possible, as a thought experiment, to conceptualize a regime of copyright liability that takes the form of a continuum, rather than discrete categories. The continuum of moral blameworthiness of defendants (or righteousness of copyright owners) would be examined by the judges (or jury) who would then base their remedial determinations on the precise degree of culpability of the defendant.

The obvious downside of creating many categories of liability, let alone of creating a continuum model, comes in the form of lengthier and more costly adjudicative decision-making, as well as associated error costs and litigation costs.[277] The more fine-grained a liability system one adopts, the more difficult it will become to draw the lines between categories with a real degree of precision. Furthermore, such a system would induce parties to expend resources on convincing courts to put them in a more lenient category if they are defendants and a harsher category if they are plaintiffs. Hence, the number of categories cannot be infinite. Grading has its costs and at some point they will outweigh the benefits. This is not to say, of course, that the optimal number of categories is three, as it is under our proposal. The virtue of our proposal is that it tracks the same categories that already exist under the Copyright Act in the context of statutory damages, and thus it minimizes transition costs. The adoption of our proposal, in other words, would not require judges to educate themselves or adopt a new set of skills. But we cannot a fortiori rule out the possibility that the long-term benefits of a more finely grained liability scheme that includes more liability categories would not outweigh the aggregate costs associated with adopting it.

2. Per Se Liability.—A different way of reforming copyright liability draws its inspiration from antitrust law. Readers familiar with antitrust jurisprudence know that antitrust violations are divided into two general categories: per se violations and rule of reason violations.[278] An example of a per se violation is horizontal price fixing.[279] An example of a rule of reason violation is the adoption of a resale price by a provider.[280] The distinction between per se violations and rule of reason violations tracks the distinction between rules and standards. Per se violations are an example of a legal rule in that they provide clear and advance definitions of activities that are categorically illegal. Rule of reason violations represent legal standards, as they require post hoc analysis of the precise effect of commercial activities.[281]

As Chris Sprigman has suggested, it is possible to introduce a similar distinction into copyright law.[282] For instance, one could argue that all cases of literal copying, bodily appropriations of copyrighted content, should be made per se illegal, while pattern copying, nonbodily appropriations consisting of the taking of concept and feel of another work, should be subject to a rule of reason. The central advantages of adopting such a system are certainty and judicial cost savings. Placing literal copying under the category of per se infringement would result in a simpler liability system than that which we have today, as literal copiers will be barred from raising defenses. This, in turn, would also economize on judicial resources. The problem with this proposal is that it may lead to inequitable and inefficient results in certain cases. Literal copying may serve important social causes. It may improve public welfare by making important content available to the public, it is essential to the creation of parodies, and as Jeanne Fromer and Amy Adler recently established, it may represent a form of self-help for artists whose works have been appropriated by others.[283]

A variant of the proposed regime will not make any act of literal copying per se illegal, but only literal copying of 100%; i.e., full bodily appropriations of content. The introduction of a quantitative modifier, the requirement for a 100% taking, would narrow down the per se category. The modified version has a lot more surface appeal, as it seems to sit well with prevailing moral intuitions. It is also consistent with current law.[284] However, it is not without its problems. First, in the case of some works—such as photos, poems, tattoos, graffiti, short slogans—copying of the entire work is a common practice and it is often difficult to take less without distorting the content, message, or meaning of the original work.[285] In other instances where it is possible to take less than 100% of the work, copyists who would quickly become aware of the new legal regime would suffice themselves with appropriation of less than the full work in order to avoid the application of the per se rule. Finally, the proposed per se rule ignores the potential social utility of full body appropriations of expressive content. As we have established elsewhere, such appropriations may be justified not only in instances of parodies but also in instances in which the defendant would like to use the plaintiff’s work or statement as proof that she actually made it, as is the case between two political rivals or two academics.[286]

3. New Defenses.—Another way in which copyright liability might be varied is along the dimension of defenses. Instead of adopting our proposal, lawmakers could keep in place a uniform system of liability but recognize new types of defenses. As we have noted earlier in this Article, historically U.S. copyright law recognized an innocent infringer defense.[287] Reintroducing an innocent infringer defense could be the functional equivalent of adopting a category of inadvertent infringements. Furthermore, it is possible to give weight to the culpability of the defendant as part of the fair use defense. In assessing fair use claims, courts primarily rely on the four statutory factors listed in § 107 of the Copyright Act, namely: (1) the purpose and nature of the use; (2) the nature of the copyrighted work; (3) the amount and substantiality taken; and (4) the actual and potential effect on the market for the copyrighted work.[288] Fair use, in other words, is a multifactored defense that leaves a lot of discretion to the court. Importantly, the enumeration of factors in § 107 is not a closed one and courts have the authority to consider additional factors.[289] Courts, therefore, have the authority to consider the culpability of users who raise the fair use defense, and, in fact, have already exercised this power. In Harper & Row, Publishers, Inc. v. Nation Enterprises,[290] for instance, the Supreme Court denied the defendant’s fair use claim, inter alia, because it acted in bad faith.[291] The case was brought after the Nation magazine had copied several short passages from a purloined copy of President Ford’s memoirs prior to the book’s publication.[292] Writing for the majority, Justice O’Connor ruled that the defendant, the Nation magazine, was fully aware that it was in possession of a stolen manuscript that was about to be published and, by deciding to use it, acted in bad faith.[293] This played a role—perhaps decisive—in the decision that the Nation could not avail itself of a fair use defense.[294]

Here, too, one should tread with caution in incorporating culpability into fair use doctrine. Presently, the key variable in fair use cases is transformativeness.[295] The more transformative a defendant’s use, the more likely it is to be considered fair. Transformative uses have been recognized as fair uses even in cases of bad faith copying. The most extreme examples to date are found in the cases of appropriation art. As noted above, courts have in no small number of cases sided with the bad faith appropriator.[296] Defenders of appropriation art claim that transformative works may be of considerable social value irrespective of the moral posture of their creators.[297] As Pablo Picasso famously quipped, “good artists copy, great artists steal.”[298] Hence, before rushing to transform the fair use defense, lawmakers should carefully assess the implications of keying fair use determinations to moral notions.

4. Contributory Fault.—Finally, it is also possible to vary copyright liability based on the degree of fault or negligence of the copyright owner. Like contemporary copyright liability, tort liability was once binary. Historically, defendants in tort cases were either liable for the full harm occasioned by their actions or omissions on their victims or not liable at all.[299] Modern tort law follows a radically different design. Today, tort law incorporates the notion of comparative fault, which allows courts to apportion responsibility for harms between tortfeasors and victims.[300] It is possible to adopt a similar design for copyright liability. For example, we could amend copyright law to provide that authors and publishers who did not affix a copyright notice to their works would not be able to sue others or would only be able to collect certain minimal damages. In fact, the Copyright Act adumbrates this possibility by affording statutory damages only to copyright owners who registered their work prior to the infringement for which they sue.[301] To be sure, such a reform would necessitate the reintroduction or augmentation of certain formalities that the copyright system abolished some forty years ago. But Chris Sprigman, as well as other academics, have presented a powerful prima facie case for effecting this change.[302] Copyright owners who have not taken reasonable measures to notify users that content is protected should be deemed partially responsible for infringements of their rights, and while they should not be denied legal relief altogether, the remedies awarded to them should be adjusted downward to reflect this fact.

In a similar vein, Congress can reform the Copyright Act to deny recourse to copyright owners who induced users to infringe their work. There exists a myriad of examples of copyright owners who uploaded content onto various Internet sites, enticed users to use the content, and then sued them.[303] The doctrine of copyright misuse can potentially address such cases.[304] However, misuse doctrine is confused and, perhaps because the consequences of misuse are extreme, courts shy away from it and rarely find misuse.[305] In this instance, therefore, improving the law requires cooperation not only from the legislature but also from the courts.

Conclusion

In this Article, we proposed a new liability regime for copyright infringements predicated on the degree of blameworthiness of the infringer. Concretely, we called for the establishment of three distinct liability categories—inadvertent, standard, and willful infringements—and proposed a specific menu of remedies for each of the categories. Under our proposal, inadvertent infringements would be remedied by compensatory damages only. Standard infringements would entitle copyright owners to a broader variety of monetary damages that includes restitution and profits, and in the alternative, limited statutory damages, but typically no injunctive relief. Willful infringements would be addressed by the full panoply of remedies available under the Copyright Act, including potentially punitive statutory damages and injunctions.

Our proposal marks a significant departure from the current regime that ignores the culpability level of infringements and treats all infringements alike for liability purposes. As we demonstrated, the existing “one-size-fits-all” design of copyright liability sets back the goals of the Copyright Act, as it causes overdeterrence and discourages future creativity. Furthermore, it fails to incentivize copyright owners to fully disclose proprietary and transactional information about their works. The result is both unfair and inefficient. Extant law metes out excessive penalties on users and creators who are ill-situated to handle the cloud of uncertainty that shrouds our copyright system, even though such users and creators often act in good faith. It likewise fails to distinguish between creators and users who try to navigate the uncertainties of copyright law, even if they ultimately fail, and blatant infringers who do not bother to partake of this quest. Most importantly, extant law discourages socially desirable uses of copyrighted works by creating an excessive fear of the penalties of copyright infringements. Our proposal of graduated liability yields a far more nuanced and just liability system, one that would also enhance use of existing works and future creativity.

  1. .Michael Bachner, Israel’s Eurovision Win in Jeopardy as US Label Claims Netta’s ‘Toy’ Was Copied, Times Israel (July 3, 2018, 10:50 PM), https://‌www.timesofisrael.com‌/israels-eurovision-win-in-jeopardy-as-us-label-claims-nettas-toy-was-copied‌/ [https://‌perma.cc‌/X6ND-MN92]; Amy Spiro, Does ‘Toy’ Sound Like ‘Seven Nation Army’?, Jerusalem Post (July 3, 2018), https://‌www.jpost.com‌/OMG‌/Does-Toy-sound-like-Seven-Nation-Army-561473 [https://‌perma.cc‌/84KJ-J4M8].
  2. .No. LA CV13–06004 JAK (AGRx), 2015 WL 4479500 (C.D. Cal. July 14, 2015), aff’d in part, rev’d in part sub nom. Williams v. Gaye, 895 F.3d 1106 (9th Cir. 2018).
  3. .Id. at *12.
  4. .Id. at *21.
  5. .Id. at *5 (limiting protection to musical elements contained in the lead sheets of “Got to Give It Up,” which was registered while the 1909 Copyright Act was in effect).
  6. .Id. at *1.
  7. .The $5.6 million award of damages was remitted to approximately $3.5 million on remittitur. Id. at *30. The running royalty of 50% of future proceedings was sustained. Id. at *45.
  8. .Williams v. Gaye, 895 F.3d 1106, 1128 (9th Cir. 2018). It should be noted that, in part, the lower-court judgment was affirmed because certain issues had not been properly raised for appeal. Id. at 1130–31.
  9. .See, e.g., Warner Bros. Entm’t Inc. v. RDR Books, 575 F. Supp. 2d 513, 527, 554 (S.D.N.Y. 2008) (holding that the copying of long quotations from a novel is copyright infringement).
  10. .See, e.g., Bridgeport Music, Inc., 2015 WL 4479500, at *1 (discussing the potential copying of Marvin Gaye’s “Got to Give It Up” by the defendants in “Blurred Lines”); Bright Tunes Music Corp. v. Harrisongs Music, Ltd., 420 F. Supp. 177, 180–81 (S.D.N.Y. 1976) (holding that George Harrison’s song “My Sweet Lord” infringed on an earlier song, even though Harrison’s copying was subconscious), aff’d sub nom. ABKCO Music, Inc. v. Harrisongs Music, Ltd., 722 F.2d 988 (2d Cir. 1983); ABKCO Music, Inc. v. Harrisongs Music, Ltd., 508 F. Supp. 798, 799 (S.D.N.Y. 1981) (ruling on the alleged infringement by George Harrison’s song “My Sweet Lord”), modified, 722 F.2d 988 (2d Cir. 1983);.
  11. .See infra Part II.
  12. .17 U.S.C. § 504(c) (2012).
  13. .See id. (prescribing statutory-damage amounts).
  14. .109 F.3d 1394 (9th Cir. 1997).
  15. .510 U.S. 569 (1994).
  16. .Id. at 572, 588.
  17. .Id. at 572.
  18. .Dr. Seuss Enters., 109 F.3d at 1396.
  19. .See id. at 1397 (quoting Katz’s declaration that The Cat in the Hat was the “object for [his] parody”).
  20. .Id. at 1400 (limiting the parody defense only to cases involving a discernable direct comment on the original).
  21. .Id. at 1406 (enjoining the defendants from “directly or indirectly” copying and distributing the book).
  22. .See infra Part II.
  23. .The one exception is statutory damages. See Pamela Samuelson & Tara Wheatland, Statutory Damages in Copyright Law: A Remedy in Need of Reform, 51 Wm. & Mary L. Rev. 439, 441 (2009) (explaining upper-end damage awards are “reserved for ‘willful’ infringers”).
  24. .See infra Part II.
  25. .See, e.g., Arnstein v. Porter, 154 F.2d 464, 468 (2d Cir. 1946) (setting out the modern test for determining whether a defendant copied the protected elements of a work); Nichols v. Universal Pictures Corp., 45 F.2d 119, 121 (2d Cir. 1930) (discussing how a court should go about deciding whether a movie has copied a previously written play).
  26. .The terminology for these two elements differs among courts and legal scholars, but the basic elements are the same. See Laureyssens v. Idea Grp., Inc., 964 F.2d 131, 140 (2d Cir. 1992) (discussing the presence of the substantial similarity requirement in both the copying and the improper-appropriation prong); Arnstein, 154 F.2d at 468 (setting out a two-prong test for copyright infringement with the first prong being copying and the second being improper appropriation); Alan Latman, “Probative Similarity” as Proof of Copying: Toward Dispelling Some Myths in Copyright Infringement, 90 Colum. L. Rev. 1187, 1198 (1990) (discussing both the vagueness of the term improper appropriation and the confusion caused by having substantial similarity as a part of both prongs).
  27. .17 U.S.C. §§ 502–504 (2012).
  28. .420 F. Supp. 177 (S.D.N.Y. 1976), aff’d sub nom. ABKCO Music, Inc. v. Harrisongs Music, Ltd., 722 F.2d 988 (2d Cir. 1983).
  29. .Id. at 178.
  30. .Williams v. Bridgeport Music, Inc., No. LA CV13–06004 JAK (AGRx), 2015 WL 4479500, at *1 (C.D. Cal. July 14, 2015), aff’d in part, rev’d in part sub nom. Williams v. Gaye, 895 F.3d 1106 (9th Cir. 2018).
  31. .Bright Tunes, 420 F. Supp. at 180.
  32. .Id.
  33. .ABKCO Music, Inc. v. Harrisongs Music, Ltd., 508 F. Supp. 798, 802 (S.D.N.Y. 1981), modified, 722 F.2d 988 (2d Cir. 1983).
  34. .See infra section I(B)(4).
  35. .See Samuelson & Wheatland, supra note 23, at 474 (discussing the “tripartite structure” of statutory-damages awards under the Copyright Act of 1976).
  36. .17 U.S.C. § 504(c) (2012).
  37. .Id.
  38. .Bright Tunes Music Corp. v. Harrisongs Music, Ltd., 420 F. Supp. 177, 181 (S.D.N.Y. 1976), aff’d sub nom. ABKCO Music, Inc. v. Harrisongs Music, Ltd., 722 F.2d 988 (2d Cir. 1983).
  39. .Williams v. Bridgeport Music, Inc., LA CV13–06004 JAK (AGRx), 2015 WL 4479500, at *21 (C.D. Cal. July 14, 2015), aff’d in part, rev’d in part sub nom. Williams v. Gaye, 895 F.3d 1106 (9th Cir. 2018).
  40. .See Dr. Seuss Enters., L.P. v. Penguin Books USA, Inc., 109 F.3d 1394, 1402–03 (9th Cir. 1997) (stating that a fair use defense was unconvincing because of the substantial similarity between the infringing work and original work).
  41. .Jordan Runtagh, Songs on Trial: 12 Landmark Music Copyright Cases, Rolling Stone (June 8, 2016, 4:24 PM), http://‌www.rollingstone.com‌/music‌/lists‌/songs-on-trial-12-landmark-music-copyright-cases-20160608‌/the-verve-vs-the-rolling-stones-1997-20160608 [https://‌perma.cc‌/ZEC4-3JGY].
  42. .Id.
  43. .Id.
  44. .See Ben Depoorter et al., Copyright Backlash, 84 S. Cal. L. Rev. 1251, 1253–54 (2011) (pointing out the gap between copyright law and the social norms pertaining to use of copyrighted content); Ben Depoorter & Sven Vanneste, Norms and Enforcement: The Case Against Copyright Litigation, 84 Or. L. Rev. 1127, 1128 (2005) (same); Ben Depoorter et al., Problems with the Enforcement of Copyright Law: Is There a Social Norm Backlash?, 12 Int’l J. Econ. Bus. 361, 362 (2005) (same); John Tehranian, Infringement Nation: Copyright Reform and the Law‌/Norm Gap, 2007 Utah L. Rev. 537, 543 (same).
  45. .See Gary S. Becker, Crime and Punishment: An Economic Approach, 76 J. Pol. Econ. 169, 170 (1968) (discussing the varying resource costs of different methods of law enforcement).
  46. .See infra Part IV.
  47. .See Ben Depoorter, Copyright Enforcement in the Digital Age: When the Remedy Is the Wrong, 66 UCLA L. Rev. 400, 402–04 (2019) (discussing the “astronomically high” statutory-damages awards copyright-infringement plaintiffs can expect).
  48. .Sometimes the law grants the rights to persons other than the author. See 17 U.S.C. § 101 (2012) (defining “a work made for hire” as “a work prepared by an employee within the scope of his or her employment,” or as a commissioned work); Cmty. for Creative Non-Violence v. Reid, 490 U.S. 730, 732, 737 (1989) (explaining the work-made-for-hire doctrine and how this could translate into the author not becoming the owner of the work he has produced).
  49. .Folsom v. Marsh, 9 F. Cas. 342, 344 (C.C.D. Mass. 1841) (No. 4,901).
  50. .Peter Pan Fabrics, Inc. v. Martin Weiner Corp., 274 F.2d 487, 489 (2d Cir. 1960).
  51. .17 U.S.C. § 101, 102(a). The 1976 Act does not preempt common law copyright protection of unfixed works, such as lectures or prepared dialogues. Id. § 301(b)(1). But cf. Estate of Hemingway v. Random House, Inc., 268 N.Y.S.2d 531, 536–37 (N.Y. Sup. Ct.) (considering but rejecting common law protection for conversations Hemingway had with the author of the biography in question), aff’d, 25 A.D.2d 719 (N.Y. App. Div. 1966).
  52. .1 Howard B. Abrams, The Law of Copyright § 8:25 (2010); see, e.g., Thomas F. Cotter, Toward a Functional Definition of Publication in Copyright Law, 92 Minn. L. Rev. 1724, 1725–28 (2008) (reasoning that while publication is no longer required to establish copyright, there are still important reasons why publication might still matter).
  53. .17 U.S.C. § 408(a) (registration not required); Abrams, supra note 52, at § 9:4 (notice not required); id. at § 10:2 (registration not required). Prior to March 1, 1989, the 1976 Act required proper notice on published works, failure of which could result in forfeiture of protection. Abrams, supra note 52, at § 9:65; see also Cotter, supra note 52, at 1725 (discussing how the Berne Convention removed the notice requirement from U.S. copyright law); Thomas M. Landrigan, Application or Registration?: Confusion Regarding the Copyright Act’s Prerequisite to Copyright Infringement Lawsuits, 44 Ind. L. Rev. 581, 583 (2011) (stating that while registration is not required, it is required before instituting a suit for copyright infringement).
  54. .See 17 U.S.C. § 101 (including “numerical symbols” in the definition of literary work); Apple Comput., Inc. v. Franklin Comput. Corp., 714 F.2d 1240, 1253–54 (3d Cir. 1983) (holding that computer operating systems are not per se precluded from copyright). See generally Dennis S. Karjala, Copyright Protection of Operating Software, Copyright Misuse, and Antitrust, 9 Cornell J.L. & Pub. Pol’y 161 (1999) (discussing the copyrightability of computer operating systems); Julian Velasco, Note, The Copyrightability of Nonliteral Elements of Computer Programs, 94 Colum. L. Rev. 242 (1994) (examining the relevant tests that courts have used to determine whether the nonliteral elements of computer programs can be the basis of a copyright).
  55. .See, e.g., Verified Complaint for Injunctive and Other Relief at 7–8, Whitmill v. Warner Bros. Entm’t Inc. (E.D. Mo. Apr. 28, 2011) (No. 4:11-cv-752), 2011 WL 2038147 (seeking to enjoin the defendant from using Mike Tyson’s facial tattoo designed by the plaintiff). For more information on copyright protection of tattoos, see generally Christopher A. Harkins, Tattoos and Copyright Infringement: Celebrities, Marketers, and Businesses Beware of the Ink, 10 Lewis & Clark L. Rev. 313 (2006).
  56. .See Kelley v. Chicago Park Dist., 635 F.3d 290, 292 (7th Cir. 2011) (holding that the garden is not protected under the 1976 Act due to the lack of authorship and fixation).
  57. .See 17 U.S.C. § 302(a) (stating that copyright in a work created on or after January 1, 1978, has a copyright lifetime consisting of the life of the author plus 70 years).
  58. .See Eldred v. Ashcroft, 537 U.S. 186, 194–96 (2003) (discussing the various changes in copyright duration from 1790 until 2003); Kevin A. Goldman, Comment, Limited Times: Rethinking the Bounds of Copyright Protection, 154 U. Pa. L. Rev. 705, 708–18 (2006) (tracing changes in copyright law from 1790 until modern times).
  59. .See, e.g., Feist Publ’ns, Inc. v. Rural Tel. Serv. Co., 499 U.S. 340, 348 (1991) (acknowledging that factual compilation can be protected if it exhibits originality in the selection and arrangement); Carol Barnhart Inc. v. Econ. Cover Corp., 773 F.2d 411, 419 (2d Cir. 1985) (recognizing the copyrightability of conceptually separable artistic aspects of a utilitarian article that are otherwise unprotected); Shine v. Childs, 382 F. Supp. 2d 602, 608–09 (S.D.N.Y. 2005) (concluding that architectural models are entitled protection).
  60. .DC Comics v. Pac. Pictures Corp., 545 F. App’x 678, 680–81 (9th Cir. 2013) (holding that the copyright termination notice of the heir of one of the cocreators of the Superman character was invalid).
  61. .Calvin Reid, DC Marks 80 Years of ‘Action Comics’ and Superman with Hardcover Tribute, Publishers Wkly. (Apr. 13, 2018), https://‌www.publishersweekly.com‌/pw‌/by-topic‌/industry-news‌/comics‌/article‌/76612-dc-marks-80-years-of-action-comics-and-superman-with-hardcover-tribute.html [https://‌perma.cc‌/T6BC-PV5P].
  62. .See Warner Bros. Entm’t, Inc. v. X One X Prods., 644 F.3d 584, 599, 603 (8th Cir. 2011) (discussing how a character’s clothing or catchphrase could have different ownership than the character itself).
  63. .An “orphan work” is a copyrighted work whose author cannot be identified or located. Orphan Works and Mass Digitization, 77 Fed. Reg. 64555 (Oct. 22, 2012); U.S. Copyright Office, Report on Orphan Works 1 (2006); Olive Huang, Note, U.S. Copyright Office Orphan Works Inquiry: Finding Homes for the Orphans, 21 Berkeley Tech. L.J. 265, 265 (2006). Orphan works cannot be used unless the fair use doctrine applies. See Authors Guild, Inc. v. HathiTrust, 755 F.3d 87, 95 (2d Cir. 2014) (stating that the fair use doctrine “allows the public to draw upon copyrighted materials without the permission of the copyright holder in certain circumstances”).
  64. .See infra subpart III(A).
  65. .See United States v. Liu, 731 F.3d 982, 985 (9th Cir. 2013) (stating the importance of proving willful action in a counterfeit case); United States v. Beltrán, 503 F.3d 1, 2 (1st Cir. 2007) (stating that copying a DVD that has an FBI copyright warning on it constitutes willful infringement). For more information on counterfeit DVDs and criminal infringement of copyright, see generally Michael M. DuBose, Criminal Enforcement of Intellectual Property Laws in the Twenty-First Century, 29 Colum. J.L. & Arts 481 (2006) and Rebecca E. Hatch, Criminal Infringement of Copyright Under 17 U.S.C.A. § 506, in 120 American Jurisprudence: Proof of Facts 181, 191 (Jennifer J. Chen ed., 3d ed. 2011).
  66. .See, e.g., United States v. Backer, 134 F.2d 533, 535 (2d Cir. 1943) (affirming conviction of the defendant for willfully infringing copyrighted figurines for profit); People v. Santiago, 9 Misc. 3d 197, 201 (N.Y. Crim. Ct. 2005) (holding that the First Amendment did not preclude prosecution for unlicensed selling of counterfeit DVDs); Press Release, Dep’t of Justice, Federal Courts Order Seizure of Three Website Domains Involved in Distributing Pirated Android Cell Phone Apps (Aug. 21, 2012), https://‌www.justice.gov‌/opa‌/pr‌/federal-courts-order-seizure-three-website-domains-involved-distributing-pirated-android-cell [https://‌perma.cc‌/6QWC-4EYT] (“The seizures are the result of a comprehensive enforcement action taken to prevent the infringement of copyrighted mobile device apps.”); see also Comment, Willful Copyright Infringement: In Search of a Standard, 65 Wash. L. Rev. 903, 907–09 (1990) (discussing various standards to measure willfulness).
  67. .Jerry Markon, Florida Man Gets Six Years in Prison for Software Piracy, Wash. Post (Aug. 26, 2006), http://‌www.washingtonpost.com‌/wp-dyn‌/content‌/article‌/2006‌/08‌/25‌/AR2006082500511.html [https://‌perma.cc‌/Y9TC-ZTTG].
  68. .Id.
  69. .Id.
  70. .Id.
  71. .Id.
  72. .See infra subpart III(A).
  73. .4 Melville B. Nimmer & David Nimmer, Nimmer on Copyright § 13.01(B) (2019).
  74. .Id. § 13.01 (footnotes omitted).
  75. .Bright Tunes Music Corp. v. Harrisongs Music, Ltd., 420 F. Supp. 177, 178 (S.D.N.Y. 1976), aff’d sub nom. ABKCO Music, Inc. v. Harrisongs Music, Ltd., 722 F.2d 988 (2d Cir. 1983).
  76. .Id. at 179–80.
  77. .Id. at 180.
  78. .Id. at 181 (“This is . . . infringement of copyright, and is no less so even though subconsciously accomplished.”).
  79. .212 F.3d 477 (9th Cir. 2000).
  80. .Id. at 480.
  81. .Id. at 484.
  82. .See Olufunmilayo B. Arewa, The Freedom to Copy: Copyright, Creation, and Context, 41 U.C. Davis L. Rev. 477, 536 (2007) (“The composition in which Bolton participated was also far removed from the Isley Brothers song in time as well . . . .”).
  83. .See id. (explaining that the Isley Brothers’ song was never listed on a Top 100 chart).
  84. .See id. (“Bolton’s acknowledged reverence of the Isley Brothers and the existence of a 1991 CD version of the Isley Brothers song are thus used to bolster fairly tenuous assumptions about access and subconscious infringement.”).
  85. .Three Boys Music, 212 F.3d at 484.
  86. .See infra subpart III(A).
  87. .See 17 U.S.C. § 102(a) (2012) (extending copyright protection to “literary works”); Salinger v. Colting, 607 F.3d 68, 83 (2d Cir. 2010) (affirming that the texts of the two works in question are substantially similar); Suntrust Bank v. Houghton Mifflin Co., 268 F.3d 1257, 1259 (11th Cir. 2001) (considering whether injunctive relief was appropriate for alleged infringement of the novel Gone With the Wind). For further discussion, see Douglas Y’Baro, Aesthetic Ambition Versus Commercial Appeal: Adapting Novels to Film and the Copyright Law, 10 St. Thomas L. Rev. 299, 315–20 (1998) (discussing whether a film’s adoption of a novel’s text is prohibited by copyright law).
  88. .See Klinger v. Conan Doyle Estate, Ltd., 988 F. Supp. 2d 879, 893 (N.D. Ill. 2013) (holding that characters that have not yet entered the public domain are protectable), aff’d, 755 F.3d 496 (7th Cir. 2014). But cf. Sinicola v. Warner Bros., Inc., 948 F. Supp. 1176, 1185 (E.D.N.Y. 1996) (stating that the less developed a character is, the less likely he is to be protected by copyright law); Note, The Protection Afforded Literary and Cartoon Characters Through Trademark, Unfair Competition, and Copyright, 68 Harv. L. Rev. 349, 357 (1954) (reasoning that the use of a literary character in a context similar to the original is almost conclusive evidence of copyright infringement, but recognizing that a less developed character is less likely to be protected).
  89. .See Dr. Seuss Enters. v. Penguin Books USA, Inc., 109 F.3d 1394, 1399–401 (9th Cir. 1997) (holding that a book describing the O.J. Simpson murder trial written in the style of the plaintiff’s children’s books is not entitled to fair use defense); Hassett v. Hasselbeck, 177 F. Supp. 3d 626, 633 (D. Mass. 2016) (holding that style is a component of the overall concept and feel of a literary work, which is protected). But see Douglas v. Osteen, 317 F. App’x 97, 99 (3d Cir. 2009) (“[T]he use of a particular writing style or literary method is not protected by the Copyright Act.”); McMahon v. Prentice-Hall, Inc., 486 F. Supp. 1296, 1302 (E.D. Mo. 1980) (“A writer may not claim a monopoly on a particular writing style by virtue of a copyright.”).
  90. .See Nash v. CBS, Inc., 899 F.2d 1537, 1543 (7th Cir. 1990) (holding that only the expression and not the analysis of history is protected by copyright law); Miller v. Universal City Studios, Inc., 650 F.2d 1365, 1369–70 (5th Cir. 1981) (holding that facts underlying a literary work cannot be appropriated through copyright law); Effie Film, LLC v. Pomerance, 909 F. Supp. 2d 273, 298 (S.D.N.Y. 2012) (holding that historical facts or interpretations do not count towards substantial similarity analysis).
  91. .See, e.g., Stromback v. New Line Cinema, 384 F.3d 283, 296 (6th Cir. 2004) (holding that depiction of parties, alcohol, and wild behavior are scènes à faire in a story about a college fraternity); Shaw v. Lindheim, 919 F.2d 1353, 1362–64 (9th Cir. 1990) (discounting claims of infringement based on copying of “‘scenes à faire’—that is, scenes that flow naturally from a basic plot premise,” while allowing other copying claims to proceed to trial). “Scènes à faire refers to ‘incidents, characters or settings which are as a practical matter indispensable, or at least standard, in the treatment of a given topic.’” Atari, Inc. v. N. Am. Philips Consumer Elecs. Corp., 672 F.2d 607, 616 (7th Cir. 1982) (quoting Alexander v. Haley, 460 F. Supp. 40, 45 (S.D.N.Y. 1978)).
  92. .17 U.S.C. § 101 (2012).
  93. .Id.; H.R. Rep. No. 94-1476, as reprinted in 17 U.S.C. § 102 (2012).
  94. .H.R. Rep. No. 94-1476.
  95. .See Kieselstein-Cord v. Accessories by Pearl, Inc., 632 F.2d 989, 993 (2d Cir. 1980) (protecting belt buckles whose artistic features are primary and the utilitarian features subsidiary).
  96. .See Brandir Int’l, Inc. v. Cascade Pac. Lumber Co., 834 F.2d 1142, 1146–47 (2d Cir. 1987) (determining that the bicycle rack was not entitled to protection under the independent-artistic-judgment test).
  97. .See Mazer v. Stein, 347 U.S. 201, 202, 217 (1954) (holding that the bases for lamps in the form of male and female dancing figures did not invalidate the plaintiff’s registration of copyright); Lamps Plus, Inc. v. Seattle Lighting Fixture Co., 345 F.3d 1140, 1142 (9th Cir. 2003) (discussing the protection of table lamps); Sunset Lamp Corp. v. Alsy Corp., 698 F. Supp. 1146, 1148 (S.D.N.Y. 1988) (discussing the protectability of floor lamp designs).
  98. .See Carol Barnhart Inc. v. Econ. Cover Corp., 773 F.2d 411, 414 (2d Cir. 1985) (recognizing the copyrightability of conceptually separable artistic aspects of a utilitarian article that are otherwise unprotected).
  99. .See Star Athletica, L.L.C. v. Varsity Brands, Inc., 137 S. Ct. 1002, 1016 (2017) (holding that the design on the relevant cheerleading uniforms warranted copyright protection because it could be “perceived as a two- or three-dimensional work of art separate from the useful article” and because it qualified “as a protectable pictorial, graphic, or sculptural work”).
  100. .See Pivot Point Int’l, Inc. v. Charlene Prods., Inc., 372 F.3d 913, 921 (7th Cir. 2004) (noting that the statutory language with respect to the copyrightability of utilitarian articles “has presented the courts with significant difficulty”). For more discussions of the topic, see generally John B. Fowles, The Utility of a Bright-Line Rule in Copyright Law: Freeing Judges from Aesthetic Controversy and Conceptual Separability in Leicester v. Warner Bros., 12 UCLA Ent. L. Rev. 301 (2005) and Barton R. Keyes, Alive and Well: The (Still) Ongoing Debate Surrounding Conceptual Separability in American Copyright Law, 69 Ohio St. L.J. 109 (2008).
  101. .Works by Arthur Conan Doyle Featuring Sherlock Holmes, Conan Doyle Info, https://‌www.conandoyleinfo.com‌/sherlock-holmes‌/list-of-sherlock-holmes-fiction‌/ [https://‌perma.cc‌/TNX8-XLYB].
  102. .Klinger v. Conan Doyle Estate, Ltd., 988 F. Supp. 2d 879, 893 (N.D. Ill. 2013), aff’d, 755 F.3d 496 (7th Cir. 2014) (extending protection to a later version of the Sherlock Holmes character when the protection term of the original version of the round character had expired).
  103. .Sherlock Holmes (Warner Bros. Pictures 2009).
  104. .Sherlock Holmes: A Game of Shadows (Warner Bros. Pictures 2011).
  105. .45 F.2d 119 (2d Cir. 1930).
  106. .Id. at 121.
  107. .Id. at 121–22.
  108. .216 F.2d 945 (9th Cir. 1954).
  109. .Id. at 948.
  110. .Warner Bros. Pictures, Inc. v. Columbia Broad. Sys., Inc., 102 F. Supp. 141, 143 (S.D. Cal. 1951), aff’d in part, rev’d in part, 216 F.2d 945 (9th Cir. 1954).
  111. .Id. at 143–44, 149. Dashiell Hammett wrote three more short stories featuring Sam Spade in 1932, but the stories were not part of the deal with Warner Brothers. Id. at 144. The 1941 film was nominated for three Academy Awards, and it was eventually recognized by the American Film Institute as one of the greatest movies ever made. See The 14th Academy Awards: 1942, Oscars.org, https://‌www.oscars.org‌/oscars‌/ceremonies‌/1942 [https://‌perma.cc‌/EWP4-GAHV] (listing three Academy Award nominations for The Maltese Falcon: Actor in a Supporting Role, Outstanding Motion Picture, and Writing (Screenplay)); AFI’s 100 Greatest American Movies of All Time, AFI, http://‌www.afi.com‌/100years‌/movies.aspx [https://‌perma.cc‌/8NXY-PN7P] (ranking The Maltese Falcon as the twenty-third on the list); AFI’s 100 Greatest American Movies of All Time: 10th Anniversary Edition, AFI, http://‌www.afi.com‌/100years‌/movies10.aspx [https://‌perma.cc‌/6E7R-NT8S] (ranking The Maltese Falcon as the thirty-first on the list).
  112. .Warner Bros. Pictures, 102 F. Supp. at 144.
  113. .Warner Bros. Pictures, 216 F.2d at 948–49.
  114. .Id. at 950.
  115. .No. 87–0592 WDKGX, 1989 WL 206431 (C.D. Cal. Apr. 25, 1989).
  116. .Id. at *1.
  117. .See id. (describing the sequence of events).
  118. .Id. at *1, *14–15.
  119. .Id. at *1.
  120. .Id. at *18.
  121. .Id. at *6–7. The court acknowledged some confusion in the doctrine, noting that the Second Circuit follows a different test for determining whether copyrighted characters are copyrightable, and that subsequent litigation in the Ninth Circuit has raised questions about whether the ruling in the Sam Spade litigation is still good law. Id. at *6. The court decided that the characters would be copyrighted under all the competing tests, making it unnecessary to determine which test is required in the Ninth Circuit. Id. at *8.
  122. .Id. at *7–8.
  123. .Id. at *8–9.
  124. .Metro-Goldwyn-Mayer, Inc. v. Am. Honda Motor Co., 900 F. Supp. 1287, 1298 (C.D. Cal. 1995). The court described the ad as follows:

    Defendants’ “Escape” commercial features a young, well-dressed couple in a Honda del Sol being chased by a high-tech helicopter. A grotesque villain with metal-encased arms jumps out of the helicopter onto the car’s roof, threatening harm. With a flirtatious turn to his companion, the male driver deftly releases the Honda’s detachable roof (which Defendants claim is the main feature allegedly highlighted by the commercial), sending the villain into space and effecting the couple’s speedy get-away.

    Id. at 1291.

  125. .Id. at 1304 (denying summary judgment under the intrinsic test because “the jury must make a factual determination”).
  126. .See infra notes 128–38 and accompanying text.
  127. .429 F.2d 1106 (9th Cir. 1970).
  128. .Id. at 1107.
  129. .Id. at 1109.
  130. .Id. at 1109–11.
  131. .Id. at 1110.
  132. .562 F.2d 1157 (9th Cir. 1977).
  133. .Id. at 1160.
  134. .Id. at 1166–67.
  135. .Id. at 1166.
  136. .Id. at 1116.
  137. .Id. at 1166–67.
  138. .274 F.2d 487 (2d Cir. 1960).
  139. .Id. at 489.
  140. .See infra subpart III(A).
  141. .17 U.S.C. § 107 (2012); see also Ty, Inc. v. Publ’ns Int’l Ltd., 292 F.3d 512, 517 (7th Cir. 2002) (stating that through the fair use doctrine a book review can copy a short quote from the original work without a license from the author); Folsom v. March, 9 F. Cas. 342, 344 (C.C.D. Mass. 1841) (No. 4,901) (“Thus, for example, no one can doubt that a reviewer may fairly cite largely from the original work, if his design be really and truly to use the passages for the purposes of fair and reasonable criticism.”); Nimmer & Nimmer, supra note 73, at § 13.05 (providing an overview of the standards governing fair use doctrine).
  142. .See Jennifer T. Olsson, Note, Rights in Fine Art Photography: Through a Lens Darkly, 70 Texas L. Rev. 1489, 1514–15 (1992) (discussing an example of appropriation art).
  143. .Cariou v. Prince, 714 F.3d 694, 699 (2d Cir. 2013).
  144. .After Walker Evans: 4, The Met, http://‌www.metmuseum.org‌/art‌/collection‌/search‌/267214 [https://‌perma.cc‌/EPP8-RDLB]; see also AfterSherrieLevine.com, http://‌www.aftersherrielevine.com‌/ [https://‌perma.cc‌/EUD6-VPTP] (“Sherrie Levine rephotographed Walker Evans’ photographs from the exhibition catalog ‘First and Last.’”).
  145. .Amy Adler, Fair Use and the Future of Art, 91 N.Y.U. L. Rev. 559, 606 (2016).
  146. .Niels Schaumann, Fair Use and Appropriation Art, Cybaris, Summer 2015, at 112, 122; Carmen Winant, Sherrie Levine’s ‘Mayhem’: A Retrospective of the Original Fake at the Whitney, WNYC (Nov. 9, 2011), http://‌www.wnyc.org‌/story‌/169656-sherrie-levines-mayhem-retrospective-original-fake-whitney‌/ [https://‌perma.cc‌/NW38-FVYV] (“Believing her photographs to be copyright infringement, the estate bought the entire sweep to prevent anyone else from doing so.”).
  147. .467 F.3d 244, 246 (2d Cir. 2006).
  148. .960 F.2d 301, 309 (2d Cir. 1992).
  149. .817 F. Supp. 370, 372 (S.D.N.Y. 1993).
  150. .No. 91 Civ. 6055(RO), 1993 WL 97381, at *3 (S.D.N.Y. Apr. 1, 1993).
  151. .714 F.3d 694 (2d Cir. 2013).
  152. .Id. at 699.
  153. .Id. at 707.
  154. .Id. at 712.
  155. .Id. The parties ultimately reached a settlement regarding the final five images. Brian Boucher, Landmark Copyright Lawsuit Cariou v. Prince Is Settled, Art in Am. (Mar. 18, 2014), http://‌www.artinamericamagazine.com‌/news-features‌/news‌/landmark-copyright-lawsuit-cariou-v-prince-is-settled‌/ [https://‌perma.cc‌/4AC7-243M].
  156. .See Warner Bros. Inc. v. Am. Broad. Cos., 654 F.2d 204, 209 (2d Cir. 1981) (holding that defendant’s parody film was sufficiently distinct from plaintiffs’ movie to defeat a copyright-infringement claim).
  157. .Compare Atari Games Corp. v. Nintendo of Am. Inc., 975 F.2d 832, 847 (Fed. Cir. 1992) (affirming the lower court’s ruling that the software in question was substantially similar enough to the plaintiffs’ program to violate copyright law), with Sega Enters. Ltd. v. Accolade, Inc., 977 F.2d 1510, 1527 (9th Cir. 1992) (holding in favor of defendant’s fair use defense). See also Sony Comput. Entm’t, Inc. v. Connectix Corp., 203 F.3d 596, 599 (9th Cir. 2000) (holding that defendant’s reverse engineering of the disputed computer software was protected fair use). at 468te.).Porter, 154 F.2d 464, 468 (2d Cir. 1946)(reasoning that copying can be inferred by proving evidence of access and su at 468te.).Porter, 154 F.2d 464, 468 (2d Cir. 1946)(reasoning that copying can be inferred by proving evidence of access and su
  158. .Leibovitz v. Paramount Pictures Corp., 137 F.3d 109, 110 (2d Cir. 1998); Triangle Publ’ns, Inc. v. Knight-Ridder Newspapers, Inc., 626 F.2d 1171, 1172 (5th Cir. 1980); Straus v. DVC Worldwide, Inc., 484 F. Supp. 2d 620, 623 (S.D. Tex. 2007).
  159. .E.g., Cambridge Univ. Press v. Patton, 769 F.3d 1232, 1237 (11th Cir. 2014); Princeton Univ. Press v. Mich. Document Servs., 99 F.3d 1381, 1383 (6th Cir. 1996); Basic Books, Inc. v. Kinko’s Graphics Corp., 758 F. Supp. 1522, 1526 (S.D.N.Y. 1991).
  160. .See Sony Corp. of Am. v. Universal City Studios, Inc., 464 U.S. 417, 425 (1984) (discussing the legality of recording publicly broadcasted television shows and movies).
  161. .A&M Records, Inc. v. Napster, Inc., 239 F.3d 1004, 1014 (9th Cir. 2001), as amended (Apr. 3, 2001), aff’d, 284 F.3d 1091 (9th Cir. 2002).
  162. .Authors Guild v. Google, Inc., 804 F.3d 202, 207 (2d Cir. 2015).
  163. .Kelly v. Arriba Soft Corp., 336 F.3d 811, 815 (9th Cir. 2003).
  164. .Am. Geophysical Union v. Texaco Inc., 60 F.3d 913, 914–15 (2d Cir. 1994).
  165. .Ringgold v. Black Entm’t Television, Inc., 126 F.3d 70, 71–73 (2d Cir. 1997).
  166. .Harper & Row, Publishers, Inc. v. Nation Enters., 471 U.S. 539, 542, 545 (1985).
  167. .Authors Guild, 804 F.3d at 229.
  168. .Harper, 471 U.S. at 545, 569.
  169. .17 U.S.C. § 110(1) (2012).
  170. .Id. § 113(c).
  171. .Id. § 110(5).
  172. .See Sailor Music v. Gap Stores, Inc., 668 F.2d 84, 86 (2d Cir. 1981) (holding that the defendant store’s square footage was too large to fall under Congress’s exemption for performance of copyrighted works in private homes).
  173. .See Authors Guild v. Google, Inc., 804 F.3d 202, 207 (2d Cir. 2015) (stating that Google copied millions of books without a license).
  174. .See Am. Broad. Cos. v. Aereo, Inc., 134 S. Ct. 2498, 2503 (2014) (finding that Aereo copied programming without licenses).
  175. .Authors Guild, Inc. v. Google, Inc., 954 F. Supp. 2d 282, 294 (S.D.N.Y. 2013), aff’d, 804 F.3d 202 (2d Cir. 2015).
  176. .Authors Guild, 804 F.3d at 229–30.
  177. .Aereo, 134 S. Ct. at 2511.
  178. .See infra subpart III(A).
  179. .See I. Fred Koenigsberg, The 1976 Copyright Act: Advances for the Creator, 26 Clev. St. L. Rev. 515, 525–26 (1977) (describing the notice requirements in the 1976 Copyright Act).
  180. .See Berne Convention Implementation Act of 1988, Pub. L. No. 100-568, § 7, 102 Stat. 2853, 2857–58 (codified as amended at 17 U.S.C. § 401 (2012)) (relaxing the notice requirements).
  181. .See generally Annemarie Bridy, Three Notice Failures in Copyright Law, 96 B.U. L. Rev. 777 (2016) (describing the lack of stringent notice requirements in current copyright law).
  182. .U.S. Copyright Office, Orphan Works and Mass Digitization 9 (2015).
  183. .U.S. Copyright Office, supra note 63, at 7.
  184. .Aaron C. Young, Copyright’s Not So Little Secret: The Orphan Works Problem and Proposed Orphan Works Legislation, 7 Cybaris 202, 203 (2016).
  185. .See id. at 204 (discussing the billions of photographs taken and shared online each year).
  186. .See Vigdis Bronder, Saving the Right Orphans: The Special Case of Unpublished Orphan Works, 31 Colum. J.L. & Arts 409, 410 (2008) (discussing problems specific to unpublished orphan works). See generally David R. Hansen et al., Solving the Orphan Works Problem for the United States, 37 Colum. J.L. & Arts 1 (2013) (discussing various solutions to the problems presented by orphan works); Randal C. Picker, Private Digital Libraries and Orphan Works, 27 Berkeley Tech. L.J. 1259 (2012) (discussing orphan works as they relate to use by public libraries); Huang, supra note 63 (discussing the status of orphan works following the United States Copyright Office inquiry in 2005).
  187. .See Stewart v. Abend, 495 U.S. 207, 238 (1990) (finding infringement when a film used the unique setting, characters, plot, and sequence of events from an earlier work).
  188. .17 U.S.C. § 201(d) (2012).
  189. .Id. § 203(a)(3).
  190. .Id.
  191. .Copyright Act of 1909, Pub. L. No. 60-349, § 23, 35 Stat. 1075, 1080.
  192. .See, e.g., Miller Music Corp. v. Charles N. Daniels, Inc., 362 U.S. 373, 375 (1960) (holding that the heirs or next of kin of a deceased author “are among those to whom § 24 has granted the renewal right, irrespective of whether the author in his lifetime has or has not made any assignment of it”).
  193. .Getty Images Makes 35 Million Photos Free to Use, British J. Photography (Mar. 5, 2014), https://‌www.bjp-online.com‌/2014‌/03‌/getty-images-makes-35-million-images-free-in-fight-against-copyright-infringement‌/ [https://‌perma.cc‌/3F78-7ZM2].
  194. .Id.
  195. .Hong Luo & Julie Holland Mortimer, Copyright Infringement in the Market for Digital Images, 106 Am. Econ. Rev. (Papers & Proc.) 140, 140 (2016).
  196. .Id.
  197. .Id. at 142.
  198. .Id.
  199. .Id. at 142–43.
  200. .See, e.g., Boosey & Hawkes Music Publishers, Ltd. v. Walt Disney Co., 145 F.3d 481, 485–86 (2d Cir. 1998) (challenging Disney’s use of Stravinsky’s The Rite of Spring in newly published video cassettes of Fantasia); Video Pipeline, Inc. v. Buena Vista Home Entm’t, Inc., 275 F. Supp. 2d 543, 549 (D.N.J. 2003) (examining whether an official distributor of trailer clips in video-rental stores could stream those clips online); see also Konrad Gatien, Internet Killed the Video Star: How In-House Internet Distribution of Home Video Will Affect Profit Participants, 13 Fordham Intell. Prop. Media & Ent. L.J. 909, 923–24 (2003) (discussing in-house internet distribution of home videos); Dominique R. Shelton, How Licensing Issues Are Affecting Recent Media Developments, L.A. Law., June 2007, at 17, 17 (asking the question: “Does the right to distribute entertainment content through one channel include the right to distribute that content through a later-developed channel with a different underlying technology?”).
  201. .See, e.g., HarperCollins Publishers LLC v. Open Road Integrated Media, LLP, 7 F. Supp. 3d 363, 377 (S.D.N.Y. 2014) (holding that the initial author’s granting of publishing rights gave the publisher exclusive right to license the e-book version); Random House, Inc. v. Rosetta Books LLC, 150 F. Supp. 2d 613, 624 (S.D.N.Y. 2001) (holding that the licensee to the right to publish in book form could not succeed on a copyright-infringement claim against someone publishing the work in electronic form).
  202. .17 U.S.C. § 204(a) (2012).
  203. .Id. § 204(e).
  204. .908 F.2d 555 (9th Cir. 1990).
  205. .Id. at 555.
  206. .Id. at 556.
  207. .Id.
  208. .Id.
  209. .Id.
  210. .Id. at 559.
  211. .786 F.3d 733 (9th Cir. 2015) (en banc).
  212. .Id. at 736–37.
  213. .Id. at 738–39, 741 n.5.
  214. .Id. at 741–44.
  215. .Id. at 743.
  216. .17 U.S.C. § 106 (2012).
  217. .Id. § 501.
  218. .See, e.g., Columbia Pictures Indus., Inc. v. Aveco, Inc., 800 F.2d 59, 60 (3d Cir. 1986) (the public-performance right); London-Sire Records, Inc. v. Doe 1, 542 F. Supp. 2d 153, 174 (D. Mass. 2008) (the public-distribution right); Walt Disney Prods. v. Filmation Assocs., 628 F. Supp. 871, 875 (C.D. Cal. 1986) (the reproduction right).
  219. .Nimmer & Nimmer, supra note 73, at § 13.01 (footnotes omitted).
  220. .17 U.S.C. § 504.
  221. .Id. § 502.
  222. .Id. § 504(c).
  223. .Id.
  224. .Id. §§ 106, 501. There are also instances in which importation of a work can constitute an infringement. Id. § 602.
  225. .See Dane S. Ciolino & Erin A. Donelon, Questioning Strict Liability in Copyright, 54 Rutgers L. Rev. 351, 351 (2002) (“In copyright’s strict liability scheme, the infringer’s faultlessness or culpability is of anomalously little relevance.”); Comment, Internet Copyright Infringement and Service Providers: The Case for a Negotiated Rulemaking Alternative, 35 San Diego L. Rev. 219, 248 (1998) (stating that the Copyright Act imposes strict liability for unauthorized copying). But see Steven Hetcher, The Immorality of Strict Liability in Copyright, 17 Marq. Intell. Prop. L. Rev. 1, 2 (2013) (arguing that although it has historically been accepted that copyright is a strict liability scheme, due to the fair use doctrine, it is actually closer to a fault standard).
  226. .R. Anthony Reese, Innocent Infringement in U.S. Copyright Law: A History, 30 Colum. J.L. & Arts 133, 133–34 (2007) (detailing the history of innocent liability in the United States).
  227. .Id. at 133.
  228. .17 U.S.C. § 501(a).
  229. .See Paycom Payroll, LLC v. Richison, 758 F.3d 1198, 1204 (10th Cir. 2014) (quoting Julie E. Cohen et al., Copyright in a Global Information Economy 291 (3d ed. 2010)) (stating that part of the analysis for determining copyright infringement is “whether the defendant copied from the plaintiff,” otherwise referred to as “copying in fact”).
  230. .Shyamkrishna Balganesh, The Normativity of Copying in Copyright Law, 62 Duke L.J. 203, 205 (2012) (emphasis omitted); see also Christopher A. Cotropia & Mark A. Lemley, Copying in Patent Law, 87 N.C. L. Rev. 1421, 1427 (2009) (“[A]ctual copying is still a fundamental first step in determining copyright infringement.”).
  231. .E.g., Laureyssens v. Idea Grp., Inc., 964 F.2d 131, 139 (2d Cir. 1992) (“It is now an axiom of copyright law that actionable copying can be inferred from the defendant’s access to the copyrighted work and substantial similarity between the copyrighted work and the alleged infringement.”); Data East USA, Inc. v. Epyx, Inc., 862 F.2d 204, 206 (9th Cir. 1988) (stating that defendant’s access to the copyrighted work and substantial similarity between the copyrighted work and allegedly copied piece were together sufficient to constitute copyright infringement). But see Sid & Marty Krofft Television Prods., Inc. v. McDonald’s Corp., 562 F.2d 1157, 1162–63 (9th Cir. 1977) (stating that relying on circumstantial proof of access combined with proof of similarity would be “untenable” and would go beyond “the scope of copyright protection”).
  232. .See O’Keefe v. Ogilvy & Mather Worldwide, Inc., 590 F. Supp. 2d 500, 515 (S.D.N.Y. 2008) (“In order to support a claim of access, a plaintiff must offer significant, affirmative and probative evidence that an alleged infringer had a reasonable probability . . . of being exposed to his work.” (internal quotation marks omitted)); Nimmer & Nimmer, supra note 73, at § 13.02(A) (“[E]ven if evidence is unavailable to demonstrate actual viewing, proof that the defendant had the opportunity to view (when combined with probative similarity) is sufficient . . . to conclude that copying as a factual matter has occurred.” (footnotes omitted)).
  233. .See O’Keefe, 590 F. Supp. 2d at 517–18 (finding that the plaintiff needed to establish substantial similarity among unique features in order to sustain a copyright-infringement action); Nimmer & Nimmer, supra note 73, at § 13.03(A) (stating that substantial similarity must exist between the copyrighted work and the allegedly copied work to sustain a copyright action).
  234. .See Selle v. Gibb, 741 F.2d 896, 901 (7th Cir. 1984) (“[T]wo works may be identical in every detail, but, if the alleged infringer created the accused work independently . . . then there is no infringement.”); Gideon Parchomovsky & Alex Stein, Intellectual Property Defenses, 113 Colum. L. Rev. 1483, 1494 (2013) (“If, perchance, a person independently creates a work that is identical to a preexisting work, she will not be liable for copyright infringement. Rather, both works would be entitled to copyright protection.”).
  235. .See Ty, Inc. v. GMA Accessories, Inc., 132 F.3d 1167, 1169 (7th Cir. 1997) (stating that “[t]he Copyright Act forbids only copying; if independent creation results in an identical work, the creator of that work is free to sell it”).
  236. .See supra notes 28–33 and accompanying text.
  237. .See supra notes 32–33 and accompanying text.
  238. .17 U.S.C. § 504(a) (2012).
  239. .Id. § 502(a).
  240. .Id. § 503.
  241. .Id. § 504(d).
  242. .Id. § 505.
  243. .Id. § 504(b).
  244. .Id.
  245. .Id.
  246. .Id. § 504(d). The statute does not use the term “punitive,” though the punitive nature of the award is plain.
  247. .Id. § 110(5).
  248. .Id. § 110(5)(B)(i).
  249. .Id. § 110(5)(B)(i)(II).
  250. .Id. § 504(d).
  251. .Id.
  252. .See id. § 504(c) (providing that the copyright owner may “elect” to recover statutory damages instead of actual damages and profits).
  253. .Id. § 504(c)(2).
  254. .Id. § 504(c)(1).
  255. .Id. § 504(c)(2).
  256. .Culpability is relevant to criminal actions under the Copyright Act. See id. § 506 (requiring willful infringement for criminal offenses).
  257. .Id. § 504(c)(3).
  258. .Samuelson & Wheatland, supra note 23, at 441.
  259. .17 U.S.C. § 504(c)(2).
  260. .Id.
  261. .See, e.g., John E. Calfee & Richard Craswell, Some Effects of Uncertainty on Compliance with Legal Standards, 70 Va. L. Rev. 965, 995 (1984) (noting that when the “probability [of liability] declines as defendants take more care, then defendants may tend to overcomply”); Richard Craswell & John E. Calfee, Deterrence and Uncertain Legal Standards, 2 J.L. Econ. & Org. 279, 289 (1986) (stating a defendant’s “reason for overcomplying is to reduce the likelihood of being found liable”); A. Mitchell Polinsky & Steven Shavell, Punitive Damages: An Economic Analysis, 111 Harv. L. Rev. 869, 873 (1998) (observing that “if injurers are made to pay more than for the harm they cause, wasteful precautions may be taken . . . and risky but socially beneficial activities may be undesirably curtailed”).
  262. .On the connection between fair use and transaction costs, see Wendy J. Gordon’s classic, Fair Use as Market Failure: A Structural and Economic Analysis of the Betamax Case and Its Predecessors, 82 Colum. L. Rev. 1600, 1628–30 (1982), for a discussion on high transaction costs as a prerequisite to a fair use finding.
  263. .For excellent discussion, see generally James Gibson, Risk Aversion and Rights Accretion in Intellectual Property Law, 116 Yale L.J. 882 (2007) and Mark A. Lemley, Property, Intellectual Property, and Free Riding, 83 Texas L. Rev. 1031 (2005).
  264. .See Oren Bar-Gill & Gideon Parchomovsky, The Value of Giving Away Secrets, 89 Va. L. Rev. 1857, 1858–61 (2003) (discussing the effect of strategic holdups on bargaining between rightsholders and potential improvers in patent law). The discussion applies with equal force to negotiations over copyrights.
  265. .Any overinvestment in precautions is socially wasteful. See Guido Calabresi, The Costs of Accidents 29 (1970) (stressing that a point exists at which precautionary measures will “incur[] costs in achieving the reduction that are greater than the reduction is worth”).
  266. .But see Eduardo Moisés Peñalver & Sonia K. Katyal, Property Outlaws, 155 U. Pa. L. Rev. 1095, 1146–47 (2007) (arguing that blatant and repeated infringement of property rights provides a valuable social service by pointing out legal areas in need of change).
  267. .See supra subpart III(A).
  268. .See supra subpart I(B).
  269. .See supra subpart II(B).
  270. .See supra subpart III(B).
  271. .Depoorter, supra note 47, at 419.
  272. .Id. at 410–11.
  273. .Samuelson & Wheatland, supra note 23, at 485 & n.238.
  274. .17 U.S.C. § 504(c)(2) (2012).
  275. .See Lieb v. Topstone Indus., Inc., 788 F.2d 151, 156 (3d Cir. 1986) (“Factors which should play a part [regarding copyright-infringement remedies] include . . . the need in particular circumstances to advance considerations of compensation and deterrence.”); Samuelson & Wheatland, supra note 23, at 445 (“The application of statutory damages . . . has focused too heavily on deterrence and punishment by holding many ordinary infringements to be willful, which has resulted in many awards that are punitive in effect and often in intent.”).
  276. .Restatement (Second) of Torts § 500 (Am. Law Inst. 1965).
  277. .See Mark A. Lemley & Christopher R. Leslie, Categorical Analysis in Antitrust Jurisprudence, 93 Iowa L. Rev. 1207, 1259–63 (2008) (discussing the costs of courts’ categorical analysis in antitrust cases and mentioning, inter alia, error costs, determination costs, and litigation costs); see also Peter M. Gerhart, Property Law and Social Morality 42–44 (2014) (discussing categorization costs on the ground of an information-costs analysis).
  278. .See Lemley & Leslie, supra note 277, at 1213 (discussing the modes of antitrust analysis); Thomas A. Piraino, Jr., Reconciling the Per Se and Rule of Reason Approaches to Antitrust Analysis, 64 S. Cal. L. Rev. 685, 685 (1991) (discussing the per se and rule of reason approaches); Richard M. Steuer, Indiana Federation of Dentists: The Per Se–Rule of Reason Continuum (and a Comment on State Action), 8 Cardozo L. Rev. 1101, 1101 (1987) (“[T]he Supreme Court labored at defining two categories of antitrust offensesthose that were illegal per se and those that violated the ‘rule of reason.’”).
  279. .Robert H. Bork, The Rule of Reason and the Per Se Concept: Price Fixing and Market Division, 75 Yale L.J. 373, 391 (1966) (“[A]ll horizontal price fixing and market division is illegal per se . . . .”).
  280. .See Leegin Creative Leather Prods., Inc. v. PSKS, Inc., 551 U.S. 877, 908 (2007) (Breyer, J., dissenting) (“Today the Court holds that courts must determine the lawfulness of minimum resale price maintenance by applying, not a bright-line per se rule, but a circumstance-specific ‘rule of reason.’”). By ruling so, the court had overruled the Dr. Miles rule that remained in force for 96 years. Id.
  281. .Piraino, supra note 278, at 685 (“Thus, pro-competitive conduct should receive the benefit of the doubt under a rule of reason that considers all of its possible justifications and beneficial effects. Anticompetitive practices, on the other hand, should be summarily condemned under the per se rule . . . .”).
  282. .Christopher Sprigman, Copyright and the Rule of Reason, 7 J. on Telecomm. & High Tech. L. 317, 325 (2009) (“[T]he introduction of a per se‌/rule of reason distinction into copyright law would incentivize the production of more information about the incentive effects of a variety of uses of copyrighted works.”).
  283. .Amy Adler & Jeanne C. Fromer, Taking Intellectual Property into Their Own Hands, 107 Calif. L. Rev. 1455, 1505–06 (2019).
  284. .One can argue that to a large degree this is the law today: bodily appropriators of a considerable percentage of a copyrighted work would find it very difficult to argue fair use successfully. See Pierre N. Leval, Toward a Fair Use Standard, 103 Harv. L. Rev. 1105, 1132 (1990) (“[T]he fair use test is difficult to predict. It depends on widely varying perceptions held by different judges.”).
  285. .See Justin Hughes, Size Matters (Or Should) in Copyright Law, 74 Fordham L. Rev. 575, 636 (2005) (“When liability for copyright infringement boils down to copying a name, a couple of choice phrases, a slogan, or a small subset of numeric evaluations, copyright law is being dragged by clever lawyers into dark alleys where it should not go.”); Leval, supra note 284, at 1123–24 (discussing short literary work); Gideon Parchomovsky & Kevin A. Goldman, Fair Use Harbors, 93 Va. L. Rev. 1483, 1517 (2007) (discussing visual works).
  286. .See Abraham Bell & Gideon Parchomovsky, The Dual-Grant Theory of Fair Use, 83 U. Chi. L. Rev. 1051, 1085 (2016) (describing political speech and truth seeking as “privileged categories” in a copyright-infringement analysis).
  287. .Reese, supra note 226, at 180–81.
  288. .17 U.S.C. § 107 (2012).
  289. .More Information on Fair Use, U.S. Copyright Off., https://‌www.copyright.gov‌/fair-use‌/more-info.html [https://‌perma.cc‌/8T9J-ML59] (last updated Sept. 2019) (“[O]ther factors may also be considered by a court in weighing a fair use question, depending upon the circumstances.”).
  290. .471 U.S. 539 (1985).
  291. .Id. at 542, 563.
  292. .Id. at 542.
  293. .Id. at 562–63.
  294. .Id.
  295. .Campbell v. Acuff-Rose Music, Inc., 510 U.S. 569, 579 (1994) (“[T]he more transformative the new work, the less will be the significance of other factors . . . .”).
  296. .Leval, supra note 284, at 1126–27.
  297. .Id. at 1126 (“The inquiry should focus not on the morality of the secondary user, but on whether . . . the secondary use is productive and transformative . . . .”).
  298. .John Mullins & Randy Komisar, A Business Plan? Or a Journey to Plan B?, MIT Sloan Mgmt. Rev., Spring 2010, at 1, 2.
  299. .See Arthur Best, Impediments to Reasonable Tort Reform: Lessons from the Adoption of Comparative Negligence, 40 Ind. L. Rev. 1, 2 (2007) (“[Contributory negligence] completely precluded a plaintiff’s recovery in any case where the plaintiff and defendant were both negligent.”).
  300. .See Li v. Yellow Cab Co. of Cal., 532 P.2d 1226, 1229 (Cal. 1975) (“The doctrine of comparative negligence is preferable to the ‘all-or-nothing’ doctrine of contributory negligence from the point of view of logic, practical experience, and fundamental justice . . . .”).
  301. .17 U.S.C. § 412 (2012).
  302. .See, e.g., Pamela Samuelson & Members of the CPP, The Copyright Principles Project: Directions for Reform, 25 Berkeley Tech. L.J. 1175, 1194–95 (2010) (“[P]rinciples of primary and secondary liability should continue to play a role in encouraging technology and service providers to participate in deploying reasonable measures and discouraging widespread infringement.”); Sprigman, supra note 282, at 336–37.
  303. .A recent example is a case in which two attorneys threatened to sue thousands of people who downloaded pornographic films without permission in order to profit from settlements. The lawyers filmed some of these movies and uploaded them to file-sharing websites with the intention of enticing the alleged violators. See Plea Agreement and Sentencing Stipulations at 2, United States v. Steele, No. 16-334 (JNE/KMM) (D. Minn. Mar. 6, 2017).
  304. .See Brad A. Greenberg, Copyright Trolls and the Common Law, 100 Iowa L. Rev. Bull. 77, 86 (2015) (suggesting a judicial application of the misuse doctrine on copyright-trolling cases). More generally on the evolution of the copyright misuse doctrine, see Kathryn Judge, Rethinking Copyright Misuse, 57 Stan. L. Rev. 901, 905–12 (2004) and David S. Olson, First Amendment Based Copyright Misuse, 52 Wm. & Mary L. Rev. 537, 570–83 (2010).
  305. .See Jason Mazzone, Copyfraud, 81 N.Y.U. L. Rev. 1026, 1090 (2006) (“This limit on the defendant’s ability to assert the misuse defense will permit much copyfraud to survive. . . . [C]ourts should consider loosening the nexus requirement.”).