A Transactional Theory of the Reader in Copyright Law

I.       Introduction

“Ultimately, the Blurred Lines case isn’t so much about the scope of copyright protection . . . It’s about the strange, unpredictable entity that is the American jury doing whatever it is an American jury does while we’re not looking.”

Keith Harris, music critic1

In 2015, a highly publicized copyright infringement lawsuit over the allegedly infringing song, Blurred Lines, led a jury to award the heirs of Marvin Gaye’s estate $7.3 million in damages.2 The case was noteworthy because of the notoriety of the song, the large amounts of money at stake, and the celebrity litigants. Yet for copyright law, it also represented a problematic allocation of authority to jurors on questions they were ill-suited to resolve, and poorly instructed to answer.3 First, the court allowed jurors to compare the works of Gaye and the Williams–Thicke team without clearly delineating what was copyrighted (Gaye’s composition only) as against the many elements that were not copyrighted in this case (Gaye’s falsetto, and any other performance embellishments not present in the sheet music) or that could not be copyrighted (all the musical ideas, or stock elements common to soul or funk as a genre more generally).4 This is a problem of miscalibrating the scope of copyright protection and overprotecting things which the defendant did not, or could never protect through copyright law, and the court properly acknowledged this error on appeal.5 Second, the court instructed jurors using 43 complex, unclear, and arguably incorrect instructions that further contributed to the doctrinal muddle.6 Indeed, the case probably presented questions more appropriate for a judge than a jury, and it may have been reversible error to send it to the jury in the first place.7

Though this case was exceptional in the sense of the fame and money involved, it was all too representative of copyright’s inconsistent operation with respect to juries. Most copyright cases do not wind up in jury deliberation, perhaps partly because of the problem of presenting juries with specialized legal questions they are ill-equipped to answer.8 Yet those cases that do wind up in jury deliberation display common problems in their assumptions about what jurors can do, as this Article will argue below. By ignoring the realities of how people actually engage with art, courts overestimate what jurors can do, and underestimate their need for guidance on both law and how to engage with art. Consequently, courts withhold the guidance juries need and mistake the proper limits of juror decision-making.

Copyright doctrine requires judges and juries to apply legal doctrines to works of art to decide whether works have been infringed. These arbiters therefore must engage in some form of experiencing or “reading” the works. However, despite the central role this reading—or viewing and listening—plays in copyright disputes, copyright law lacks a robust theory of the act of reading, and of the proper role for the “reader.” In this sense, copyright cases seem to rely on naïve intuitions about reading. Yet copyright cases require a particular kind of reading that can be better understood by understanding general theories of reading better.

It has been said that we do not read the newspaper, but rather step into it like a hot bath.9 Notwithstanding the homespun idea that reading is a comforting, easy, perhaps passive activity, psychologists and scholars of literature who spend time developing models for how humans read point to the complex social, biological, and emotional dimensions of reading. As an activity, it is highly complex, and composed of many active steps readers may take for granted or forget about entirely once they have become adept at it. It unfolds through time and evolves; it is, in the words of one scholar, “more like playing a game of tennis than solving a math problem, in that reading and playing can be assessed differently at different points in time.”10 Some readers may never have thought through just how much they do every time they experience reading, though for parents who witness young children acquiring literacy over time, the complexity is striking. Early readers may use their body (squinting, tracing a finger, moving lips) to do what later on they may think of as a mental, not physical practice.11 They may avoid reading and resort to memorization. They may shut down when they “fail” to decode something. They may rely on rhymes and guesses to detect words they do not yet know.12 The process is complex even when readers are decoding works that are purely or primarily informational, such as signs and instruction manuals. The complexity of the process becomes clearer still in light of issues such as counterintuitive spelling, homonyms, and puns. Adding to that, when readers confront literary works of art, they must make sense of tone, metaphors, narrative, allusion, and other issues pertaining to meaning. A work of art contains “stimuli” but it is also urging its recipient “toward a special kind of intense and ordered experience—sensuous, intellectual, emotional—out of which social insights may arise.”13 Reading or encountering art adds affective complexity to the existing cognitive challenges of reading.14 This experience is one that has an impact on readers, emotionally, and perhaps physically, too.15 Reading is a highly idiosyncratic activity that benefits from empirical and theoretical accounts that unpack its psychological, cognitive, and physical elements.16

It turns out that experiencing art other than literary works, such as music, film, photography, theater, and dance is just as complex.17 For instance, the same problems of how to make sense of the experience persist even when the allusions are to musical patterns rather than to phrases of poetry; when the narrative is conveyed through choreographic movement or film editing rather than sections of a novel; and when the metaphors are rendered through visual symbols rather than words. Aesthetic and psychological theories have developed the particular issues in each of the relevant artistic fields,18 and broader theories of artistic reception have also established some of the commonalities inherent in the experience of art, whatever the artistic medium.19 Reader response theory, in particular, has offered a bridge from theories of textual reading to the practice of receiving other forms of art, to concentrate on how the “reader”—or viewer, or listener—actively participates in making meaning as she perceives the work of art.20

A theory of reading is necessary to, and missing from, copyright law because the kind of complex reading this Article has been describing plays an important role in determining outcomes. Both judges and juries are called upon to “read” works in particular ways in the course of copyright litigation. And how they read matters to individual cases.

For example, finding copyright infringement when two works are not identical requires a determination of the works’ similarity. If the works are not identical, and not similar, there is no plausible infringement case. Since not all similarities are actionable under copyright law, determining whether two works are similar requires drawing some legal conclusions about what to count as protected and unprotected. This determination, in turn, requires some basic non-legal decisions about how to define the boundaries of the work, what methods to use to determine similarities, and what evidence to use to inform both of those judgments.21 Though these decisions end up having legal significance, they are more properly identified as aesthetic or interpretive choices.22

Among the most important decisions about how to read is the decision of whose perspective to adopt. Should one make these determinations imagining oneself to be an expert in the relevant field (and perhaps rely on expert evidence for assistance), or should one adopt a nonexpert perspective? On the surface, copyright might be said to instruct readers clearly: in most cases, the law demands that courts and juries apply the “ordinary observer” standard when determining whether works are sufficiently similar to be infringing.23 In the Second Circuit Court of Appeals, the ordinary observer standard is actually a test, and it has been adopted by other circuits as well.24 In some other jurisdictions, the perspective of an ordinary, or “average,” or “lay” observer influences judicial analysis in other important ways, even absent a formal test.25 Even in the Ninth Circuit Court of Appeals, where the test for substantial similarity differs in name, the structure is the same: two steps, with analysis in the first step (including by experts), and subjective impressions of the work in the second (by the jury).26 Thus reading matters in copyright cases in at least two crucial ways: first, many courts rely on the “ordinary observer” standard to determine infringement, which requires figuring out or assuming how an ordinary observer would read the works at issue. Second, most courts characterize a key part of infringement analysis as a matter for the jury, largely on the basis of the jury’s ability to adopt the perspective of the ordinary observer. In spite of the centrality of the ordinary observer concept, however, it has not received much attention as a feature in copyright law.

Thus even though copyright does not have a robust theory of reading, it definitely has a reader: the ordinary observer, whose amorphous role is poorly understood and inconsistently applied. No scholarship has attempted to articulate a broad theory of what copyright case law’s judge and jury do as readers when they address the question of infringement. In some accounts, the work of determining improper appropriation consists merely of comparison of two works, set side by side, and it seems as though any person could do it.27 In other accounts, the process looks something more like a balancing analysis, analogous to application of tort law’s reasonable person standard which is, of course, a long-established legal fiction that the jury or factfinder applies on the basis of everyday experience.28 Under this tort-like account, the factfinder compares two works, and weighs their similarity from the perspective of a hypothetical “ordinary observer.”29 The purpose of the hypothetical observer is not to capture, empirically, how an actual audience member would react to a work, but rather to approximate what seems representative of reasonable behavior in a particular community at a particular time.30 At other times, however, the ordinary observer starts to resemble the empirically constructed consumer in trademark law, albeit with an important difference. Trademark law depends on a concept of the consumer because trademark infringement is largely determined by whether a consumer is actually or likely to be confused by competing marks.31 The consumer reflects an aggregate of responses collected through surveys administered by the parties.32 The consumer in trademark law, therefore, may be used to assess whether two marks are so similar that their similarity is likely to confuse consumers to the point where the original mark owner’s market will suffer. To be clear, copyright does not look at actual readers’ responses to works, the way trademark would, but in some cases it does look at intended audiences for the work, adjusting the ordinary observer standard based on the works or markets involved. In such cases, courts characterize the audience by describing the attributes of the average audience for a work. For instance, one case held that the average purchaser for a martial arts video game was male, likely to be 17.5 years old, “knowledgeable, critical, and discerning.”33 In that sense, when copyright takes empirical evidence about an author’s intended audience or actual market into account, it resembles trademark law. As this distinction between the tort and trademark-like accounts shows, sometimes the ordinary observer is wholly imagined (one might say it is general, and prescriptive or aspirational of what is reasonable), and sometimes it is based on empirical reality (particular, and descriptive of what is actual).

Because copyright in the typical case does not rely on any empirical evidence about the ways actual individual readers “read” works—and in many instances excludes any such evidence when it is available—there is no “real” reader on which to base our understanding of the role of the reader in copyright law.34 Hence the need for a theory to describe what judges and factfinders do in the course of the infringement inquiry is all the more pressing.35 This Article relies on a law-and-humanities approach to propose that reader response theory, a body of scholarship that arose in literary studies, can help us better understand what readers of art actually do. Using reader response theory to describe reading, in turn, allows for a better normative theory of copyright’s reading practices to emerge, and helps build the case for what tasks are better suited for the judge or jury, when expert evidence should be admitted, how to improve jury instructions, and how to reduce the “black-box” lack of transparency around jury findings of substantial similarity, given that they are often found to be improper or erroneous after the fact.

Part II asks readers to engage in a brief thought experiment to experience different kinds of reading for themselves. Part III begins by assessing the ordinary observer’s role in copyright law, and it demonstrates how copyright law lacks cohesion in its reading practices, especially around the shifting, amorphous purposes to which courts put the ordinary observer. It argues that the operation of the ordinary observer standard is inconsistent as applied by courts, and often unrealistic in its operation in the demands it places on jurors. Part IV delves into reader response theory to expose a surprising gap in copyright scholarship, which has mostly overlooked this body of work. Though scholars have quite fully explored theories of the author,36 and there have been some articles on theories of interpretation and aesthetics, broadly speaking,37 there has been only very modest attention to readers in copyright in light of humanistic theories of reception.38 Louise Rosenblatt’s theory surfaces in the context of reader response as one of the most practical and applicable of the theories of reception developed in the humanities. Part V provides an elaboration on the differences between the kinds of reading readers experience, and it uses the thought experiment from Part II to show how the kind of reading being done places certain demands on the reader. Part VI applies Rosenblatt’s theory to copyright law’s infringement doctrine to illustrate the need for a clearer division of labor between judge and jury, to suggest that more reliance on expert evidence may be a good idea, to call for greater attention to what is being asked of the jury as it serves the needs of copyright litigation, and to propose that courts use special verdicts to guide and cabin jury deliberation.

II.     A Thought Experiment

Copyright lacks, but needs, a coherent theory of its reader. In order to offer such a theory, this Part sets forth a brief, three-part thought experiment. Please read the following descriptions of three works of art, in preparation for experiencing them for yourself. This thought experiment is directly relevant to the thesis of the Article, so your participation will help you assess the merits of the Article’s conclusion and proposal.

First, a song, entitled La Dernière Minute, by Carla Bruni.39 A sultry female voice half-sings, half-whispers in French, for exactly 60 seconds.40 The lyrics tumble out rapidly, staccato and urgent.41 In the background, the sound of a drum ticks like a clock, and the words seem to be delivered as though in a race against time.42 “Juste encore une minute,” French for “just one more minute,” recurs as a refrain, and the rest of the lyrics discuss the singer’s sense of panic about the passage of time.43 The song is an entreaty for just one more minute to experience life, from the perspective of the singer when she imagines herself at death’s doorstep, regretting not having gotten more out of her life.44 The song’s melody is simple and repetitive, and the acoustic guitar accompaniment insistent, but minimalist. One reviewer calls the album from which the song is drawn, “somewhere between singer-songwriter and a dinner-jazz version of Django Reinhardt: a guitar equivalent of Norah Jones, sung in French.”45 The style is part dinner lounge, part folk, layered with soft vocals by a voice that sounds as though it lacks firepower, and wisely chooses to deliver silk and smoke instead.

Second, a poem, entitled Turbulence, by the late Adrienne Rich, a highly respected American poet.46 An unknown narrative voice uses the conceit of airplane turbulence to discuss the experience of coping with emotional loss. The poem moves between different registers of language—from literal to metaphorical and back again—and it includes some language borrowed and reworked from the Victorian poet, Gerard Manley Hopkins.47 The language goes from being contemporary and easy to understand to archaic and difficult. Some of the poem’s language seems out of place, and the allusion to Hopkins helps explain why the language changes back and forth in tone and register. The language shows that Rich intentionally drew on Hopkins’ poem, which is one of his “Terrible Sonnets,” known to have been written during a period of great personal depression for him.48 Rich’s work often uses a complex metaphor for the process of managing emotions such as grief, loss, love, and anger. Her choice of airplane turbulence as that complex metaphor plays out in imagery of heights, falling, breaking, and oxygen loss, all effectively deployed to focus the reader’s attention on how it feels, physically and emotionally, to experience loss.

Third, a self-portrait by Jacob Lawrence, an influential African-American visual artist.49 The painting depicts the artist himself at work in a surreal version of his studio in Seattle. There are at least seven shades of light brown for the wood of the walls and the ceiling, conveying warmth and a pleasantly uneven quality to the color and light in the space. The figure of the artist rests his left arm on a large fire-engine red wooden railing as he stands in the stairwell just beneath the studio. The railing cuts a striking diagonal line down through the tableau. The artist wears eyeglasses and a shirt in shades of bright blue, and several of his paintings are propped or hung in view, attesting to a productive and experienced artist, someone whose work takes up room in his life. His figure is somewhat skeletal, with a bony face and jaw. His right hand holds paintbrushes; his left holds a drawing compass. The painting’s use of color is bold and cheery; even the cityscape visible through the window set in the background center of the painting features little glimmers of yellow, terra cotta brown, and pale silver in the far-off architectural details. Above the skyscape, the view through the window is a dull mass of heavy grey fog, with a dark night sky behind it creating depth and distance underneath strokes of lightly textured grey. The scene through the window depicts the artist’s native Harlem, suggesting the stylized nature of the depiction since Seattle and Harlem are the artist’s two home cities, but they lie far apart geographically. For those trained to recognize such styles, the work will evoke cubism, and the artists of the Harlem Renaissance school who influenced Lawrence,50 with the painting’s sharp edges, broad, flat colors, and intense shapes and shades.51 For those who know Lawrence’s work, the depiction of his paintings within the tableau, which show images of carpentry and architecture, will echo insistent themes from the rest of his body of work.

The first part of this thought experiment is over. If pressed, you, reader, could now tell another listener what each of these works was, on some general level, “about.” You have derived information about these works sufficient for a basic familiarity with them. In one case, you’ve been informed about the relationship of the work to the creator’s other works (the Lawrence painting and its carpentry themes); in another you’ve been informed of the allusion one work makes to another pre-existing work (the Rich poem, vis-à-vis Hopkins). And in the case of the song, you’ve been given the names of other artists whose work may be familiar to you, and thus may assist you in conjuring up this work (the Bruni song). You’ve been given a summary of the texts’ features, and a small amount of context in which to situate those texts. With more information provided to you about the authors, or the genres, or these particular works, you would increase your knowledge of the works, but you would still not have experienced the works for yourselves.

Here comes the second part of the thought experiment. Please listen to the song, chosen in part for its exhilarating brevity—60 seconds—and read the poem once or twice; it, too, is short, consisting of a tightly wrought 14 lines. Finally, take a few moments to view the painting. All are available, or cited to, in the Appendix. The original Hopkins poem is there as well, in case it interests you to read that one too. If you plan to read the rest of this Article, you will find it worthwhile to have spent the time experiencing these works for yourself, since parts of the argument below refer to the works again. When you have had the full experience of your own encounter with each of the works, making even the most basic sense of each of them for yourself, you have completed the second step in the thought experiment.

The third and last step asks you to return, briefly, to each of the works, in order to derive information from them sufficient to answer the following questions:

How many times does Bruni sing the title phrase “dernière minute”?

Make a note of all verbs Rich uses that are acts of destruction or damage.

Where in the painting are the straight lines perfectly straight, versus appearing to be hand-drawn and imperfect?

When you have completed this quicker second reading, you will have fully participated in two kinds of engagement with artistic works described in a theory of reading developed by Louise Rosenblatt.

After reading brief introductory material designed to assist your reading, you performed the first kind of reading, from a phenomenologically (or experientially) oriented stance that Rosenblatt calls “the aesthetic stance.” You, as an active reader, experienced a particular, idiosyncratic version based on what you brought to the text and the cues that the text contains, which you noticed and “activated.” This Article leveled the playing field somewhat with the background information that many readers find helpful before encountering a work, so that certain cues would be more available to your reading. Then in the final step of the experiment, you performed a different kind of reading, which was largely informational and instrumental, not experiential.52 Rosenblatt calls this reading from “the efferent stance.” To use Rosenblatt’s own explanation, it is efferent because the reader “bears away” information (efferre is Latin for to bear or carry something away). This Article explores these stances in further detail below, and it is helpful for you to have experienced them, and the three central works of art, for yourself first.

These differences in how you approached the work are relevant both to theories of reading and, as this Article will argue, to copyright’s reading practices. Rosenblatt’s reader response theory was unique and prescient in recognizing that readers are particular, not general, and in separating the aspects that could be deemed structural/‌analytical or objective from those deemed intuitive or subjective. Her separation of efferent and aesthetic reading emphasized that the former was a kind of work that could be done with consistency and conformity while the latter was one whose subjective, idiosyncratic nature was built into the very experience. Further, Rosenblatt advocated for a view of reading as a transaction between reader and text a good half century before most other theorists were emphasizing the active role readers play. When the literary academy did finally embrace reader response theory, it did so with less nuance than Rosenblatt, generally holding that either texts, or readers, contain dispositive power over what a text means. Eschewing the polar positions to either side, namely, that meaning was entirely text-controlled or entirely reader controlled, Rosenblatt argued that the reader created a third thing in evoking the work, something idiosyncratic to the reader but nonetheless faithful to the text because constrained by cues it offered the reader.53

This Article argues that Rosenblatt’s theory of reading could play a part in reforming copyright litigation. Rosenblatt’s work was “radically interdisciplinary, shattering all sharp borderlines dividing philosophy, the social sciences, and literary criticism and pedagogy.”54 This interdisciplinary approach offers a better understanding of how audiences—including observers and readers—actually experience works of art, even in the expressly teleological contexts of litigation. Although Rosenblatt’s model was developed for verbal texts, it can be argued that it should apply broadly to all works of art on the basis of the existing scholarship that has extended reader response and reception theories to forms of art other than literature. If Rosenblatt’s theory is more accurate for describing how audiences, and thus jurors, experience works of art, it may improve predictions for juror interaction with art, and it could inform jury instructions to increase the likelihood that jurors do what they are asked, and are asked to do what they can.

III.     An Assessment of the Ordinary Observer Standard in Copyright Law

This Part describes how the ordinary observer standard plays an important role in infringement analysis, especially in the many actions involving works that are not identical. The reader’s role in copyright law has not received much direct treatment as a concept in need of clearer articulation, and there has been little theoretical work on the larger question of how it actuates copyright’s purposes. Contemporary scholars characterize copyright as predominantly utilitarian: it exists to incentivize creation, and it does so by offering limited monopolies.55 Some courts acknowledge that the ordinary observer standard introduces into copyright’s infringement analysis a subjective, particularized element. This is the Fourth Circuit approach, which tailors substantial similarity to a work’s intended audience.56 The dominant approach, however, is to frame the ordinary observer from both an objective and generalized perspective, as the Second and Ninth Circuits do.57 Confusingly, however, the objective, generalized take on the ordinary observer, which any reasonable juror is meant to be able to apply, is also subjective in the nonlegal sense of intuitive, impressionistic and gestalt as opposed to analytical and dissective.58

Courts applying the intended audience approach root their reasoning in an economic rationale, seeking to protect an author’s market.59 Losing sight of the market rationale, under that view, “would be to allow the imprecise ‘ordinary lay observer’ label to effect a betrayal of the fundamental purposes of copyright doctrine.”60 Accordingly, in their own review of the contested work, such courts attempt to approximate the experience of an actual purchaser, for instance, by listening to a song from start to finish.61 Only some courts apply the ordinary observer standard as though it meant the actual purchaser; others proceed as though the observer simply means a member of the jury.62 Hence folded into the operation of the ordinary observer standard are confusing terms, unclear notions about what jurors are supposed to do, unrealistic ideas about what jurors can do, and resulting confusion about which parts of copyright law are properly reserved for judge or jury. The muddle around the ordinary observer standard contributes to the general mess in substantial similarity law, and thus frustrates coherence in copyright law generally.

A.     Background for the Ordinary Observer in Copyright Litigation

It may be helpful to provide a little background for nonspecialists to see the role that the ordinary observer plays in copyright’s overall infringement analysis. In a copyright infringement action, a plaintiff must prove that he owns a valid copyright and that the defendant copied protected parts of that work.63 If the defendant’s work is identical, the plaintiff’s task will be easier because the similarity between the works will be indisputable rather than needing to be asserted and proven.64 In many cases, however, the defendant’s work is not identical, and only partially resembles the plaintiff’s. Assuming the defendant did copy the plaintiff’s work, the court’s task will be to assess whether what the defendant copied is protected by copyright. If it is, the court will find substantial similarity, which is a legal conclusion.65 The standard for arriving at that conclusion is “whether the accused work is so similar to the plaintiff’s work that an ordinary reasonable person would conclude that the defendant unlawfully appropriated the plaintiff’s protectible [sic] expression by taking material of substance and value.”66

Substantial similarity is the core substantive requirement in every copyright infringement action67 that does not involve literal similarities or identical works.68 On the surface, copyright infringement’s main analysis is straightforward: assessing substantial similarity would seem to require no more than a comparison of the two works’ to determine their similarity.69 Despite its first-glance simplicity, substantial similarity is one of the hardest conclusions to draw in copyright law70 and remains for many courts “an elusive concept.”71 The party tasked with assessing similarity must apply a legal test in assessing the extent and nature of similarity. The doctrine is a challenge for many courts because many tests exist, there is insufficient consensus on which test to apply, and there is little guidance on the proper way to apply each test. At present, the tests for substantial similarity are largely driven by jurisprudence in the Second and Ninth Circuit Courts of Appeals.72 The tests from these two circuits dominate the case law in and beyond their own courts.73 In these two circuits alone, there are five main tests, including: (1) the abstractions test;74 (2) the abstraction-filtration-comparison test developed for computer software;75 (3) the improper appropriation test (whose second step relies on the ordinary observer test);76 (4) the total look and feel test;77 and (5) the extrinsic/‌intrinsic test.78

It is nominally one of these tests that concerns this Article the most—the ordinary observer test that occurs in step two of Arnstein’s improper appropriation test. However, as the next Part explains, Arnstein’s influence has extended beyond the formal test to a general two-step approach and an emphasis on intuition-driven jury determinations in the guise of the ordinary observer.

B.     Arnstein’s Legacy and the Ordinary Observer

Though precursor references to the ordinary observer test exist, Arnstein v. Porter offered a test that seemed to consolidate the existing legal reasoning around the lay experience, while adding an emphasis on market harm.79 Accordingly, Arnstein now serves as the influential source for the ordinary observer test in copyright law.80

Ira Arnstein was an experienced composer with a “persecution complex” who had, before this case, begun five copyright lawsuits, each claiming plagiarism and each failing to offer evidence in support of his assertions.81 This case, brought against the wildly successful American composer, Cole Porter, was no different.82 Arnstein made allegations that he had been followed, spied upon, and burgled, but when asked for evidence, he could point to nothing more than his gut instinct that these things had happened. He appeared pro se and demanded the ridiculous-on-its-face sum of $1 million in damages.83 In a case of bad facts making bad law, the court held that it was too soon to dismiss Arnstein’s complaint because these contested questions of fact should go to the jury. The court seemed sensitive to the fact that Arnstein might actually be suffering from mental illness, as prior litigants seem to have asserted against him.84 In the course of announcing that the case needed further fact finding by a jury, however, the court also severely cabined the use of expert evidence in a sweeping move that has, though somewhat haphazardly, established the reigning norm in copyright infringement cases today.85

Arnstein’s legacy has been mixed: the summary judgment portion of its holding has been overruled by the Supreme Court,86 but the contributions to copyright doctrine have become entrenched.87 In particular, two aspects of the opinion deserve focus: the first contribution is the two-step test for whether copyright infringement occurs.88Arnstein laid out a test that asked courts to determine, as a matter of law, first, whether copying had occurred, and second, as a matter of fact, whether that copying was improper copying.89Arnstein held that the test for whether the copying was improper should use the lay observer’s response to the litigated works as the standard for whether infringement took place.90 The test “attempt[s] to gauge the reaction of the ordinary ‘man on the street’ to the two works.”91 The factfinder, whether judge in a bench trial or jury in a jury trial, was to “decide[] whether an average lay observer would recognize the alleged copy as having been appropriated from the copyrighted work.”92 The Arnstein test, while modified, has been embraced in some form in most circuits.93 Arnstein’s two major contributions, therefore, have been its two-step analysis, and its focus on the ordinary observer.94

The Arnstein orthodoxy has been surprisingly steadfast. Arnstein’s author, Judge Jerome Frank, was a noted legal realist who would become most famous, perhaps, for his deep fact-skepticism and his view of judging as ineradicably idiosyncratic.95 Furthermore, he was, elsewhere in his jurisprudential writings, known for being a jury-skeptic who sought to resolve issues as matters of law whenever possible.96 This makes his ruling all the more surprising since it would seem to carve space for jury deliberation in any copyright dispute that clears substantial similarity’s first hurdle, namely, copying. What is more likely is that Judge Frank ruled this way in this case, on these facts, to produce a particular outcome—a method of adjudication very much characteristic of him. That Arnstein’s logic has endured is nonetheless plain, even as courts have struggled to amend parts of it in the face of what have appeared to them to be difficult facts or subject matter areas. As courts have tweaked the infringement test and, in particular, the way the audience or observer is to be understood, they have done much to muddy the jurisprudential waters further, and little to clarify them.

It has not escaped notice that the law of substantial similarity is a mess. Riddled with inconsistencies and manifestly illogical in certain respects, its doctrine contains many areas in need of improvement, from the test to apply and the method for applying it to the role and admissibility of expert evidence. Many of the “tests” for determining substantial similarity are scarcely tests, but malleable standards, unclear dicta, or unconstrained space for the impressions of the judge or jury. The ordinary observer is perhaps one of the most unwieldy aspects at the heart of substantial similarity analysis. Turning to the humanities provides insights into how the ordinary observer standard masks important processes of reading that ought to be unpacked, disaggregated, and better allocated.

IV.     Reader Response Theory and Its Relevance for Copyright Law

This Part briefly introduces reader response theory and then delves into the work of Louise Rosenblatt, a somewhat overlooked scholar whose work deserves greater attention and influence in the legal academy.97 As the reader will recall from Part I, Rosenblatt conceived of reading as a “transaction” between readers and texts, a subtle dialogue that distributed power between the work and its interpreter. Rosenblatt’s transactional theory of reading is grounded in empirical evidence culled from her long career of teaching literature. That her theory is so phenomenologically oriented makes it somewhat unusual in the broader reader response movement, and ultimately, more applicable to the real-world demands of copyright law discussed in Part V.

A.     Reader Response Theory and Louise Rosenblatt

Reader response theory can most helpfully provide guidance for copyright jurisprudence if it is set into a brief—indeed, telescopic—intellectual history that illustrates what was once radical about it. This overview of the critical landscape of literary studies will also highlight Rosenblatt’s unique position outside the canon of literary criticism. Her insistence on creating a theory of reading grounded in practical experience, which may have made her work less popular among academic elites, makes it especially helpful for addressing concrete problems in copyright litigation.

The literary world has witnessed major shifts in how it conceives of what it does in the study of literature and in the work of literary criticism.98 Before the 19th century, a dominant subset of Western scholars from Plato to M.H. Abrams had focused on the text as a kind of mirror of the real world.99 Approaches may have varied, since some scholars focused on philology, and others on myth or religion as themes one could discover within a work, but the primary emphasis was on the text as a thing that reflected the world back to its inhabitants.100 With Romanticism blooming in the late 18th century, the focus could be said—at a high level of abstraction—to have shifted from the text to the poet.101 In Professor Abrams’ famous formulation, instead of looking at the text as a mirror of the world, scholars were to look at the poet as the light that illuminates that world.102 Accordingly, critical inquiry tended to foreground intentionalist and biographical approaches to understanding a creative work since only through investigation of what an author intended, or what she might have experienced, could a work’s meaning come across fully.

Romanticism waned in the later 19th century, as the Victorian era emphasized art’s moral messages and didactic elements.103 Critics emphasized the text as an instrument of cultural education, and its author as the source of morally or religiously infused teaching. Sometimes this instruction was intended literally, as in the case of language instruction in British and French imperial expansion, and sometimes the ideas were more implicitly conveyed.104 These critical vicissitudes did not disturb the belief that the text lay at the center of inquiry in literary studies, and that it was a determinate object whose meaning could, through various techniques, be accessed by anyone, with a little effort. In this sense, the critical default could be said to be a general formalism, or an emphasis on the aspects of the work that were discoverable either through exegesis, or through other research and engagement with the work.

In the beginning of the 20th century, the literary world witnessed a turn towards a narrower kind of formalism known as the New Criticism. The New Critics cast the text as an object that could and should be understood apart from the authorial claims of the Romantic era, and the moral claims of the Victorian era. The text was no longer to be read as a reflective object, as it had been before the Romantic era, and indeed, it was best read as a thing that did not require context or history to be understood. This movement held that the text was a thing that yielded insights when a sufficiently well-equipped reader came along and analyzed it, regardless of other factors.

New Critics viewed the poem as a well-wrought urn, impersonal and ahistorical, and the very existence of the poem implied that in some platonic fashion there also existed—waiting to be discovered by the diligent critic who studied the construction of the poem for efficiency and balance—some ideal reading of that object.105

New Criticism fundamentally reoriented academic thinking about how to approach works of art, not only literature, in its calling for attention to the form of the text apart from other factors like context or authorial intention. Moreover, its main method of investigation, the close reading, arguably remains the dominant tool taught to students and practiced by scholars in the humanities today.106 Thus its lasting impact is hard to overstate even though the movement peaked in the 1940s and 1950s.107

Yet in the 1960s, New Criticism witnessed a steady decline in prestige, despite certain habits of reading it institutionalized.108 Part of what caused New Criticism’s decline had to do with political and cultural context creeping in to create tension around the notion that a text could be read hermetically, and apolitically, let alone that the text must be read that way.109 As New Criticism waned, the text was no longer solely at the center of the inquiry. In its place arose various theories of reading with a focus on readers, how readers read works, and more generally, what “response” to a work of art involves.110 Reader response theories shifted focus from the text to its impact on readers.111 At a minimum, readers were to be considered equally as important as the texts themselves.112 In some cases, the readers trumped the text. Some of these theories, initially, centered on the author, and what the author expected the reader to do, or on the text, and the constraints it placed, or attempted to place, on the readers. Known loosely as “reader response theory,” this scholarly movement “share[d] the agenda of moving from theories of text to the study of reading.”113 Its proponents, such as I.A. Richards, D.W. Harding and Louise Rosenblatt, began advocating for greater emphasis on the reader as early as the 1920s and 1930s.114 Yet it would not become a mainstream position until nearly half a century later.115

Many strands of reader response theory exist, including formalism, structuralism, post-structuralism, phenomenology, and psychoanalysis.116 Later iterations focused more on pedagogy, and on identity aspects of reading, such as gender, race, and sexual orientation.117 Diverse as these sub-movements may have been, they all reject the idea that meaning is created purely by the text.118 Reader response theory came under attack for what appeared to some to be its advocacy of interpretive anarchy: a text could be made to mean anything its reader said it did. This is a common caricature of the anti-reader response position, but it captures the concerns that vesting interpretive authority in the reader can be a slippery slope.119 The range of theories under reader response theory is actually great, and the various theories are far more nuanced. For the purposes of copyright law, there is one theorist whose work is most practical and most potentially applicable.

B.     Rosenblatt’s Reader and the Transactional Theory of Reading

In light of the contested nature of interpretive authority that exists within the diverse reader response movement, Rosenblatt strikes a pragmatic balance. Her view of reader response was that readers were active, but textually delimited, agents. She wrote:

[T]he text is a necessary condition, but it is not a sufficient condition, for the re-creation of a particular work. The text is merely an object of paper and ink until some reader responds to the marks on the page as verbal elements. That is why those [scholars] who seek in the texts alone the elements that differentiate between the aesthetic and the nonaesthetic arrive at only partial or arbitrary answers. They assume the very thing that should be highlighted—the character of the reader’s relationship to the text during these various kinds of reading events.120

Rosenblatt redirects attention to the reader as a locus for study: to Rosenblatt, the reader is as important as the text in understanding how the text comes to produce meaning or exist in the world beyond its author. Her attention to the reader, in turn, affords her insight into how the ways we read impose different demands on us, and invite our focus to go in different directions. Further, Rosenblatt offers a way of understanding the reader’s experience functionally: how she reads will be conditioned by the purpose to which she puts the text. Hence she describes two kinds of reading that readers do in connection with texts: efferent reading (in which they draw out information) and aesthetic reading (in which they experience emotions or seek some sort of affective, rather than informational, response). She describes an efferent transaction in this way: “As the reader responds to the printed words or symbols, his attention is directed outward . . . toward concepts to be retained, ideas to be tested, actions to be performed after the reading.”121 The difference in the means of reading involves the reader’s focus: in efferent reading, the focus is external, while in aesthetic reading, the focus is primarily internal.122 Efferent reading trains the reader to focus “on what will remain as the residue after the reading—the information to be acquired, the logical solution to a problem, the actions to be carried out.”123 One mode is instrumental, seeking answers about the work. The other mode is experiential: an awareness of one’s own evolving impressions of the work, as those evolve, is the goal. The two modes of reading lend themselves to many different contexts, and they map onto copyright’s infringement analysis, as explored below in Part V. Rosenblatt’s theory helps describe what readers actually do when they engage with works of art. The descriptive force of her theory opens a window into copyright law that, ultimately, allows a normative theory of copyright’s ordinary observer to emerge, as detailed later in Part VI.

V.      An Application of Rosenblatt’s Theory of Reading

Rosenblatt’s theory of reading is rooted in the empirical study Rosenblatt did in the four decades of teaching literature, between 1938, when she published her first foray into reader response theory,124 and 1978, when she published her formulation of the transactional theory, complete with efferent and aesthetic modes of engagement.125 Because it is a theory of reading that is phenomenologically oriented, an effective way to understand how it works is to try to approximate the experience of reading actively yourself, in both modes. You have done that. This Part discusses a range of possible experiences—possibly including yours and mine—reading our three shared texts to demonstrate the differences between efferent and aesthetic reading, and to highlight how a range of possible readings does not mean interpretive anarchy.

An efferent reading derives its value from some measure of correctness and, in some cases, special knowledge will assist in determining how to read a piece correctly. For instance, I can describe a romantic comedy as a violent psychological thriller, but unless it is a rare mashup of the two very different genres, I am wrong in my description. Though an efferent reading can be wrong, there is not likely to be a set of correct answers in an aesthetic reading. Put another way, when one shifts into measuring correctness in one’s response, one shifts gears to the efferent stance. An aesthetic reading need not be clever, or even all that sophisticated, to constitute a transaction. It need merely to chart an interpretive course that responds to cues in the text, rather than cues imagined by the reader. The reader’s active engagement, rather than any flourishes of critical insight or specialized knowledge, is what makes the text into a creative work activated into its full potentiality by the reader. It is a deeply subjective process, by design. The skills and experiences we use differ as we read from these separate stances, as you have experienced for yourself in participating in the thought experiment above. These differences will turn out to matter, for copyright litigation, as discussed in Part V, below. First, though, this Part fleshes out what Rosenblatt’s transactional reading looks like, whether it is efferent or aesthetic, by discussing a range of possible responses to the Bruni song, the Rich poem, and the Lawrence painting you experienced above.

A.     A Range of Possible Readings

This Article has cast the thought experiment as a two-step reading exercise in which you read the works from two different stances, the aesthetic and the efferent. In point of fact, however, when you read the introductions, superficial substitutes for the texts though they were, you were already reading a version of the works. You were reading to grasp what the text would be, or to complete the steps in the assignment assigned to you, rather than seeking to experience the works of art for yourself. You organized the information as information, and you waited to be told what to do with that information. You did not experience the works of art yourself in that step. You read the summaries efferently.

By contrast, when instead you experienced the texts firsthand, by listening to, reading, and viewing them, respectively, you had some sort of experience in which you allowed yourself to perceive, and then respond to cues in the works. As you let each work sink in, you found various cues—a sound, a shape, a choice of word or an arrangement of words, a line break—that you used as you actively continued to chart a perceptual course, correcting for meanings you might have misread or not seen, continually adding new layers as you learned more about each work. Your experience relied on your memory of each prior part of the experience. The line or measure before it; the thought you had as you reviewed one part of the painting, and then another. It was idiosyncratic, and shaped by your prior experience with that medium, perhaps that genre, and even that artist. Perhaps you experienced frustration at listening to a song in another language, or perhaps you enjoyed the French (whether you understood it or not; this disparity among what readers bring to texts they read is among the points the exercise highlights). You may have reread the poem, or listened to the song again, or covered the image again from frame to frame, poring over the work and seeing what else might reveal itself to you. You may have gone back to my summaries, to test your informational and experiential versions of the work (or, on the basis of your experience of the work, to test the accuracy of my informational account now that you were in a position to create your own such summary).

This readerly work all probably happened fairly quickly, but it still happened in connection with the aesthetic stance. Then you shifted back into the efferent stance when you completed step three of the exercise, and you returned to the works to collect what you needed for it. Doing so imposed an order on the work, and it also flattened the uniqueness of your individual experience with the work. Anyone could have read the works efferently, as you did, and substituted their reading for yours. Put another way, their summary of the work could have substituted for your reading.126 Efferent readers are fungible.

This reading was active labor on your part, and this experience of mental creativity and labor, for Rosenblatt, is what fully “evoked” those works of art. Rather like the proverbial tree in the forest, whose falling might be said not to make a sound if no human is there to experience it, the work the author creates requires the reader to experience it to bring it into meaning. The constructive experience constitutes that work, and it depends equally on a text, or work of some kind, and on the reader’s agency. Rather than having the text, or the author, dictate a fixed meaning, the reader uncovers and creates it.

To be clear, this work may have looked different from reader to reader; aesthetic reading is highly subjective. My own experience listening to the Bruni song is largely to be ignorant of what is happening musically, but to experience naïve pleasure at the language, whose lyrics I (perhaps idiosyncratically) find appealing. Someone with greater experience in music, as a student or performer, would doubtless have a richer experience encountering cues and responding to them (whether or not they like the song or dislike it). My universe of cues is small with this work, and, to be fair, certain kinds of works offer fewer cues than others.127 Intuitively, we may experience such works as “simpler,” or “easier,” works, or perhaps we find them glib, or unsatisfying. When I view the Lawrence painting, I find I respond to the colors, and I also find myself musing about what and who is out of sight. (Is anyone downstairs below the studio?). If I were equipped with more information about post-Harlem-Renaissance art, no doubt I would “see” more in the painting, but it does not invalidate my reading of the painting that I am only able to activate some of its cues. If you bring more background to the work than I, you likely had a richer experience. You may have made the work “mean” even more fully.

Similarly, the poem will elicit a range of experiences on the aesthetic side. The voice addresses itself to a listener: the second person “you.” The speaker positions herself as a dispenser of a kind of unconventional folk wisdom whose rhetoric includes predictions (“You’ll drop /‌ your book to hold your /‌ water bottle steady”) as well as the advisory language offered at the start of airplane flights (“put on /‌ the child’s mask first.”). While the language itself, in the context of a flight, seems clear enough (parents, be sure you can in fact care for your kids; if you pass out, you will fail to do so), its use reminds the reader of the stark possibility of incapacitation and the necessity of prioritizing oneself in the act of self-preservation. The poem repeats references to losing control (three lines end with indirect hints of destruction: “You’ll drop,” “fall” “break”), and in so doing, it plays out the parallel between a plane crash and an emotional trauma. Like the plane, which is built “to shudder,” the recipient learns he or she is “designed to tremble too.” A layer of literary allusion is built into the middle of the poem, when its lines “Your /‌ mind, mind has mountains, cliffs of fall /‌ may who ne’er hung there let him /‌ watch the movie.” The referent is Gerard Manley Hopkins’s poem, “No Worst, There is None, Pitched Past Pitch of Grief,” and its lines are: “O the mind, mind has mountains; cliffs of fall /‌ Frightful, sheer, no-man-fathomed. Hold them cheap /‌ May who ne’er hung there.” Rich’s poem repeats the Hopkins line (though she personalizes the mind, converting it from “the” to “your”), but she adds humor to deflate the mood. Instead of letting the line end with the idea in the original (that the sublime, scary, enormous landscape within the mind can not be ignored—“held cheap”— if one has experienced it), she literalizes the airplane and suggests an alternative. Someone afraid of literal turbulence has a palliative option, and can escape rather than experiencing the emotion: “watch the movie.” The poem’s concluding line, “Breathe normally “ [sic] ends without punctuation, with a deliberate space, as if mid-sentence, suggesting interruption, or abrupt, unexpected conclusion. While it is a direct quotation from at least some typical safety scripts airlines use, its abruptness here inverts its effect: on a plane, the idea in that context is to avoid panicking, and it usually comes accompanied with language about how oxygen is flowing at that point, even if it is not immediately obvious.128 Breathing normally is an instruction to avoid panic.

I focus a great deal on the experience of the poem when it is read aloud, noticing its sound and shape. The poem looks somewhat cliff-like, and its choppy phrasing gives effect to the theme of instability. I note the way the lines seem mostly to end with a dramatic or violent word, or short syntax (until “normally,” which seems paradoxically abnormal, syntactically, by that point). I experience pleasure at the sounds and repetitions Rich uses (“shudder, shoulder,” “mind, mind”), and her alliterative phrases “lungs labor, heights hurl vistas,” which cause the reader reading aloud to breathe differently. Even the title creates a repetition that is almost reassuring when read to flow directly into the poem: “Turbulence. There’ll be turbulence.” Finally, I experience a chill of sorts when the last line breaks off abruptly, as though the line hangs there the way it does because the speaker finally believes it to be empty advice, a futile thing to try to do when in the midst of trauma. No doubt others see more, or fewer, or different, cues. All our idiosyncratic readings are valid so long as they are interpreting cues in the work, not making them up from private associations. This is not a poem, for instance, that dramatizes shipwreck as a conceit for emotional volatility or pain; it deliberately features planes, and sky imagery, not ships, and water imagery.129 Nor is it a poem that that seems to explore the ups and downs of passion, say, or the unlikelihood of love, as it could through an airplane metaphor differently deployed.130 Those readings would be affirmatively incorrect. Efferent readings may lead to a single correct answer, or there may be in rarer cases multiple correct answers, as when experts use rigorous and well-accepted methods but arrive at divergent conclusions, and thus may be thought to have good reasons for answering efferent questions in ways that differ. Still, among the possible efferent readings, there will no doubt be many incorrect answers; by contrast, aesthetic readings, so long as tethered to textual evidence, will always leave room for subjectively differing experiences without calling any one of those necessarily wrong.

B.     A Straightforward Reading to Make Basic Sense of the Text

Lest it seem that only complicated readings count as active, or transactional, note that simply reading through the poem requires the engaged, creative work of the reader. For instance, when the reader encounters the first phrase, “There’ll be turbulence” she right away must conjure a speaker, but at that point the speaker could as easily be a pilot on a plane, or a seat mate making a lay prediction based on their life experience, or an attendant informing passengers what lies up ahead on a flight. The very next word instructs the reader that in the world of the poem, she is the recipient of the speech, thus asking her to believe that she is either to imagine herself on the plane, being spoken to, or to accept the idea of turbulence as metaphorical. The voice informs her that she will drop, which sounds intransitive (that is, does not take a direct object), as though she herself will be the thing that falls. The next two words, “your book,” make clear that the verb, “drop” is transitive (that is, does take a direct object), and its object is “book.” She adjusts her reading accordingly, and perhaps she notes that the author placed a line break between “drop” and “your book,” expressly causing her to engage in the rethinking of what is going to drop, just as she has done. This is what it looks like on the page so far:


There’ll be turbulence. You’ll drop
your book to hold your
water bottle steady.131

The poem next shifts gears from the description of the plane’s turbulence to something else the reader must work to understand on the most basic level; it moves from concrete things, a book and bottle, to an intangible thing of a different kind: mind. Soon that intangible thing is wrapped up in metaphor, too. The next lines read: “Your /‌ mind, mind has mountains, cliffs of fall /‌ may who ne’er hung there let him /‌ watch the movie.”132 The reader cannot keep reading without doing work to assimilate the shift, and even if she is not sure what to make of it, she now knows definitively that she has shifted out of literal gear because a mind does not literally have mountains and cliffs. If she were reading strictly in the efferent stance seeking information, she would put the poem down, dismayed by its scientific carelessness or exaggeration. Instead, she continues, aware on some level that her reading the poem is an experience of encounter with art, and exists on the symbolic, not literal level. The lines borrowed from Hopkins are archaic and also unusual in their syntax, in a way that may be recognizable to readers of Hopkins, who used sound in his poetry in a unique fashion.133 The lines are likely to need rereading by any reader merely trying to follow along who is doing something other than simply advancing to the next line without understanding the meaning of the sentence. Through reading the line again, a reader might sort out that two of the phrases point in different directions: “cliffs of fall” belong to the mind’s mountains, and “may who ne’er hung there,” is an independent statement meaning “let the person who has not hung on those cliffs . . . .” The cliffs and mind thus cohere as a landscape of the mind with its potentially large emotional scope; the line about who has hung there imagines people in pain experiencing themselves stuck (or hung) on that landscape, which requires an additional transposition of the metaphor, from an imagined landscape in the brain to a place within the brain within which another metaphor, the body within the brain, can hang. The old-fashioned nature of the language, including the start of the line with “may,” and the use of the contraction for “ne’er” signal the language’s difference from the commonplace phrases that belong in the same linguistic register as the instructions about turbulence.

The mind and cliffs—and for that matter, the heights that hurl vistas—are on one side of a linguistic divide; the book, bottle, plane, movie, and mask are on the other. Even readers who do not stop—and likely most will not—to sort through this difference, will experience the poem’s moving between different registers from concrete to intangible and back to concrete, with the phrase, “watch the movie.” The poem may eventually cause the reader to move back through it with fuller understanding of the way the concrete references to the plane in the poem work to effectuate the symbolic, metaphorical parts, that is, the parts that finally give the greatest amount of meaning to the poem. Without them, the poem would be a short, not very effective instruction manual about how to manage airplane turbulence.

For readers who make it through the poem with the barest schema of meaning pieced together, a great number of those cues must be experienced simply to make the poem mean. That is what Rosenblatt means when she calls reading a transaction in which a text is a necessary, but not sufficient condition.134 The reader comes along and makes the text mean something, but she cannot do so with total freedom, or disregard for the textual cues. On the contrary, she is the one who perceives the cues, fills gaps around them, and breathes life into the poem, making it mean as much as it possibly can, for her. It may mean more, or less, to other readers, but in order to make it mean at all, to actuate it as a work of art, the reader must actively engage to produce that meaning. Recall that Rosenblatt does not subscribe to an approach that would hold that the text can mean anything, that is, to a view of semiotic indeterminacy.135 On the contrary, there are cues there that can mean, and perhaps must mean. Not all cues will mean, for all readers, but those cues exist as constraints shaping the reader’s experience in the transaction.

VI.    A Transactional Approach to Copyright’s Infringement Analysis

This Part applies Rosenblatt’s theory to suggest four changes to copyright law: (1) more work should be done by judges and experts as a matter of law, thus limiting or narrowing the role of the jury in determining infringement; (2) expert evidence ought to be permitted to play a greater role in copyright litigation; (3) the jury should be instructed to do a different, more informed kind of work when it evaluates works of art for infringement; and (4) courts should consider using special verdicts to restrict jury verdicts to appropriate tasks and to render jury decision-making more transparent. In light of a growing consensus that copyright’s infringement analysis is full of inconsistency and illogic, it may be time to reexamine some of its first principles, beginning here with the ordinary observer standard.

The first thing to note is that copyright’s infringement analysis is widely acknowledged to be flawed in concept and operation. The heart of it— substantial similarity analysis—is incoherent. Multiple explanations exist for why substantial similarity is so complex, inconsistent, and confusing. At a minimum, copyright infringement analysis has been recognized as an area of law with a deeply normative aspect.136 Scholars bemoan the range of confusing or regrettable ways infringement is decided, often lamenting the illogical exclusion of expert evidence on questions where such testimony could be helpful.137 Further, there is a growing consensus that jurors cannot do what courts ask of them.138

Scholars have argued that the answer may be to do away with substantial similarity altogether,139 perhaps replacing it with a pure reproduction right.140 Assuming such radical change is unlikely,141 and that substantial similarity is likely to remain good doctrine for the foreseeable future, some changes within the doctrine could still help produce more logical and fairer outcomes. Indeed, numerous scholars have proposed various doctrinal fixes, though most of those have focused on the elements of substantial similarity other than the ordinary observer.142 A promising, but nascent, literature discusses what jurors are actually called upon to do, and whether they seem to be able to do it.143 Still less scholarship has theorized the audience in terms of a broader affirmative concept.144 This Article proposes that reform efforts target the ordinary observer standard, which could improve the method by which courts determine substantial similarity because of the important functions associated with it: allocation of decision-making to the jury rather than the judge, exclusion of expert evidence, and a black-box mentality to the jury’s impressions of the work. A transactional theory of the reader could shape reform efforts by providing a better descriptive model of how judges and juries read works of art, and how that reading is shaped by the purpose of the reading: are they filtering and reading efferently? Or are they intended to read subjectively for a holistic or “intrinsic” response, that is, reading aesthetically? The distinction could divide the allocation of decision-making helpfully, as shown in the next two subparts.

A.     Efferent Reading in Copyright’s Infringement Analysis

In copyright infringement cases, courts have clear moments when they read from an efferent mode, such as when they are deciding whether a work draws on public domain sources, or trying to assess whether two works contain similarities sufficient to warrant the inference of copying. Rosenblatt’s theory illustrates that the kind of reading judges are called upon to do in copyright infringement litigation as a matter of law is efferent, that is, informational, or instrumental reading. Deciding whether a work contains unprotectible scènes à faire—which is a copyright term of art for stock elements necessary to a genre or idea—for instance, or filtering protected from unprotected elements, involves reading a work to extract an answer (to “bear away” an answer, again per the Latin roots of “efferent”).145 Thus, as a default, the questions involving informational, filtering, or doctrinal conclusions should be treated as efferent and, in my view, decided by a judge as a matter of law. This would align with the way dissection-of-works is currently treated in most circuits.146 It would also align with the well-settled rules dividing work between judges and juries: judges resolve questions of law and jurors consider questions of fact. Informational questions that have correct answers (or questions where there will be some incorrect answers, even if there are competing correct answers) are more like questions of law; subjective impressions (and aesthetic reading) are more idiosyncratic, context-dependent, and variable, and they are more like questions of fact. They are particular to each given reader, every time.

If the foregoing statements are correct, then it may make sense to allow expert testimony to play a bigger role, at the trial court’s discretion. Currently, expert testimony is disallowed during the subjective or intuitive second step of the infringement analysis.147 It is disfavored even in the first step, unless works merit special treatment because of their complexity or impenetrability.148 Adopting Rosenblatt would justify expanding the use of experts, because the informational investigation of texts can be helpfully guided when those possessing deeper knowledge of a text, or a field, assist the decision-maker.149 Anyone can read efferently, as you will recall. However, it is important to provide tools to assisting the reader in finding the correct answer, or the maximum possible accurate information. This would not foreclose the possibility of allowing expert testimony to play a role in assisting jurors in their reading, however. Just as I provided you with background to the works you read in the thought experiment above, experts can provide context and additional information that conditions courts to read efferently with greater efficiency and accuracy, and condition aesthetic readers to activate more cues in the works they are evoking through their reading.150 There are, of course, costs to expert testimony, but here their risks may be outweighed by the confusion and inconsistency riddling infringement analysis.151 Importantly, a respected federal judge, the Honorable Denise M. Cote, published a brief article advocating for greater reliance on expert testimony in copyright litigation, suggesting that it is feasible, and that other judges might welcome it just as fervently.152

Filtering of the works should occur at this first (efferent) stage too, since abstraction and filtration of protected and unprotected elements is clearly efferent, rather than aesthetic, in nature. Only after filtering has occurred should jurors be exposed to the works, and even then, given the idiosyncratic aspects of aesthetic reading, the scope of jurors’ decision-making should be cabined. Narrow questions concerning the aesthetic appeal of the work can be answered by jurors as questions of fact, in order to assist a judge in a larger infringement analysis. This has the virtue of avoiding what one case—where the parties were litigating the originality of and relevant market for choral arrangements of spiritual music—has called “the practical evil of having an unaided uninformed finder of fact deciding the crucial issue in a case.”153

B.     Aesthetic Reading in Copyright’s Infringement Analysis

The prior Part argued that the court ought to be allowed to do efferent reading as a matter of law, perhaps with recourse to expert testimony. Nonetheless, there is an important role for the jury, or factfinder, in copyright litigation: reading works aesthetically. Many courts presently divide infringement into dissective (or objective) and intuitive (subjective) phases.154 The role for the jury, however, is not well understood in terms of theories of reading. At times the jury is given lengthy and unclear instructions as to what it should do, and at other times, it is simply incapable of doing the work that is asked of it.155 There is a notion that juries ought to be able to look at works and experience a gut reaction.156 In point of fact, Rosenblatt’s theory of aesthetic reading shows that even intuitive reading unfolds over time, not in the gut, but in the back-and-forth transaction between the reader and the work as the reader “evokes” the work, responding carefully to its cues.

With all of this in mind, courts could limit the jury to reading aesthetically only—or to reading efferently, but only with respect to questions of fact that they are clearly well-suited to answer as laypeople. Distinguishing between efferent and aesthetic could help courts more accurately decide who can do what, and may justify decisions about the scope of authority delegated to the jury as well. Of course, the question that lurks behind what questions go to the jury is how those questions are sent there, and what kinds of instructions are most effective.

C.     Instructing Juries More Effectively

Jury instructions are considered to be among the most important aspects of a trial.157 Instructions perform a key translating role, explaining the law to nonlawyers, and shifting the jury’s attention just as jurors are “turning from listening to deciding.”158 Juries in copyright litigation cases at present are instructed on a range of complex issues in inconsistent and unclear ways.159 The expert’s role in assisting the court and the factfinder remains contested and ad hoc. Part of the problem is the reigning view that copyright cases regarding artwork do not require special expertise; another part of the problem is that juries are often asked to do more than they can realistically do, especially when denied the assistance of expert testimony. Jury instructions in copyright cases often fail in terms of being comprehensible to jurors, being feasible for jurors, and being fully accurate with respect to copyright law. A transactional theory as applied would hold that jury instructions should do more to signal to juries that they will be doing complex work, that they may require assistance, and that they, as individual jurors, may experience works differently from their fellow jurors. Applying this theory could help improve comprehensibility, feasibility, and accuracy.

It is helpful to consider existing model instructions for jurors as a means of determining the feasibility of implementing Rosenblatt’s theory. Given its prominence in copyright law, and the fact that it is undergoing review of its instructions currently, the Ninth Circuit Court of Appeals provides a good point of investigation. Focusing for present purposes on the first step of substantial similarity analysis, the model instructions read:

Instruction [insert cross reference to the pertinent instruction, e.g., Instruction 17.4 [[the Elements of Infringement]] states that the plaintiff has the burden of proving that the defendant copied original elements from the plaintiff’s copyrighted work. The plaintiff may show the defendant copied from the work by showing by a preponderance of the evidence that the defendant had access to the plaintiff’s copyrighted work and that there are substantial similarities between the defendant’s work and original elements of the plaintiff’s work.160

Applying Rosenblatt might lead to the addition of a few sentences underscoring (without mentioning that highfalutin word, “efferent,” directly) the efferent nature of the analysis that takes place in this stage. With an emphasis on efferent reading, one can ask the jury to do some work: make a list, consider which things are creating the perception of similarity, and so on. The proposed language is:

To determine whether the defendant’s work is similar to the plaintiff’s work, take note of what is similar about the works. Though the list of similarities you make is not conclusive evidence, it helps point the court to the specific things you find similar. Instruction # [insert cross-reference to the pertinent instruction, e.g., an instruction explaining the idea expression dichotomy] states that only the particular [way of expressing] or [expression of] an idea can be copyrighted and protected. Making a list of similarities helps the court identify which parts of the work receive copyright protection.

Any such instruction should be followed by an instruction asking jurors to list dissimilarities as well, so as to counteract a cognitive bias in favor of finding similarity. A jury instruction could also address the way expert evidence might affect juror decision-making here, since experts might testify as to similar and dissimilar features of given works. Rosenblatt’s theory shows that efferent reading requires special knowledge of the sort that typically experts have and judges and juries lack. Jury instructions can attempt to cure for this lack of knowledge, but they will inevitably only go so far.161

Juries, even when they encounter works in the intrinsic mode, should be instructed not to simply consult with their gut to discern whether the “total concept and feel” of two works is similar. Instead, they should be made aware that the process of their engagement with the works is aesthetic in the sense Rosenblatt describes. Their reading will unfold over time—whether minutes, hours, days, or weeks—subject to constraints already existing with the works. Further, it may differ from juror to juror. Indeed, Rosenblatt would expect it to differ in this way. Thus jurors should be prepared to explain, in their own words, their own experiences (recall that there are no single “correct” readings in the aesthetic mode, simply readings that do respond to cues and ones that ignore them and fail to engage on the most basic level with those works). Of course, this must be done in such a way as to state the law correctly, truthfully, comprehensively, and comprehensibly, no small task.162

Note that improving instructions will not solve all jury issues in copyright law. Scholars have pointed out that among the various biases that distort juror decision-making, there is an order bias that makes the second step—intrinsic, in the Ninth Circuit, or improper appropriation, in the Second—more likely to weigh in favor of the plaintiff, if the first step produced a finding of copying (that is, found for the plaintiff).163 It would require an overhaul of copyright case law, or at a minimum, some doctrinal innovation, to reverse the order of the steps of analysis. Yet once a jury is involved, it may make good sense to do this to overcome or limit the effects of jury biases. Presenting a jury first with the work requiring only a subjective response would allow Rosenblatt’s aesthetic reading to take place free from the informational bias that arises once jurors know—from the evidence typically presented in step one—that the defendant had copied and the works are deemed sufficiently similar to continue at trial. If no doctrinal change arises to cure this cognitive bias, it may be worth thinking about how to limit the effect of this bias, perhaps through disclosure to jurors.

Under copyright law’s current doctrinal structure, however, the intrinsic (subjective) analysis would follow the extrinsic (dissective) analysis addressed in the proposed jury instruction above. Applying Rosenblatt in the second step would require additional treatment. The Committee on Model Jury Instruction for the Ninth Circuit Court of Appeals once offered its own tailored, complex jury instruction on the extrinsic and intrinsic test for similarity, but withdrew it and instructed that courts offer case-by-case particularized instructions.164 The Committee reasoned that “the general statement of the test embodied in the former instruction was not helpful in light of the diverse facts that might arise at trial pertinent to a substantial similarity assessment,” and accordingly, it concluded “that the court and counsel would be best served by specifically crafting instructions in this area based upon the particular work(s) at issue, the copyright in question, and the evidence developed at trial.”165 Following a transactional theory, a jury instruction might instruct jurors to focus on their own subjective experiences. Without using terms of art that  jurors are unlikely to understand (such as “intrinsic” or “aesthetic”), jury instructions could explain to jurors that this part of their reading is intended to be holistic and intuitive, even as it unfolds through time. This instruction would differ from the prior proposed instruction in steering clear of asking juries to list similarities or produce analysis of the works. Instead, the charge to jurors would be to experience the work and then offer analysis of their reactions to the work. Tracking Rosenblatt, they would be shifting focus from the text or author to the reader’s experience of that text.

Jury instructions remain a powerful tool that could help take insights gleaned from a transactional theory of reading and operationalize them to improve copyright litigation. Much more work can and should be done to improve jury instructions and rationalize their use in copyright litigation. This Part has simply taken up the very first steps in imagining what a transactional theory of the reader could look like, as applied.

D.     Incorporating Special Verdicts

Among the difficulties of copyright’s current ordinary observer problem is that juries are handed one of the hardest parts of the infringement analysis, and then effectively given unfettered discretion over whether, intrinsically, they experience two works as similar.166 This results in the scope of copyright protection often being poorly calibrated since jurors are ill-equipped to perform the filtering copyright doctrine requires, and better suited to speak to holistic similarity that tends to stack the deck against defendants.167 This black-box mentality to the jury’s impressions of the work means that when juries find substantial similarity, a judge cannot at that point ask whether their finding sufficiently distinguished between protectable and unprotectable elements. It is too late to address the problem of overprotection. Nor is it clear that a jury can accurately distinguish among those elements even when instructed to do so. This Part urges courts to consider the use of special verdicts that would gather information from jurors about their subjective experiences of the works at issue. There is nothing wrong or unusual about jurors having subjective experiences of works of art. Indeed, that they do so comports with Rosenblatt’s theory of aesthetic reading. But when that experience is mapped against the role that courts typically ask juries to play, it is clear that it goes too far because juries are routinely asked to evaluate whether the kind and quality of copying constitutes infringement. The flip side of that proposition is that juries are determining the scope of copyright protection in the complaining work. They are doing so on the basis of what Rosenblatt’s theory shows us is an aesthetic experience, reading works seeking improper appropriation as a question of fact. The danger is that the subjective, intuitive approach to these works ends up deciding what is, or should be a question of law: how much protection does a particular work receive? Requiring jurors to “show their work” would help minimize the risks inherent in subjective experiences of work being responsible for calibrating the scope of copyright protection.

Usually juries return a general verdict, defined as “a finding for or against the plaintiff that does not state the grounds for the jury’s decision.”168 By contrast, a special verdict requests that juries answer specific questions submitted to them by the court.169 Under the Rule 49 of the Federal Rules of Civil Procedure, a federal judge may decide to direct a jury to return a special verdict or a general verdict with interrogatories.170 Special verdicts arose as a device to curb the perceived excesses of the general jury verdict, which was perceived to:

[C]onfer[] on the jury a vast power to commit error and do mischief by loading it with technical burdens far beyond its ability to perform, by confusing it in aggregating instead of segregating the issues, and by shrouding in secrecy and mystery the actual results of its deliberations.171

Prior to Rule 49’s adoption, special verdicts were available but not often used. One explanation for why the general verdict remained the dominant process had to do with the special verdict’s transparency: it was actually one of the benefits of the general verdict that a jury could do its work under the relative protection of black-box deliberation. One scholar wrote:

[T]he great technical merit of the general verdict . . . [is that it] covers up all the shortcomings which frail human nature is unable to eliminate from the trial of a case. In the abysmal abstraction of the general verdict concrete details are swallowed up . . . [it] is valued for what it does, not for what it is. It serves as the great procedural opiate, which draws the curtain upon human errors and soothes us with the assurance that we have attained the unattainable.172

This tension between producing a decision and the perhaps chaotic decisional process “behind the curtain” calls forth larger jurisprudential questions about form and content in law. Should it matter how the jury reaches a decision if it reaches a “good” decision? The answer to that may depend on whether one adopts a realist or formalist approach to law. And it will inevitably and swiftly lead to debates over how to define a good or bad outcome. The process will matter a great deal if the decisional outcomes are thought to be consistently skewed and the decisions are, by consensus, not “good” or unfair.  In copyright law, as the previous parts of this Article have argued, jury decisions often mistake the scope of copyright protection, skew outcomes, and ultimately reflect an improper allocation of decision making as between judge and jury.

Copyright law, therefore, might be particularly well-suited to use of the special verdict, or the general verdict with interrogatories. By asking juries to answer questions about their experiences of a work of art, the court could determine whether what juries found similar about two works was in fact protectable or not. The court could also take into account differences in jurors’ aesthetic reading of the works, noting their subjective reactions but setting those in the larger framework of the efferent analysis the court must do to manage the proper scope of copyright protection. This Article leaves for further work the task of specifying what such verdicts would ask jurors to answer, but the existing body of literature provides guidance on how to decide when not to use a general verdict and how to avoid some of the drawbacks associated with the use of these narrowing devices as alternatives to the general jury verdict.173

In conclusion, Rosenblatt’s theory holds that a transaction occurs between the reader and the work. The dialogic nature of her theory of reading helps us see that the jury in copyright law is typically being asked to do interpretive reading that copyright case law does not acknowledge, and thus does not sufficiently theorize as reading. Instead, copyright appears to assume that instantaneous, gut-level intuition, and the ordinary observer perspective, provides all that jurors need. Jury instructions and the general verdict form are designed to steer jury intuitions in the right direction, but not necessarily more than that. Rosenblatt’s theory would call for us to acknowledge that that leaves too much discretion for jurors and provides insufficient guidance. Particular readers have particular, idiosyncratic responses when they engage with a work affectively, and it takes some time and effort to process those responses. Applying Rosenblatt’s insights could translate into better jury instructions,  that aim for greater comprehensibility (understanding the task) and greater feasibility (applying the law). Rosenblatt’s insights could also translate into more use of special verdicts, with the aim of increasing transparency and accuracy. The transactional theory could further be used to argue that the law needs to be not just improved upon but fundamentally overhauled such that the two steps of copyright’s infringement analysis are reversed. Doing so would allow jury deliberation to focus not on improper appropriation but on the questions of fact surrounding copying and access—that is, not whether copying was improper but whether it happened at all. That larger reform is beyond the scope of this Article: its aims are rather more modest and pragmatic.

This Part has argued that operationalizing the division of functions in a reader’s experience, between efferent and affective reading modes, could help improve copyright law’s infringement analysis by better framing the roles allocated for judge, expert, and jury. Further, it has begun the work of tackling jury instructions that might more effectively guide juries toward outcomes based on what juries can and do actually do when encountering and evaluating works of art. It could—with some doctrinal tweaking—correct for some of the juror bias introduced through the current order of issue resolution by juries. It would also have the benefit of aligning legal practice and procedure better with human behavior with respect to works of art. Finally, it has argued that courts would do well to consider the use of special verdicts or general verdicts with interrogatories, in order to guide, and also constrain, jury decision-making with respect to how juries read works in copyright infringement cases.

VII.     Conclusion

The ordinary observer standard is not a model of conceptual clarity. Courts and scholars alike do not seem to have a reliable theory for what it is, why it exists, and how to operate it. This is despite the real complexity of reading works of art of all kinds, and the difficult and divergent tasks we set for judges and juries in copyright infringement cases. Yet it skews outcomes, contributes to the miscalibration of scope in copyright, and literally demands the impossible of juries. For those reasons, scholars and courts should care about improving their understanding of the ordinary observer, and rethinking its role in copyright litigation.

If the offhand remark at the start of this Article—about reading being like bathing—presents a view of reading as waterlogged, relaxed, and comfortable, Rosenblatt’s view could not be more different. She concludes her second major book on reader response theory with a call to arms, a view of the reader as so active a participant in creating—or “evoking”—the book that the image she uses is that of a gymnast:

Books are to be call’d for, and supplied, on the assumption that the process of reading is not a half-sleep, but, in highest sense, an exercise, a gymnast’s struggle; that the reader is to do something for himself, must be on the alert, must himself or herself construct indeed the poem, argument, history, metaphysical essay—the text furnishing the hints, the clue, the start or frame-work. Not the book needs so much to be the complete thing, but the reader of the book does.174

Rosenblatt’s great insight was to shift attention away from the text as the dominant source of authority in the reading experience, to the reader. The reader’s experience contained clues not only to her particular experience, but to generalizable truths about how we read, and what reading actually demands and provides, depending on why we are doing it. By studying the readers in her classrooms, Rosenblatt opened the way for a theory of reading that was more particularized, more empirically accurate, more practical, and more balanced than any other models in the academic literature. These qualities make Rosenblatt’s transactional reader an apt fit for copyright law, which lacks a theory of its reader, and greatly needs one.

This Article has presented a view of the ordinary observer in copyright law, to argue that it is poorly understood and in need of more sustained attention from courts and scholars. Greater awareness of copyright’s readerly gymnastics could help create more effective rules around the allocation of authority between judge and jury, the possible admissibility of expert evidence, and the work juries actually have to do in order to render their verdicts. An improved understanding of the ordinary observer could help build a case for improved jury instructions, as well as the use of special verdicts and interrogatories that could provide more guidance to juries as they do the difficult work of engaging with works of art, that is, the work of reading that copyright needs them to do. This better understanding would assist judges in exercising their own authority and knowing when delegating work to juries makes sense, versus when it will mean jurors cannot properly do the task set for them, or cannot do so unassisted. In sum, starting to define copyright’s reading practices, and to conceptualize its readers, could improve one big problem area in copyright litigation, thus rendering it more streamlined, more accurate, and fairer. That said, Rosenblatt’s insights hold the possibility of improving copyright in problem areas well beyond the ordinary observer standard. Greater awareness on the part of judges in how they do and should read could guide them in their evaluations of summary judgment motions, their determinations of questions as being matters of law versus matters of fact, and their assessments of what sorts of evidence is both relevant and necessary. Rosenblatt’s theory helps us see that the ordinary observer standard is one broken part of copyright litigation, and future work studying its application in other troublesome areas of copyright could reveal its power to reshape how courts fairly and efficiently adjudicate copyright disputes, while retaining but rethinking the role the jury can play.


 1.     The song in question is Italian-French singer-songwriter Carla Bruni’s La Dernière Minute.175

The lyrics to the song read as follows:
Quand j’aurai tout compris, tout vecu d’ici-bas,
Quand je serai si vieille, que je ne voudrai plus de moi,
Quand la peau de ma vie sera creusée de routes,
Et de traces et de peines, et de rires et de doutes,
Alors je demanderai juste encore une minute . . .
Quand il n’y aura plus rien qui chavire et qui blesse,
Et quand même les chagrins auront l’air d’une caresse,
Quand je verrai ma mort juste au pied de mon lit,
Que je la verrai sourire de ma si petite vie,
Je lui dirai “écoute ! Laisse-moi juste une minute . . .”
Juste encore une minute, juste encore une minute,
Pour me faire une beauté ou pour une cigarette,
Juste encore une minute, juste encore une minute,
Pour un dernier frisson, ou pour un dernier geste,
Juste encore une minute, juste encore une minute,
Pour ranger les souvenirs avant le grand hiver,
Juste encore une minute . . .sans motif et sans but.
Puisque ma vie n’est rien, alors je la veux toute.
Tout entière, tout à fait et dans toutes ses déroutes,
Puisque ma vie n’est rien, alors j’en redemande,
Je veux qu’on m’en rajoute,
Soixante petites secondes pour ma dernière minute.
Tic tac tic tac tic tac

I’ve provided my own very basic translation of La Dernière Minute (The Last Minute), just below:

When I’ve seen [understood] it all, and lived it all here on earth,
When I’m so old even I wouldn’t want me,
When my skin is creased with deep lines
And with traces and sorrows, and with laughter and doubt,
Then I’ll ask for one more minute. . .
When there’s nothing left that can capsize or wound me,
And even sadnesses will feel like caresses,
When I’m facing my death right there at the foot of my bed,
And I see death smile at me, and my insignificant life,
I’ll tell it, “listen! Leave me just one minute. . .”
Just one more minute, just one more minute
To pretty myself up, or have a cigarette,
Just one more minute, just one more minute
For a last thrill, or a last gesture,
Just one more minute, just one more minute
To tidy up my memories before the great winter,
Just one more minute, without motive or goal.
Because my life is worth nothing, well, I want it all.
In fullest form, completely, and with all its foibles [routs],
Because my life is worth nothing, well, I’m asking for more of it.
I want to have some added on,
Sixty little seconds for my last minute.
Tic tac tic tac tic tac

2.  The poem discussed above is by the late Adrienne Rich.176 I have reproduced it in full here:


There’ll be turbulence. You’ll drop
your book to hold your
water bottle steady.  Your
mind, mind has mountains, cliffs of fall
may who ne’er hung there let him
watch the movie.  The plane’s
supposed to shudder, shoulder on
like this. It’s built to do that. You’re
designed to tremble too. Else break
Higher you climb, trouble in mind
lungs labor, heights hurl vistas
Oxygen hangs ready
overhead. In the event put on
the child’s mask first. Breathe normally

The poem by Gerard Manley Hopkins to which Rich’s poem alludes follows here:

No worst, there is none. Pitched past pitch of grief177

No worst, there is none. Pitched past pitch of grief,
More pangs will, schooled at forepangs, wilder wring.
Comforter, where, where is your comforting?
Mary, mother of us, where is your relief?
My cries heave, herds-long; huddle in a main, a chief
Woe, wórld-sorrow; on an áge-old anvil wince and sing —
Then lull, then leave off. Fury had shrieked ‘No ling-
ering! Let me be fell: force I must be brief.”‘
O the mind, mind has mountains; cliffs of fall
Frightful, sheer, no-man-fathomed. Hold them cheap
May who ne’er hung there. Nor does long our small
Durance deal with that steep or deep. Here! creep,
Wretch, under a comfort serves in a whirlwind: all
Life death does end and each day dies with sleep.

3.    The work of visual art, The Studio (1996), a self-portrait by Jacob Lawrence, is owned by the Seattle Art Museum, and out on loan elsewhere at the moment. A small replica of the image is pasted in below.178

 Appendix pic Said

  1. [1]. Keith Harris, The Blurred Lines Verdict Proves Only One Thing: You Can’t Second-Guess a Jury, Guardian, (last modified July 13, 2016, 3:51 AM).

  2. [2]. Ted Johnson, Judge Rejects New ‘Blurred Lines’ Trial, Variety (July 14, 2015, 7:24 PM), ThYet some docause of the problem of presenting juries with specialized legal questions they are ill equipped to answerecording, ae jury initially awarded $7.3 million in damages, but the court later reduced damages to $5.3 million. Id.

  3. [3]. Wendy Gordon, Opinion, The Jury in the ‘Blurred Lines’ Case Was Misled, Newsweek, (Mar. 18, 2015, 2:30 PM),; Kal Raustiala & Christopher Jon Sprigman, Squelching Creativity: What the “Blurred Lines” Team Copied is Either Not Original or Not Relevant, Slate (Mar. 12, 2015, 12:27 PM),
    politics/jurisprudence/2015/03/_blurred_lines_verdict_is_wrong_williams_and_thicke_did_not_infringe_on.single.html; Tim Wu, Why the “Blurred Lines” Copyright Verdict Should be Thrown Out, New Yorker (Mar. 12, 2015),

  4. [4]. Wu, supra note 3 (noting that Gaye had never complied with the formalities necessary to registering a copyright in the sound recording, and thus owned only the copyright in the composition, or sheet music, not in his performance of it); see also Mike Masnick, Blurred Lines Copyright Lawsuit Gets Funky as Judge Delves into the Blurred Lines of What’s Really Copyrighted, Techdirt (Feb. 2, 2015, 8:02 AM), (“The judge properly noted that only part of the song is actually covered by copyright, and it would be hellishly unfair to use the elements of the song that are not covered by copyright (including Gaye’s voice) to prejudice the jury.”).

  5. [5]. Williams v. Bridgeport Music, Inc., No. LA CV13–06004 JAK (AGRx), 2015 WL 4479500, at *17­–18 (C.D. Cal. July 14, 2015).

  6. [6]. Gordon, supra note 3.

  7. [7]. Wu, supra note 3 (“[A] serious error has been made: the judge overseeing the case should never have let the case go before a jury. The ruling against Thicke was a mistake, and it should, and likely will, be reversed on appeal.”).

  8. [8]. Jeremy D. Larson, 8 Artists That Could Sue ‘Uptown Funk’ on the Same Grounds as ‘Blurred Lines’, 98.7 AMP Radio (Mar. 11, 2015, 7:13 PM), (“Most copyright infringement cases are settled out of court so the decision doesn’t lay in the hands of the jury, who are just regular humans largely unfamiliar with copyright laws.”); see also generally Shyamkrishna Balganesh, The Questionable Origins of the Copyright Infringement Analysis, 68 Stan. L. Rev. 791 (2016) (current citations point to prior draft on file with the author).

  9. [9]. This is a phrase often attributed to once-trendy technology visionary and aphorist, Marshall McLuhan, but none of the sources this author has consulted have confirmed that, or stated where or when he said it. See If It Works, It’s Obsolete, Marshall McLuhan,
    mcluhanisms (last visited Oct. 12, 2016).

  10. [10]. Susan L. Feagin, Reading with Feeling: The Aesthetics of Appreciation 34–35 (1996) (describing the differences among readers that make their responses particular to them).

  11. [11]. Michael Pressley & Peter Afflerbach, Verbal Protocols of Reading: The Nature of Constructively Responsive Reading 12–13 (1995); Derry Koralek & Ray Collins, How Most Children Learn to Read, Reading Rockets, (last visited Oct. 12, 2016).

  12. [12]. See generally Anne E. Cunningham & Keith E. Stanovich, What Reading Does for the Mind, 1 J. Direct Instruction 137 (2001).

  13. [13]. Louise M. Rosenblatt, Literature as Exploration 31 (5th ed. 1995).

  14. [14]. Many different terms could be used as a proxy for the recipient of art’s experience. This Article chooses reading because it calls forth the demonstrated complexity of reading textual works; it draws on the rich body of literature in reader response theory which originated in the study of literary texts but has been widely embraced in fields beyond literature; it emphasizes an active stance on the part of the recipient, who is not passively receiving, but actively constructing meaning; and finally, because it is more concise and accurate than “experiencing,” “encountering,” “engaging with,” or any of the other plausible candidates I considered.

  15. [15]. Univ. of Gothenburg, Fiction Reading as Medicine, ScienceDaily (Oct. 21, 2013), (describing a study that found that “bibliotherapy,” or reading fiction when sick as part of an array of therapeutic activities may increase rehabilitation and speed recovery).

  16. [16]. Rosenblatt, supra note 13, at 31, 67, 75 (describing the differences among readers that make their responses particular to them).

  17. [17]. See generally Emotion and the Arts (Mette Hjort & Sue Laver eds., 1997) (collecting essays on psychologically inflected approaches to reception across all artistic media); Reception Study: From Literary Theory to Cultural Studies (James L. Machor & Philip Goldstein eds., 2001) (collecting essays that theorize the complexity of reception from literature to film, music, television, and popular culture generally).

  18. [18]. See generally, e.g., Mieke Bal & Norman Bryson, Looking In: The Art of Viewing (2001) (theorizing audience response to both visual and literary art); Michael Fried, Absorption and Theatricality: Painting and Beholder in the Age of Diderot (1980) (reinterpreting 18th century painting in light of the spectator’s role); Alois Riegl, The Group Portraiture of Holland (Evelyn M. Kain trans., 2000) (arguing that Dutch painters were engaging with their imagined viewer and positing a new role for the beholder of a work of art); Laura Mulvey, Visual Pleasure and Narrative Cinema, in Film Theory and Criticism: Introductory Readings 833 (Leo Braudy & Marshall Cohen eds., 1975) (offering a groundbreaking theory of audience response in cinema in the form of the male gaze).

  19. [19]. Hans Robert Jauss, Toward an Aesthetic of Reception (Timothy Bahti trans., 1982); Zahr Kassim Said, Only Part of the Picture: A Response to Professor Tushnet’s Worth a Thousand Words, 16 Stan. Tech. L. Rev. 349, 351–53 (2013).

  20. [20]. See generally Reader-Response Criticism: From Formalism to Post-Structuralism (Jane P. Tompkins ed., 1980).

  21. [21]. See generally, e.g., Justin Hughes, Size Matters (or Should) in Copyright Law, 74 Fordham L. Rev. 575 (2005); Pamela Samuelson, A Fresh Look at Tests for Nonliteral Copyright Infringement, 107 Nw. U. L. Rev. 1821 (2013).

  22. [22]. I have argued in prior work that these interpretive choices structure copyright’s infringement analysis beneath the surface. Zahr K. Said, Reforming Copyright Interpretation, 28 Harv. J.L. & Tech. 469, 470 (2015); see also Christine Haight Farley, Judging Art, 79 Tul. L. Rev. 805, 807–08 (2005); Robert Kirk Walker & Ben Depoorter, Unavoidable Aesthetic Judgments in Copyright Law: A Community of Practice Standard, 109 Nw. U. L. Rev. 343, 346 (2015) (describing the “subjective and qualitative assessments” copyright law demands of judges); Alfred C. Yen, Copyright Opinions and Aesthetic Theory, 71 S. Cal. L. Rev. 247, 252 (1998).

  23. [23]. Ellis v. Diffie, 177 F.3d 503, 506 (6th Cir. 1999) (“[T]he ‘ordinary observer’ test... is the traditional standard of copyright infringement.”).

  24. [24]. 3 William F. Patry, Patry on Copyright § 9:69 n.8 (2016); see also Elga A. Goodman et al., 49 New Jersey Practice Series: Business Law Deskbook § 11:2(h) (2015) (“At least three Circuits, namely, the Fourth, Eighth, and NinthCircuits, have employed a two-step articulation of the substantial similarity test under which courts first apply the ‘extrinsic test,’ which ‘focus[es] on objective similarities in the details of the works,’ and, second, they apply the ‘intrinsic test,’ which ‘depend[s] on the response of the ordinary, reasonable person to the forms of expression.’” (footnotes omitted)).

  25. [25]. For instance, a court may resolve a copyright infringement dispute as a matter of law, on an early motion, so long as either there is nothing copyrightable in what the defendant has borrowed, or so long as no reasonable juror could find that the two works are substantially similar. Peter F. Gaito Architecture, LLC v. Simone Dev. Corp., 602 F.3d 57, 63 (2d Cir. 2010). Hence, a judge deciding a case can resolve it by adopting the perspective of the reasonable juror.

  26. [26]. Jane C. Ginsburg & Robert A. Gorman, Copyright Law 135 n.29 (2012) (Krofft’s test is one that “is rather confusing and misleading... applied inconsistently, and in any event largely [reductive] to an analysis very much like that in Arnstein.”).

  27. [27]. Folio Impressions, Inc. v. Byer Cal., 937 F.2d 759, 766 (2d Cir. 1991) (opining that “what is required [to determine substantial similarity] is only a visual comparison of the works, rather than credibility, which we are in as good a position to decide as was the district court”).

  28. [28]. La Resolana Architects, PA v. Reno, Inc., 555 F.3d 1171, 1180 (10th Cir. 2009) (“[T]he ‘ordinary observer,’ like the ‘reasonable person’ in tort law, is a legal fiction; it is the measure by which the trier of fact judges the similarity of two works.”).

  29. [29]. Carol Barnhart Inc. v. Econ. Cover Corp., 773 F.2d 411, 422 (2d Cir. 1985) (Newman, J., dissenting) (“Of course, the ordinary observer does not actually decide the issue; the trier of fact determines the issue in light of the impressions reasonably expected to be made upon the hypothetical ordinary observer.”).

  30. [30]. Walker & Depoorter, supra note 22, at 377–78.

  31. [31]. AMF Inc. v. Sleekcraft Boats, 599 F.2d 341, 348 (9th Cir. 1979) (listing evidence of consumer confusion as one of the factors considered in determining trademark infringement).

  32. [32]. Copyright law approximates this approach when it insists on the intended audience and market, moving from a hypothetical to a more empirically accurate construct. See, e.g., Kohus v. Mariol, 328 F.3d 848, 857 (6th Cir. 2003); Lyons P’ship, L.P. v. Morris Costumes, Inc., 243 F.3d 789, 801–02 (4th Cir. 2001). It still steers clear of empirical evidence demonstrating what actual audiences thought of particular works, emphasizing instead the intended market as a proxy for the author’s harm.

  33. [33]. Data E. USA, Inc. v. Epyx, Inc., 862 F.2d 204, 210 n.6 (9th Cir. 1988).

  34. [34]. Warner Bros. Inc. v. Am. Broad. Cos., 720 F.2d 231, 239 (2d Cir. 1983) (discussing the lower court’s decision to exclude plaintiff’s survey of audience responses to the works in question, on the grounds that “they were too general to be of any benefit”).

  35. [35]. In atypical cases, or specialized cases, copyright law departs from this rule. Kohus, 328 F.3d at 857 (technical illustrations); Lyons P’ship, 243 F.3d at 794–95 (Barney costumes bought by adults for children); Dawson v. Hinshaw Music Inc., 905 F.2d 731, 737–38 (4th Cir. 1990) (musical composition for choir directors). These departures are by no means uniform, however. See Williams v. Crichton, 84 F.3d 581, 590–91 (2d Cir. 1996) (holding that substantial similarity for children’s books should still be measured from the perspective of the ordinary, that is, adult observer).

  36. [36]. For a partial list, see generally Lionel Bently & Laura Biron, Discontinuities Between Legal Conceptions of Authorship and Social Practices: What, if Anything, is to be Done?, in The Work of Authorship 237, 237–76 (Mireille van Eechoud ed., 2014); see also generally Rosemary J. Coombe, The Cultural Life of Intellectual Properties: Authorship, Appropriation, and the Law (1998); Mark Rose, Authors and Owners: The Invention of Copyright (1993); David Saunders, Authorship and Copyright (1992); The Construction of Authorship: Textual Appropriation in Law and Literature (Martha Woodmansee & Peter Jaszi eds., 1994); Martha Woodmansee, The Author, Art, and the Market: Rereading the History of Aesthetics (1994); Oren Bracha, The Ideology of Authorship Revisited: Authors, Markets, and Liberal Values in Early American Copyright, 118 Yale L.J. 186 (2008); Christopher Buccafusco, A Theory of Copyright Authorship, 102 Va. L. Rev. 1229 (2016); Margaret Chon, The Romantic Collective Author, 14 Vand. J. Ent. & Tech L. 829 (2012); Carys J. Craig, Reconstructing the Author-Self: Some Feminist Lessons for Copyright Law, 15 J. Gender, Soc. Pol’y & L. 207 (2007); and Peter Jaszi, Toward a Theory of Copyright: The Metamorphoses of “Authorship”, 1991 Duke L.J. 455.

  37. [37]. See, e.g., Buccafusco, supra note 36; Farley, supra note 22; Yen, supra note 22.

  38. [38]. One notable exception is Laura A. Heymann’s piece, Everything Is Transformative: Fair Use and Reader Response, 31 Colum. J.L. & Arts 445, 466 (2008). Heymann brings reader response to bear on fair use’s transformative use analysis, and while her insights are sharp and thoughtful, she concedes that she does not delve into the field in detail for the purposes of that essay. Two other excellent articles that raise reader response theory to varying degrees, are Walker & Depoorter, supra note 22 and Yen, supra note 22.

  39. [39]. Carla Bruni, La Dernière Minute, on Quelqu’un M’a Dit (2005).

  40. [40]. Id.

  41. [41]. Id.

  42. [42]. Id.

  43. [43]. La Dernière Minute, Carla Bruni’s Songs in Translation,
    carlabrunissongsintranslation/la-derniere-minute (last visited Oct. 12, 2016).

  44. [44]. Id.

  45. [45]. Charlie Gillett, Carla Bruni: Quelqu’un M’a Dit, Guardian: Observer (March 21, 2004),,,1171263,00.html.

  46. [46]. David Orr, Adrienne Rich: Resolution Amid the ‘Turbulence’, NPR: Books (Mar. 29, 2012, 2:35 PM),

  47. [47]. Id.

  48. [48]. See generally Hilary E. Pearson, The ‘Terrible Sonnets’ of Gerard Manley Hopkins and the Spirituality of Depression, 46 Way 23 (2007).

  49. [49]. Jacob Lawrence: Prints, 1963–2000 A Comprehensive Survey, Hudson River Museum, (last visited Oct. 21, 2016).

  50. [50]. Artists, Harlem Renaissance,
    html (last visited Oct. 12, 2016).

  51. [51]. Susan Freudenheim, Jacob Lawrence: Artist’s Works Chronicled History of U.S. Blacks, L.A. Times (June 10, 2000),; Jacob Lawrence, Foremost American Painter, U. Wash., (last visited Oct. 21, 2016).

  52. [52]. Philip C. Kissam, Thinking (by Writing) About Legal Writing, 40 Vand. L. Rev. 135, 152‍‍–‍53 (1987) (“Much of our reading in everyday life, including the everyday life of the law, is instrumental in the sense that we read to obtain directly useful or transferable knowledge: knowledge that we can put to certain uses without reflecting extensively upon what we have read, without needing to interpret this knowledge, or without needing to evaluate or criticize what we have read.”).

  53. [53]. Louise M. Rosenblatt, The Reader, The Text, The Poem: The Transactional Theory of the Literary Work 96 (1978) (“But as we read we cannot confine our responses within the scope of the ideas and images directly evoked from the words. The emotional overtones are, we may say, too powerful.”).

  54. [54]. Wayne Booth, Foreword, in Literature as Exploration xi (5th ed. 1995).

  55. [55]. See Jeanne C. Fromer, Expressive Incentives in Intellectual Property, 98 Va. L. Rev. 1745, 1746–47 (2012); Jeanne C. Fromer & Mark A. Lemley, The Audience in Intellectual Property Infringement, 112 Mich. L. Rev. 1251, 1268 (2014).

  56. [56]. Universal Furniture Int’l, Inc. v. Collezione Europa USA, Inc., 618 F.3d 417, 435 (4th Cir. 2010).

  57. [57]. See, e.g., Litchfield v. Spielberg, 736 F.2d 1352, 1356 (9th Cir. 1984); Well-Made Toy Mfg. Corp. v. Goffa Int’l Corp., 210 F. Supp. 2d 147, 162 (E.D.N.Y. 2002) (“[The] subjective paradigm has not taken root in the Second Circuit.”); Mark A. Lemley, Our Bizarre System for Proving Copyright Infringement, 57 J. Copyright Soc’y U.S.A. 719, 723 (2010).

  58. [58]. Walker & Depoorter, supra note 22, at 374.

  59. [59]. Lyons P’ship, L.P. v. Morris Costumes, Inc., 243 F.3d 789, 802 (4th Cir. 2001).

  60. [60]. Dawson v. Hinshaw Music Inc., 905 F.2d 731, 738 (4th Cir. 1990).

  61. [61]. Copeland v. Bieber, 789 F.3d 484, 491–92 (4th Cir. 2015) (“[B]ecause the general public typically encounters popular music songs by hearing them from start to finish, we undertake that analysis by listening to the songs in their entirety and side by side, to determine whether a reasonable jury could find that they are subjectively similar.”).

  62. [62]. Lemley, supra note 57, at 729.

  63. [63]. Feist Publ’ns, Inc., v. Rural Tel. Serv. Co., 499 U.S. 340, 361 (1991).

  64. [64]. Robert C. Osterberg & Eric C. Osterberg, Substantial Similarity in Copyright Law § 2.5.1 (2004).

  65. [65]. Id. § 1.1 (stating that substantial similarity “is not a formula or a test. It is a conclusion... the jury in a copyright infringement case is called upon to determine whether unauthorized copying rises to the level that it should be compensable”).

  66. [66]. Country Kids ‘N City Slicks, Inc. v. Sheen, 77 F.3d 1280, 1288 (10th Cir. 1996) (quoting Atari, Inc. v. N. Am. Philips Consumer Elecs. Corp., 672 F.2d 607, 614 (7th Cir. 1982) (emphasis added)); see also Ideal Toy Corp. v. Fab-Lu Ltd., 360 F.2d 1021, 1022 (2d Cir. 1966) (stating the standard to be “whether an average lay observer would recognize the alleged copy as having been appropriated from the copyrighted work”); Peter Pan Fabrics, Inc. v. Martin Weiner Corp., 274 F.2d 487, 489 (2d Cir. 1960) (holding that substantial similarity will be found where “the ordinary observer, unless he set out to detect the disparities, would be disposed to overlook them, and regard their aesthetic appeal as the same”).

  67. [67]. Osterberg & Osterberg, supra note 64, at § 1:1.

  68. [68]. Samuelson, supra note 21, at 1821.

  69. [69]. Value Grp., Inc. v. Mendham Lake Estates, L.P., 800 F. Supp. 1228, 1233 (D.N.J. 1992).

  70. [70]. See, e.g., 4 Melville B. Nimmer & David Nimmer, Nimmer on Copyright § 13.03[A] (2015); Katherine Lippman, Note, The Beginning of the End: Preliminary Results of an Empirical Study of Copyright Substantial Similarity Opinions in the U.S. Circuit Courts, 2013 Mich. St. L. Rev. 513, 515.

  71. [71]. Steinberg v. Columbia Pictures Indus., Inc., 663 F. Supp. 706, 711 (S.D.N.Y. 1987).

  72. [72]. Lippman, supra note 70, at 525.

  73. [73]. See Osterberg & Osterberg, supra note 64.

  74. [74]. Nichols v. Universal Pictures Corp., 45 F.2d 119, 122 (2d Cir. 1930).

  75. [75]. Comput. Assocs. v. Altai, Inc., 982 F.2d 693, 706–11 (2d Cir. 1992).

  76. [76]. Arnstein v. Porter, 154 F.2d 464, 469–71 (2d Cir. 1946).

  77. [77]. Roth Greeting Cards v. United Card Co., 429 F.2d 1106, 1110 (9th Cir. 1970).

  78. [78]. Sid & Marty Krofft Tele. Prods., Inc. v. McDonald’s Corp., 562 F.2d 1157, 1164 (9th Cir. 1977).

  79. [79]. Arnstein, 154 F.2d at 469–71. Early references to the ordinary or average observer in copyright law did exist, but they do not appear to have had the traction in case law that Arnstein has had. See Twentieth Century-Fox Film Corp. v. Stonesifer, 140 F.2d 579, 582 (9th Cir. 1944) (“The two works involved in this appeal should be considered and tested, not hypercritically or with meticulous scrutiny, but by the observations and impressions of the average reasonable reader and spectator.”).

  80. [80]. Dawson v. Hinshaw Music Inc., 905 F.2d 731, 733 (4th Cir. 1990); see also Michelle V. Francis, Comment, Musical Copyright Infringement: The Replacement of Arnstein v. Porter—A More Comprehensive Use of Expert Testimony and the Implementation of an “Actual Audience” Test, 17 Pepp. L. Rev. 493, 494–95 (1990).

  81. [81]. Balganesh, supra note 8, at 11–12.

  82. [82]. Id.

  83. [83]. Id.

  84. [84]. See Arnstein v. Am. Soc’y of Composers, Authors, & Publishers, 29 F. Supp. 388, 391 (S.D.N.Y. 1939).

  85. [85]. See Balganesh, supra note 8, at 3–4; see also Denise Cote, Making Experts Count, 58 J. Copyright Soc’y U.S.A. 223, 225 (2010).

  86. [86]. Balganesh, supra note 8, at 3–4.

  87. [87]. Id. at 3.

  88. [88]. Arnstein v. Porter, 154 F.2d 464, 468 (2d Cir. 1946).

  89. [89]. Id. (noting that, in the copying step, “[i]f there is evidence of access and similarities exist, then the trier of fact must determine whether the similarities are sufficient to prove copying”).

  90. [90]. Id.

  91. [91]. Osterberg & Osterberg, supra note 64, § 3.1.1.A.

  92. [92]. Id.

  93. [93]. Ginsburg & Gorman, supra note 26, at 135.

  94. [94]. Arnstein, 154 F.2d at 468.

  95. [95]. Balganesh, supra note 8, at 28; see also Neil Duxbury, Patterns of American Jurisprudence 135–36 (1995).

  96. [96]. Balganesh, supra note 8, at 5; see also Robert Dudnik, Comment, Special Verdicts: Rule 49 of the Federal Rules of Civil Procedure, 74 Yale L.J. 483, 490 (1965).

  97. [97]. “I doubt that any other literary critic of this century has enjoyed and suffered as sharp a contrast of powerful influence and absurd neglect as Louise Rosenblatt. Has she been influential? Immensely so: how many other critical works first published in the late thirties have extended themselves, like this one, to five editions, proving themselves relevant to decade after decade of critical and pedagogical revolution?” Booth, supra note 54, at vii.

  98. [98]. Rosenblatt, supra note 53, at 1.

  99. [99]. Id. at 2; see also generally M.H. Abrams, The Mirror and the Lamp: Romantic Theory and the Critical Tradition (1953).

  100. [100]. See generally Erich Auerbach, Mimesis: The Representation of Reality in Western Literature (Willard R. Trask trans., 1953).

  101. [101]. Paul Goring et al., Studying Literature: The Essential Companion 407 (2d ed. 2010).

  102. [102]. Abrams, supra note 99.

  103. [103]. See generally Matthew Arnold, Culture and Anarchy: An Essay in Political and Social Criticism (1869); John Ruskin, Modern Painters (1843).

  104. [104]. See generally Alastair Pennycook, The Cultural Politics of English as an International Language (1994).

  105. [105]. Todd F. Davis & Kenneth Womack, Formalist Criticism and Reader-Response Theory 21 (Julian Wolfreys ed., 2002).

  106. [106]. Id. at 14.

  107. [107]. Id.

  108. [108]. Id.

  109. [109]. In the post-World War II era, with the rise of Existentialist literature and increased emphasis on political commitment in the arts—sometimes known as la littérature engagé—an apolitical theory of reading became harder to defend. Littérature engagée, Encyclopaedia Britannica Online, http:// (last visited Oct. 13, 2016).

  110. [110]. Reader-Response Criticism: From Formalism to Post-Structuralism, supra note 20, at ix.

  111. [111]. Id.

  112. [112]. The movement’s rallying cry can arguably be found in the first line of a foundational essay that Wolfgang Iser published in 1971. See Wolfgang Iser, The Reading Process: A Phenomenological Approach, 3 New Literary Hist. 279, 279 (1972) (“The phenomenological theory of art lays full stress on the idea that, in considering a literary work, one must take into account not only the actual text but also, and in equal measure, the actions involved in responding to that text.”).

  113. [113]. Linda M. Scott, The Bridge From Text to Mind: Adapting Reader-Response Theory to Consumer Research, 21 J. Consumer Res. 461, 463 (1994).

  114. [114]. Reader-Response Criticism: From Formalism to Post-Structuralism, supra note 20, at x.

  115. [115]. Id.

  116. [116]. Id. at ix; see also generally Susan Rubin Suleiman & Inge Crosman, The Reader in the Text: Essays on Audience and Interpretation (1980).

  117. [117]. Rosenblatt, supra note 13, at 4; see also Wanda Brooks & Susan Browne, Towards a Culturally Situated Reader Response Theory, 43 Child. Literature Educ. 74, 74–75 (2012).

  118. [118]. Reader-Response Criticism: From Formalism to Post-Structuralism, supra note 20, at 201.

  119. [119]. Walker & Depoorter, supra note 22, at 356–57 (“The most extreme version of [reader response] theory holds that works of art never have a fixed or universally accepted meaning. Because no two readers share the same set of aesthetic assumptions, there can be no consensus as to the ‘correct’ meaning of a work, and interpretation is a mere function of the reader’s preferences. Thus, all aesthetic disagreements are unresolvable matters of personal taste.” (footnotes omitted)); see also 1 William F. Patry, Patry on Copyright § 2:24 (2016) (discussing the standard critique of radical response theory thus: “because there is, allegedly, no ‘there there’ in the text, all interpretations are subjective and all equally correct since there is no foundational object against which to measure the validity of an interpretation”).

  120. [120]. Rosenblatt, supra note 53, at 23.

  121. [121]. Id. at 24.

  122. [122]. Id. at 23.

  123. [123]. Id.

  124. [124]. See generally Rosenblatt, supra note 13.

  125. [125]. Rosenblatt, supra note 53, at 23.

  126. [126]. Alas, students for generations have believed such summaries to be acceptable substitutes even for the aesthetic experience of the work; the authors of the Cliffs Notes and the Spark Notes and such have counted on it. Rosenblatt’s theory illustrates why the two experiences—efferent and aesthetic—ought to be distinguished. This smug pronouncement, by the way, issued by an author of one of the Spark Notes on a classic novel of African literature.

  127. [127]. Rosenblatt, supra, note 53, at 24.

  128. [128]. See, for example: “Place it firmly over your nose and mouth, secure the elastic band behind your head, and breathe normally. Although the bag does not inflate, oxygen is flowing to the mask. If you are travelling with a child or someone who requires assistance, secure your mask on first, and then assist the other person.” Inflight Passenger Announcements,, (last visited Oct. 13, 2016) (emphasis added).

  129. [129]. See, e.g., Emily Dickinson, Shipwreck, in Poems, Series Two 48 (1891) (“It tossed and tossed, — / A little brig I knew, — / O’ertook by blast, / It spun and spun, / And groped delirious, for morn. / It slipped and slipped, / As one that drunken stepped; / Its white foot tripped, / Then dropped from sight. / Ah, brig, good-night / To crew and you; / The ocean’s heart too smooth, too blue, / To break for you.”); Sophie Hannah, Tide to Land, in First of the Last Chances 19 (2003) (“I know the rules and hear myself agree / Not to invest beyond this one night stand. / I know your pattern: in, out, like the sea. / The sharp north wind must blow away the sand.... It’s not as if we were designed to be / Strolling along the beach front, hand in hand. / Things change, of natural necessity. / The sharp north wind must blow away the sand / And every storm to rage, however grand, / Will end in pain and shipwreck and debris / And each time there’s a voice I have to strand / On a bare rock, hardened against its plea. / I know the rules.”); see also Adrienne Rich, Diving into the Wreck,, (last visited Oct. 13, 2016). In this poem, Adrienne Rich wrote about a scuba diver exploring a shipwreck, thus allowing Rich to map emotional self-searching onto a literal experience of plumbing deep water on a kind of quest.

  130. [130]. Such as Yehuda Amichai’s poem, A Pity. We Were Such a Good Invention, in which the final five lines run, “A pity. We were such a good / And loving invention. / An aeroplane made from a man and wife. / Wings and everything. / We hovered a little above the earth. / We even flew.” The Selected Poetry of Yehuda Amichai 57 (Chana Bloch & Stephen Mitchell trans., 2013).

  131. [131]. Adrienne Rich, Turbulence, in Tonight No Poetry Will Serve: Poems 2007–2010 23, 23–24 (2011).

  132. [132]. Id.

  133. [133]. Compare Gerard M. Hopkins, No Worst There Is None. Pitched Past Pitch of Grief, in A Mind Apart: Poems of Melancholy, Madness, and Addiction 180 (Mark S. Bauer ed., 2009) (“O the mind, mind has mountains; cliffs of fall / Frightful, sheer, no-man-fathomed. Hold them cheap / May who ne’er hung there.”), with Rich, supra note 131, at 23–24 (“Your / mind, mind has mountains, cliffs of fall / may who ne’er hung there let him/watch the movie.”).

  134. [134]. See Rosenblatt, supra note 53, at 23.

  135. [135]. Id.

  136. [136]. See generally Shyamkrishna Balganesh, The Normativity of Copying in Copyright Law, 62 Duke L.J. 203 (2012).

  137. [137]. Lemley, supra note 57, at 741.

  138. [138]. Id. (“It is far from clear that juries can do that line-drawing justice, even with the aid of expert testimony and jury instructions telling them to do so.”).

  139. [139]. See generally Nicole K. Roodhuyzen, Note, Do We Even Need a Test? A Reevaluation of Assessing Substantial Similarity in a Copyright Infringement Case, 15 J.L. & Pol’y 1375 (2009).

  140. [140]. Rebecca Tushnet, Worth a Thousand Words: The Images of Copyright, 125 Harv. L. Rev. 683, 738–40 (2012).

  141. [141]. Id. at 740.

  142. [142]. See, e.g., Jeannette Rene Busek, Comment, Copyright Infringement: A Proposal for a New Standard for Substantial Similarity Based on the Degree of Possible Expressive Variations, 45 UCLA L. Rev. 1777, 1787–91 (1998); Lemley, supra note 57, 730–41 (urging broad adoption of the filtration test used for software, even beyond software); see also generally Aaron M. Broaddus, Eliminating the Confusion: A Restatement of the Test for Copyright Infringement, 5 DePaul-LCA J. Art & Ent. L. 43 (1995); Laura G. Lape, The Metaphysics of the Law: Bringing Substantial Similarity Down to Earth, 98 Dick. L. Rev. 181 (1994).

  143. [143]. Jamie Lund, Fixing Music Copyright, 79 Brook. L. Rev. 61, 91 (2013) (“Lay jurors are equally likely to find infringement when different compositions are performed similarly as they are to find infringement when identical compositions are performed differently.”); see also generally Shyamkrishna Balganesh et al., Judging Similarity, 100 Iowa L. Rev. 267 (2014); Irina D. Manta, Reasonable Copyright, 53 B.C. L. Rev. 1303 (2012); Austin Padgett, Note, The Rhetoric of Predictability: Reclaiming the Lay Ear in Music Copyright Infringement Litigation, 7 Pierce L. Rev. 125 (2008).

  144. [144]. An excellent pioneering contribution in that direction is Fromer & Lemley, supra note 55, at 1251.

  145. [145]. Walker v. Time Life Films, Inc., 784 F.2d 44, 50 (2d Cir. 1986).

  146. [146]. Lemley, supra note 57, at 726–29.

  147. [147]. Id. at 726 (citing Kohus v. Mariol, 328 F.3d 848 (6th Cir. 2003) which “applies a higher standard than most for purposes of admitting expert testimony”).

  148. [148]. Id. at 730.

  149. [149]. As a separate matter, the criteria courts have historically used for determining which works count as “technical” and which ones do not, smacks of disciplinary bias. See generally Said, supra note 22 (arguing that copyright wrongly characterizes software as complex while treating artistic works as though they lacked interpretive complexity).

  150. [150]. Experts could assist aesthetic reading, too, but the goal of such assistance would not be to derive a correct answer to an informational or threshold question as much as it would be to assist a jury to find cues in a work that it could then use to help it evoke the work aesthetically.

  151. [151]. Samuelson, supra note 21, at 1844.

  152. [152]. See generally Cote, supra note 85.

  153. [153]. Dawson v. Hinshaw Music Inc., 905 F.2d 731, 737 (4th Cir. 1990).

  154. [154]. Litchfield v. Spielberg, 736 F.2d 1352, 1357 (9th Cir. 1984) (characterizing the second stage of the infringement analysis as intrinsic, or subjective, focusing on “the total concept and feel of the works” and excluding expert evidence).

  155. [155]. Gordon, supra note 3; see also generally Balganesh et al., supra note 143.

  156. [156]. 4 Melvin Nimmer & David Nimmer, Nimmer on Copyright § 13.03[E][1][a] (2015) (“If the works in issue are directed to a particular audience, then the ‘spontaneous and immediate’ reaction of that audience is determinative.”).

  157. [157]. Erika A. Khalek, Note, Searching for a Harmless Alternative: Applying the Harmless Error Standard to Alternative Theory Jury Instructions, 83 Fordham L. Rev. 295, 300 (2014).

  158. [158]. Id. at 300 (quoting 1 Kevin F. O’Malley et al., Federal Jury Practice and Instructions § 7 App. (6th ed. 2006)).

  159. [159]. See Manta, supra note 143, at 1335–36 (“The jury instructions used in various circuits [vary]. Some of them ask jurors to determine what an ‘ordinary reasonable person’ would find substantially similar, whereas others make no reference to the reasonable person and tell jurors to determine whether two works are substantially similar in the abstract, and yet others speak of the ‘average lay observer’ but equate that concept with the jurors themselves.” (footnotes omitted)).

  160. [160]. Model Civ. Jury Instrs. 9th Cir. § 17.15 (2007).

  161. [161]. Lemley, supra note 57, at 738 (“While in theory jury instructions can instruct jurors to exclude all these elements, in practice jurors aren’t going to know what things are, for example, scenes a faire in the music industry without some testimony on standard chord progressions.”).

  162. [162]. See Chuman v. Wright, 76 F.3d 292, 294 (9th Cir. 1996) (“Jury instructions must be formulated so that they fairly and adequately cover the issues presented, correctly state the law, and are not misleading.”).

  163. [163]. Balganesh et al., supra note 143, at 268–69.

  164. [164]. Balganesh, supra note 8, at 3.

  165. [165]. Model Civ. Jury Instrs. supra note 160, § 17.17.

  166. [166]. Lemley, supra note 57, at 739–40.

  167. [167]. Mark A. Lemley & Mark P. McKenna, Scope, 57 Wm. & Mary L. Rev. 2197, 2226–39 (2016) (showing the effects of jury bias in copyright cases).

  168. [168]. Donald Olander, Note, Resolving Inconsistencies in Federal Special Verdicts, 53 Fordham L. Rev. 1089, 1089 (1985).

  169. [169]. Id.

  170. [170]. Dudnik, supra note 96, at 483.

  171. [171]. Edson R. Sunderland, Verdicts, General and Special, 29 Yale L.J. 253, 261 (1920).

  172. [172]. Dudnik, supra note 96, at 488 (quoting Sunderland, supra note 171, at 262).

  173. [173]. Sunderland, supra note 171, at 264–65; see also Olander, supra note 168.

  174. [174]. Rosenblatt, supra note 53, at 175 (quoting Democratic Vistas, in 2 Prose Works 1892, at 361, 424–25 (Floyd Stovall ed., 1964)).

  175. [175]. Bruni, supra note 39.

  176. [176]. Rich, supra note 131.

  177. [177]. Gerard Manley Hopkins, Poems and Prose (W.H. Gardner ed., 1953).

  178. [178]. An image of the work is also available online, where you may be better able to discern details. Jacob Lawrence: Prints, 1963–2000 A Comprehensive Survey, supra note 49.


Associate Professor, University of Washington School of Law. 

Many thanks to Clark Asay, Stephanie Bair, Shyam Balganesh, Chris Buccafusco, Mike Burstein, Ryan Calo, Julie Cohen, Rebecca Curtin, M.J. Durkee, Dave Fagundes, Andrew Gilden, Brad Haque, Paul Heald, Laura Heymann, Anita Krug, Mark Lemley, Jacqueline Lipton, Lisa Manheim, Liz Porter, Betsy Rosenblatt, Matt Sag, Pamela Samuelson, Andrew Stauffer, Rebecca Tushnet, and David Ziff. Special thanks to the redoubtable Jane Pryjmak, for truly outstanding research assistance. This Article benefited enormously from feedback received on various versions of the work at the Junior Scholars in Intellectual Property Workshop at Michigan State University (2016); the BYU IP Colloquium (2016); the IP Colloquium at Cardozo University (2016); the University of Pennsylvania First Annual Copyright Roundtable (2015); the Jurors were asked to make rence,a particular kind of reading that can be better understood by understanding theories of reading Law and Society Association Annual Conference (Seattle, 2015); a Northwest Copyright Society Panel (Seattle, 2015); the Intellectual Property Scholars Conference (De Paul, 2015); the University of Washington Legal Methods Workshop (2015); and WIPIP (UW Law, 2015). UW Law librarians provided fantastic research assistance; they spoil our faculty every day.