Articles

Authors at Work: The Origins

of the Work-for-Hire Doctrine

Catherine L. Fisk*

I. INTRODUCTION

The death of the author was announced in literary circles quite

some time ago.1 Rumors of the author's demise were, in my view,
premature. The author isn't dead; he just got a job. Unfortunately, as
if in a company-man dystopia, he has been subsumed into the
identity of his corporate employer. His disappearance is by now

* Professor of Law, University of Southern California. I gratefully acknowledge financial

and institutional support of the schools where I taught when I wrote this: Duke Law School
and Loyola Law School, Los Angeles. Thanks to Bob Chang, Erwin Chemerinsky, Bob
Gordon, Zorina Khan, Dan Klerman, Kim Krawiec, David Lange, Greg Mark, Barak Orbach,
Jeff Powell, Mark Rose, Simon Stern, and participants at law school workshops at Duke
University, New York University, the University of North Carolina, and Loyola Law School
for commenting on drafts, to the editors of the Yale Journal of Law & the Humanities for
thoughtful edits, and to Loyola and Duke students Eric Compere, Ian Fried, An Le, Lin Lee,
Max Rieger, and Mary Roccapriore for research assistance. Whatever may be said about
individual authorship, individual responsibility for errors and shortcomings is mine.
1. ROLAND BARTHES, THE RUSTLE OF LANGUAGE 49 (Richard Howard trans., 1989)
(referring to "the death of the author").
 Yale Journal of Law & the Humanities [Vol. 15:1

almost complete. Although he has gone on writing, the corporation

has become the author of his oeuvre.
Yet the desire both to create and to be recognized as a creator is
irrepressible. The creative process is both inherently individual and
inescapably social. So even as the author has been submerged as
employee, she claims recognition in modern copyright law in other
ways--perhaps as a joint author or through moral rights. The history
I will tell here of the nineteenth-century law's efforts to mediate the
inevitable tension between individual and collective creative rights
tells us something about law and the ownership of ideas today. As I
explain in the pages that follow, nineteenth-century lawyers tried to
short-circuit the difficult and complex issue of balancing individual
and collective rights through a neat doctrinal rule granting all rights
to the employee, but the tidiness did not last. When, at the turn of
the twentieth century, courts and Congress adopted the modern
work-for-hire doctrine allocating copyrights to the employer, lawyers
again hoped that bright-line rules would prevail. Again doctrinal
confusion surfaced, revealing the underlying tension. Today, the
same tension appears in the debate about the desirable scope of joint
authorship.' The author is not dead, and authorship is not a fiction. It
has never been easy to divvy up the honor and the money among the
many people, both natural and corporate, who contribute to the
creative process.
Recent scholarship in law and literary studies has traced the
process by which the modern notion of an author evolved with the
development of copyright law.3 This literature argues that the

2. I am indebted to David Lange for this insight about the relationship between work-for-
hire and joint authorship, and to his article, At Playin the Fields of the Word- Copyrightand
the ConstructionAuthorship in the Post-Literate Millennium, 55 LAW & CONTEMP. PROBS.
139 (1992), for the notion that authorship is irrepressible. On the contemporary debate over
the scope of joint authorship, see Rochelle Cooper Dreyfuss, Collaborative Research:
Conflicts on Authorship, Ownership. and Accountability, 53 VAND. L. REV. 1162 (2000);
Roberta Rosenthal Kwall, "Author-Stories" Narrative'sImplications for Moral Rights and
Copyright's Joint Authorship Doctrine, 75 S. CAL. L. REV. 1 (2001); and Mary LaFrance,
Authorship. Dominance, and the Captive Collaborator. Preserving the Rights of Joint
Authors, 50 EMORY L.J. 193 (2001). Peter Jaszi did seminal earlier work on the tension
between individual and collaborative work in the joint authorship area. See, e.g., Peter Jaszi,
On the Author Effect: Contemporary Copyright and Collective Creativity, in THE
CONSTRUCTION OF AUTHORSHIP: TEXTUAL APPROPRIATION IN LAW AND LITERATURE 29-
35 (Martha Woodmansee & Peter Jaszi eds., 1994). Substantially the same article appears
under the same title in 10 CARDOzO ARTS & ENT. L.J. 293 (1992).
3. Prominent histories of copyright in American, British, and European law include JAMES
J. BARNES, AUTHORS, PUBLISHERS AND POLITICIANS: THE QUEST FOR COPYRIGHT
AGREEMENT, 1815-1854 (1974); JAMES BOYLE, SHAMANS, SOFTWARE, AND SPLEENS: LAW
AND THE CONSTRUCTION OF THE INFORMATION SOCIETY (1996); THE CONSTRUCTION OF
AUTHORSHIP: TEXTUAL APPROPRIATION IN LAW AND LITERATURE, supra note 2; JOHN
FEATHER, PUBLISHING, PIRACY AND POLITICS: AN HISTORICAL STUDY OF COPYRIGHT IN
BRITAIN (1994); BENJAMIN KAPLAN, AN UNHURRIED VIEW OF COPYRIGHT (1967); MARK
ROSE, AUTHORS AND OWNERS: THE INVENTION OF COPYRIGHT (1993); PAUL K. SAINT-
2003] Fisk

development of copyright law in England and Europe from the

eighteenth4 century onward privileged the concept of the individual
"author." Michel Foucault's influential essay, What is an Author?
launched a wave of literary and historical studies on the origins of
the modern concept of authorship.' Most have argued that the legal
and literary emphasis on the individual author is overblown and that
collaborative and communal practices of creation have been
insufficiently noticed.'
Turning from literary history to economic history, one sees a
similar interest in the process by which property rights in creative
works developed over the course of the nineteenth century.'

AMOUR, INTELLECTUAL PROPERTY AND THE LITERARY IMAGINATION (2003): DAVID

SAUNDERS, AUTHORSHIP AND COPYRIGHT (1992); and, JACK STILLINGER, MULTIPLE
AUTHORSHIP AND THE MYTH OF SOLITARY GENIUS (1991). Particularly important analysis of
copyright from the perspective of economic history, rather than literary or legal history, is
being done by B. Zorina Khan. Her work is unique in giving a macro-perspective on how
copyrights functioned and how, in the absence of copyright for works of foreign authors,
American publishers created their own form of monopoly rights. See B. ZORINA KHAN, "THE
FUEL OF INTEREST": PATENTS AND COPYRIGHTS IN AMERICAN ECONOMIC DEVELOPMENT
(forthcoming 2003); B. Zorina Khan, Do Property Rights Matter? International Copyrights
and the Development of the American Economy, 1790-1910 (unpublished manuscript on file
with author). On the debate between economically and literarily-inclined approaches to
copyright law, see for example Mark A. Lemley, Romantic Authorship and the Rhetoric of
Property 75 TEX. L. REV. 873 (1997) (reviewing JAMES BOYLE, SHAMANS, SOFTWARE, AND
SPLEENS: LAW AND THE CONSTRUCTION OF THE INFORMATION SOCIETY (1996)).
4. See. e.g., SAINT-AMOUR, supra note 3 (surveying the history of copyright literature and
studying appearances of the notion of copyright in nineteenth and twentieth-century English
and American poetry and novels); Peter Jaszi, Toward a Theory of Copyright. The
Metamorphosesof "Authorship,"1991 DUKE L.J. 455 (1991) (United States.); Mark Rose, The
Author as Proprietor.Donaldson v. Becket and the Genealogy of Modern Authorship, 23
REPRESENTATIONS 51 (1988) (England); Martha Woodmansee, The Genius and the
Copyright. Economic and Legal Conditions of the Emergence of the 'Author." 17
EIGHTEENTH CENTURY STUD. 425 (1984) (Germany).
5. Michel Foucault, What is an Author?, in TEXTUAL STRATEGIES: PERSPECTIVES IN
POST-STRUCTURALIST CRITICISM 141 (J. Harari ed. & trans., 1979).
6. The origins of the literary and legal discussions of the history of copyright and the
notion of authorship, and the convergence of the two bodies of scholarship, are ably traced by
Jaszi, supranote 2 at 29-32.
7. See, e.g., KHAN, supra note 3; Robert Merges, One Hundred Years of Solicitude:
Intellectual PropertyLaw. 1900-2000,88 CAL. L. REV. 2187 (2000). One sees similar interest in
economic histories of intellectual property rights other than copyrights. See. e.g., B. Zorina
Khan & Kenneth L. Sokoloff, The Early Development of IntellectualPropertyInstitutions in
the United States, 15 J. ECON. PERSP. 233 (2001); Naomi R. Lamoreaux & Kenneth L.
Sokoloff, The Geographyof Invention in the American Glass Industry 1870-1925,60 J. ECON.
HIST. 700 (2000); Naomi R. Lamoreaux & Kenneth L. Sokoloff, Inventors, Firms, and the
Market for Technology in the Late Nineteenth andEarly Twentieth Centuries, in LEARNING
BY DOING IN MARKETS, FIRMS, AND COUNTRIES (Naomi L. Lamoreaux et al. eds., 1999);
Naomi R. Lamoreaux & Kenneth L. Sokoloff, Long-Term Change in the Organization of
Inventive Activity, 93 PROC. NAT. ACAD. SCI. 12686 (1996); Naomi R. Lamoreaux & Kenneth
L. Sokoloff, Market Trade in Patentsand the Rise of a Class of Specialized Inventors in the
lgh-Century United States, 91 AM. ECON. ASS'N PAPERS & PROC 39 (2001); Kenneth L.
Sokoloff, Inventive Activity in Early IndustrialAmerica:Evidence from Patent Records. 1790-
1846, 48 J. ECON. HIST. 813 (1988); Kenneth J. Sokoloff & B. Zorina Khan, The
 Yale Journal of Law & the Humanities [Vol. 15:1

Economic historians have noted the collaborative nature of creation

and the process by which property rights were allocated so as to
enable the development of efficient markets for creative products.
From the economic perspective, the ability of the firm to obtain
property rights in its employees' creative products is a significant
feature of an efficient intellectual property regime.'
Attribution of authorship may be a fiction in the eyes of literary
critics and a business transaction in the eyes of economists, but in the
eyes of the law it is, for the most part, simply a question of evidence.'
Under modern copyright law, "copyright in a work ...vests initially
in the author or authors of the work,"1 and as far as lawyers are
concerned, identifiable people are in fact the creators of all work
eligible for copyright protection. Attribution of authorship is a
matter of proof, not a process of cultural attribution, and when
corporations are deemed authors, it is because they have hired
people who created works.
One significant exception to the law's disregard for the fiction of
authorship is in the case of "works made for hire." Here, the
Copyright Act expressly recognizes the author as a legal fiction. If a
work meets the statutory definition of a work made for hire, "the
employer or other person for whom the work was prepared is
considered the author" for purposes of federal copyright law.' The
statute specifies that a work is "made for hire" under two
circumstances: (1) if an employee (a term of art) prepares it "within
the scope of his or her employment,"12 or (2) if one of nine statutorily
enumerated types of work prepared by persons not meeting the legal
definition of employee was "specially ordered or commissioned."' 3

Democratization of Invention During Early Industrialization: Evidence from the United

States, 1790-1846,50 J. ECON. HIST. 363 (1990).
8. I do not mean to suggest a unanimity of opinion among economists (or anyone else) on
what allocation of intellectual property rights between employer or employee is desirable or
efficient, or even on whether there should be any such rights at all. See.e.g., Jack Hirshleifer,
The Privateand Social Value of Information and the Reward to Inventive Activity, 61 AM.
ECON. REV. 561 (1971); Neal Netanel, Copyright and a Democratic Civil Society, 106 YALE
L.J. 283 (1996).
9. There are significant exceptions to this. See F. Jay Dougherty, Not a Spike Lee Joint?
Issues inthe Authorship of Motion Pictures Under U.S.Copyright Law,49 UCLA L. REV. 225
(2001).
10. 17 U.S.C. § 201(a).
11. Id.§ 201(b).
12. Id.§ 101(1).
13. Id.§ 101(2) (emphasis added). The nine categories are (1) contributions to a collective
work, (2) motion pictures or other audiovisual works, (3) translations, (4) supplementary
works, (5) compilations, (6) instructional texts, (7) tests, (8) answer material for a test, and (9)
atlases. However, the parties must expressly agree in a written instrument signed by them that
the work shall be considered a work made for hire. Id.Thus, the default rule is that works by
independent contractors are not works for hire.
2003] Fisk

On the surface, there is nothing especially noteworthy in the

statute's choice to designate employers as "authors" of works for
hire. "Author" is simply a term of art, the operative term for the
proprietor of statutory rights. Statutes often do their substantive
work of according rights and protections by defining terms. On
further consideration, however, it becomes apparent that the legal
fiction that asserts the employer (usually a corporation) is the
"author" of a copyrighted work has greater significance than an
ordinary term of art. The notion of authorship is freighted with
connotations of creative genius and the moral entitlements that flow
from it. Congress chose to call the employer the author for reasons in
addition to ease in statutory drafting; it obviated some constitutional
questions, and it elided some moral issues as well.
One of the claims I make in this article is that the legal fiction of
corporate authorship does what Lon Fuller suggested legal fictions
always do. 4 That is, it persuades lawyers that the corporate employer
has a legitimate right to the copyright-the moral and legal
entitlements that flow from the exalted status of being an
author-without the necessity of explaining why. The legal fiction of
employer authorship was not always recognized as a fiction,
particularly in cases in which factual authorship might have been
shared among many employees of a firm. But even if employer
authorship is mainly fictional, it works as what Lon Fuller called an
"intellectual shortcut" to persuade when a statement of the actual
reasons for the ruling are difficult to explain." The legal fiction is an
expedient motivated by "the impulse to preserve a form of statement
that will make the law acceptable to those who do not have the time
or the capacity for understanding reasons which are not
obvious--and this 16 class sometimes includes the author of the
statement himself.'
A second claim I make in this article is descriptive: The creation of
the modern doctrine preceded its first appearance in the federal
Copyright Act of 1909. It neither was invented by the drafters of the
1909 Act, nor was it well recognized in the cases before 1909. The
concept began to appear after 1860, though no case actually applied
such a rule until a pair of cases did so in 1899 and 1900. As one legal
rule (a default rule of employee ownership) was replaced by its exact
opposite (a default rule of employer ownership) over the course of

14. LON L. FULLER, LEGAL FIcrIONS 9-19 (1967).

15. Id.at 26.
16. Id. An excellent modern work on legal fictions, including Fuller's theory and Jeremy
Bentham's vigorous attack on them, is Nomi Maya Stolzenberg, Bentham's Theory of
Fictions-A Curious "Double Language," 11 CARDOZO STUD. L. & LIT. 223, 231-36 (1999).
See also Aviam Soifer, Reviewing LegalFictions,20 GA. L. REV. 871 (1986).
 Yale Journal of Law & the Humanities [Vol. 15:1

seventy-five years, courts struggled to accommodate corporate

control over intellectual property to a misty-eyed regard for the
prerogatives of authors, artists, and composers. The process repeated
an old pattern in copyright history. Eighteenth-century booksellers
had conjured up and deployed the modern author-a creative genius
possessed of moral and legal entitlements to his creations-and used
him to argue for a copyright law protecting against cheaper reprints. 7
Early twentieth-century firms used that same mythic genius in their
effort to assert corporate control over an increasingly wide range of
intellectual property, while at the same time downplaying or ignoring
individual creative genius so as to assert corporate ownership over
those copyrighted works. The employee writer, artist, or composer
was simultaneously cloaked with the aura of the romantic genius and
merged into the corporate entity.
In the long discussion in legal and literary studies about the
mutually constitutive roles played by copyright law and popular and
academic understandings of authorship, many have noted but few
have studied the role of employed authors and employment
relations. Most legal, literary, and economic approaches to copyright
history note the development of the work-for-hire doctrine in late-
nineteenth-century American copyright law as an exception to the
individual authorship paradigm. 8 I ask: To what extent does the
employment relation make both individual authorship andcorporate
authorship fictional, and what role have metaphors about the nature
of authorship played? Both the dominant copyright metaphors of
authorship-the work as property (Locke's mixing one's labor with
the soil of culture) and the work as progeny (Defoe's "brat of my
brain")-break down when it is someone else's brain and ideas being
tilled or reproduced.19
The work-for-hire doctrine challenged the rhetoric of individual
authorship in more serious ways than is conventionally
acknowledged. Peter Jaszi has said that the work-for-hire doctrine
has been justified by courts in the usual Romantic authorship
paradigm: The "identification of employer as author is more than a
crude, instrumental fiction-rather, it is a logical (if perverse)
working out of the underlying assumption that the essence of
'authorship' lies in original, inspired creative genius."2 In his view,
"it is the 'employer's contribution as the 'motivating factor' behind

17. Rose, supra note 4, at 56. See alsoROSE, supra note 3.

18. As Peter Jaszi put it, the work-for-hire doctrine is "a dramatic example of the ways in
which the ideology of 'authorship' has been manipulated." Jaszi, supra note 2, at 34.
19. I am grateful to Mark Rose for illuminating the power of these metaphors for me.
20. Jaszi, supranote 2, at 34.
2003] Fisk

the work (in the words of one decision) that matters, rather than the
mere drudgery of the 'employee.' 2 ' In other words, Jaszi says, courts
still focus on the employer's contribution, when they insist that it is
the right to control the work of the employee that makes the
employer the author.22 As I will show, in the development of the
work-for-hire principle some courts did focus on the employer's
contribution or the employer's right to control. But over time many
came to rely simply on the legal fiction that employment renders the
employer the author. The "crude legal fiction" was not that the
employer's right to control made it the author; it was that the
employer was the author.
In addition to my jurisprudential claim about legal fictions and my
descriptive claim about what the law was and when it changed, I
make five interrelated explanatory claims about why the law started
out favoring employees and then changed. Multiple forces converged
to generate the radical change in the default rules of copyright
ownership between the early nineteenth century and the early
twentieth.
First, I argue that the facts of the cases presenting the question of
employer ownership of employees' creative products profoundly
affected the courts' perceptions of the competing claims. The
overwhelming majority of nineteenth-century cases involved either
legal publishing (court reporters and treatise authors) or theater.
The discourse of authorship really meant something concrete to
judges when it came to law reports, legal treatises, and theater; they
thought they knew exactly what counted as authorship. At the same
time, the law developed as it did because courts had a very definite
idea of what they wanted to accomplish in allocating property rights
in texts. In particular, early nineteenth-century judges thought
allocating copyright to employees would facilitate the development
of law publishing. Toward the end of the nineteenth century, and at
the beginning of the twentieth, cases about advertisements became
more common. Thus, a crucial part of the development of the work-
for-hire doctrine was the expansion of the types of works eligible for
copyright protection to include collaboratively produced commercial
media. The expansion of copyright made copyright ownership an
important asset to an ever-growing array of businesses and,
therefore, simultaneously created incentives for firms to claim
ownership of their employees' products and made it easier for courts
to accept these claims.
Second, a related explanation for the pattern of doctrinal change

21. Id.
22. Id. (citing Cmty. for Creative Non-Violence v. Reid, 490 U.S. 730, 751 (1989)).
 Yale Journal of Law & the Humanities [Vol. 15:1 .

involves dearly-held beliefs about the moral worth and inherent

individuality of the creative process. The romantic conception of
authorship as the expression of creative individual genius played a
major role in the early period. In the later period, courts were able to
reconcile employer ownership with authors' rights in some cases by
elaborating a fiction of employer authorship and, in others, by
positing the existence of collective creation. This is how legal fictions
about authorship played a crucial mediating role. These legal fictions
are a fundamental part of the common law process, in which change
is incremental, analogies are essential but contestable, and the facts
of the cases are of great importance.
Third, the relatively pro-employee rule of the nineteenth century
would have been unthinkable in the context of nineteenth-century
master-servant law but for the elevated social status of the people
who created the works. That the dominant cases involved the
Reporter of Supreme Court decisions and some of the most
celebrated mid-nineteenth century actor-playwrights was crucial in
garnering judicial respect for the rights of employee authors.
Although I refer to the employed authors as employees, most of
them would be independent contractors in modern parlance. Few,
especially in the early cases, were the sort of people who would have
been governed by master-servant law in the nineteenth century.
Thus, the copyright rules described here were not in conflict with
master-servant law so much as completely independent of
it--doctrinally, conceptually, and socially.
The increasingly bright distinction between employees and
independent contractors played a role late in the century as courts
reevaluated old assumptions about the nature of the employment
contract, the process by which employment contracts are negotiated,
and the role of contractual default rules. At the same time, the
ascendance of an objective theory of contract in the late nineteenth
century significantly expanded the ability of judges and lawyers to
make the conceptual leap from individual authorship to collective
creation that, as an inherent feature of the employment contract,
entitled the employer to own the intellectual property. The results of
some of the cases might not have differed between the early-
nineteenth century and now, inasmuch as modern law provides that
works by independent contractors are not deemed works made for
hire absent an express written agreement to that effect.23 But the
early cases, which involved workers who would today be
independent contractors, were sometimes relied on in later cases that
did involve workers who would today be deemed employees. Courts

23. See 17 U.S.C. § 101(2).

2003] Fisk

did not draw the distinction until the statute was later amended to
make it plain.24 In virtually all of the cases, whether the hired party
had a contractual relationship of independence or one of full-time
salaried employment, the hired party occupied a position that might
customarily have been filled by a person of considerable
independence. In a sense, then, over time the kind of workers to
whom the doctrinal categories applied changed as much as the
doctrinal categories themselves.
Fourth, closely related to the changes in employment practices and
employment law was the rise of the corporate form of business
organization. When the dispute over authorship was between one
natural person and another, judges tended to think that one of them
had to be the real author. But if the work was created
collaboratively, and one of the claimants was the corporation that
employed the creators, it was much easier to characterize the
corporate employer, in its role as the legal representative of the
collective, as the author than it would have been if the employer
were a natural person.
Fifth, the doctrinal change could occur because judges revamped
their thinking about writing, authorship, and the production of works
subject to copyright, and changes in the understanding of the nature
of ideas made that revamping easier. Pragmatist thought reflected a
new approach to ideas which made a new approach to owning them
seem more plausible: Ideas are created socially; they are not "out
there" waiting to be uncovered, and they are not the product of one
individual's solitary genius.' They are produced because of the

24. Three of the earliest cases involved "independent contractors" who, the courts held,
owned their copyrights. Wheaton v. Peters, 33 U.S. 591 (1834); Little v. Gould, 15 F. Cas. 612
(C.C.N.D.N.Y 1852); Atwill v. Ferrett, 2 F. Cas. 195 (C.C.S.D.N.Y.1846). One early case
involved an "independent contractor" who did not own his copyright. Heine v. Appleton, 11 F.
Cas. 1031 (C.C.S.D.N.Y. 1857). In two cases from the middle period, the employment status of
the author was ambiguous, and he might today be characterized as employee, contractor, or
even possibly partner. Boucicault v. Fox, 3 F. Cas. 977 (C.C.S.D.N.Y. 1862); and Roberts v.
Myers, 20 F. Cas. 898 (C.C.D. Mass. 1860). In another middle period case, the author was
probably an employee, and he was held not to own the rights to his work. Keene v. Wheatley,
14 F. Cas. 180 (C.C. E.D. Pa. 1861). In another case, the author was a contractor, and the court
held the parties had expressly contracted for ownership by the employed person. Lawrence v.
Dana, 15 F. 26 (C.C.D. Mass. 1869). In two decisions from late in the middle period, the
authors were salaried employees, and the courts held that they rather than their employers
owned the copyrights. Callaghan v. Myers, 128 U.S. 617 (1888); Peters v. Borst, 9 N.Y.S. 789
(Sup. Ct. 1889). The earliest case discussing the distinction between employees and
independent contractors as relevant to ownership of literary property was Edward Thompson
Co. v. Clark, 109 N.Y.S. 700, 702 (Sup. Ct. 1904), which held that because the author was an
employee rather than an independent contractor, the employer owned all written work and
could prevent the employee from publishing it elsewhere even though the employer had
rejected the work as unpublishable.
25. My thinking about the cultural changes associated with the early development of
pragmatism in the late nineteenth century has been heavily influenced by LOUIS MENAND,
THE METAPHYSICAL CLUB (2001). On pragmatism and American law, especially the
 Yale Journal of Law & the Humanities [Vol. 15:1

individual's past experience and social context. If they are created

collectively, and if exposure to a particular environment is crucial,
then they are neither the exclusive property of the person who wrote
them nor the brat of the solitary writer's brain. Moreover, ideas are
produced because they are useful. If, as Holmes believed, all thought
"is at once 'social' and 'on its way to action,' 26 then both the
generation and the practical use of ideas in the workplace are
inherently part of the business enterprise. The belief that knowledge
or an idea is a tool like any other makes it easier to see that ideas
and their embodiment in texts are the product of the group setting in
which employees work and which the employer arguably has
purchased through wages and the investment in tools and materials.
These five interrelated causal claims are woven throughout this
article, but the organization follows the trajectory of the descriptive
claim about the evolution of the doctrine. Because the development
of the doctrine can be divided into three periods, the article, which is
based on every reported American decision involving an employed
author before 1910, is too. Part I covers the period before 1860, in
which courts quite clearly asserted that persons hired to create works
owned the copyright to their works, not the person who hired the
creator. Copyright registrations suggest that the actual practices of
employers and firms may have recognized a default rule of employer
ownership of copyrights, but the case law did not. So few copyright
cases were litigated, however, that it is difficult to know whether the
courts' and the parties' views of the default rule were the same or
different. The paucity of reported cases and the tremendous
influence of the few that exist both enable and require me to explore
them in depth. I aim to shed light both on why courts adopted the
rule and on whether the reported decisions were or were not
representative of the way other writers and employers would have
understood their rights. Part II treats the period of transition
between 1860 and 1900 when courts began to recognize the interests
of employers in copyright ownership. Several courts stated a
principle of employer ownership in dicta, but the holdings still found
that only an express contract was sufficient to locate the copyright in
the employer. During this middle period, the rules regarding
intellectual property ownership were uncertain and contradictory,
thus belying the conventional assumption that certain and definite

pragmatism of Holmes, see Thomas C. Grey, Holmes and Legal Pragmatism, 41 STAN. L. REV.
787 (1989). The literature on law and pragmatism is vast, and a full citation to the literature
will have to await a full discussion of the influence of pragmatism on the development of the
legal ownership of ideas. For better or worse, this discussion will not be found in this article.
26. Here I quote Thomas Grey's characterization of Holmes's own summary of the
essential tenets of pragmatism. Grey, supra note 25, at 798.
2003] Fisk

property rules were necessary to and a crucial factor in the rapid

economic expansion of the late nineteenth century. Part III treats the
beginning of the modern era, in which courts recognized and
Congress codified that the fact of employment was sufficient to vest
copyright in the employer, even in the absence of a contract
expressly allocating ownership.27

II. THE EARLY PERIOD: BEFORE 1860

During the first half of the nineteenth century, the hegemony of
author as owner ensured that courts saw no employer claim to the
creative products of their employees. Published cases determined
that employees were entitled to the copyrights in their works, even
though the works had been done during the course of employment
or on commission. As will be noted, however, data on copyright
registrations show that employers did sometimes copyright work
prepared by their employees. The practices in some industries-I use
the example of mapmaking--suggest that where authorship was
communal, employers claimed copyrights. Following a brief look at
copyright and collective authorship before 1830, this section
examines the only published decisions on copyright ownership
involving employed authors, which involved law publishing and
theater. Then, by way of comparison, I examine copyright practices
in the area of map publishing. Although the example of map
publishing suggests that actual practices of copyright ownership were
complicated, the rules articulated by courts in cases where employers
and employees disputed copyright ownership were not. Courts were
quite clear that, absent an express and clear contrary agreement, the
employee-author was entitled to the copyright.

27. The doctrinal change that occurred in these three periods loosely corresponds to
similar changes that occurred at the same time in the development of other branches of
intellectual property and employment law. See Catherine L. Fisk, Removing the "Fuelof
Interest"from the "Fireof Genius": Law and the Employee Inventor, 1830-1930,65 U. CHI. L.
REV. 1127 (1998) [hereinafter Fisk, Law and the Employee Inventor]; Catherine L. Fisk,
Working Knowledge: Trade Secrets, Restrictive Covenants in Employment, and the Rise of
Corporate Intellectual Property. 1800-1920, 52 HASTINGS L.J. 441 (2001) [hereinafter Fisk,
Working Knowledge]. Similar reasons for the change explain the law of employee patents,
trade secrets, and cognate doctrines. The problem of reconciling corporate authorship with
individual creation was more acute in the field of copyright than it was when the creative
products were patented or with trade secrets because the courts struggled more consciously
with the question of whether the kinds of materials that firms would create were sufficiently
original, creative, or even artistic to be worthy of copyright protection. Unlike in patent law,
where the only requirements were nonobviousness, novelty, and usefulness, courts seemed to
think that copyright required some modicum of artistic expression or creativity. Thus, when
courts came to justifying corporate authorship, they had to analogize advertisements to oil
paintings and corporations to great artists.
 Yale Journal of Law & the Humanities [Vol. 15:1

A. Copyright and Collective AuthorshipBefore 1830

For centuries before my study begins, people had been employed
to create texts and other works that today would qualify for
copyright protection. Literary histories of copyright and authorship
note that prior to the Renaissance the collective nature of the
creation of texts was widespread. If texts were divinely inspired,
original creation was not understood to be the essential characteristic
of authorship. And the reproduction of texts involved copying rather
than conjuring. The nature of book production was thus inherently
collective. 8
Evolving notions of authorship, together with Lockean labor
theories of value, gradually created the ideas that the author is the
one who does the original and creative work of imagining and
writing, and that property rights are justified by the labor expended
in the creation.29 That understanding of the nature of authorship
appeared in Blackstone's Commentaries on the Laws of England.
Blackstone, of course, was a contemporary of the eighteenth-century
English advocates of authors' rights, and he was an author as well.
Blackstone famously stated the classic mixed-labor and genius-based
justification for copyright:
When a man by the exertion of his rational powers has
produced an original work, he has clearly a right to dispose of
that identical work as he pleases, and any attempt to take it
from him, or vary the disposition he has made of it, is an
invasion of his right of property. °
But as the labor theory of property melded with the Romantic
notion of authorship, no attention was paid to the question of how to
allocate property rights when the labor was that of an employee.

28. One description of the process reveals the similarity between book production and
wage labor in the pre-Renaissance period:
With the so-called "book revolution of the twelfth century and university supervision of
copying, there came a 'putting out' system. Copyists were no longer assembled in a single
room, but worked on different portions of a given text, receiving payment from the
stationer for each piece .... Through a sort of guilt by association, we can see how wage
labor might metonymically become associated with the capacity to possess the
knowledge which is the source and aim of the labor. We can also envision how the
"employer" of copyists might become a metonym for the owner of the knowledge and
thus of the authority supposedly contained within the text.
Marion B. Ross, Authority and Authenticity Scribbling Authors and the Genius of Print in
Eighteenth-Century England, in THE CONSTRUCTION OF AUTHORSHIP: TEXTUAL
APPROPRIATION IN LAW AND LITERATURE 231, supra note 2 (quoting I ELIZABETH
EISENSTEIN, THE PRINTING PRESS AS AN AGENT OF CHANGE 12-13 (1979)).
29. See generally THE CONSTRUCTION OF AUTHORSHIP: TEXTUAL APPROPRIATION IN
LAW AND LITERATURE, supra note 2; ROSE, supra note 3.
30. 2 WILLIAM BLACKSTONE, COMMENTARIES ON THE LAWS OF ENGLAND *405-06.
Blackstone had, as a lawyer, represented one of the parties in a prominent English case on the
scope of copyright. See ROSE, supranote 3, at 75.
2003] Fisk

The modern notion of the author as the individual creator of text

ex nihilo was a political and legal strategy employed by publishers to
galvanize legislative and judicial support for monopoly rights over
books. As Mark Rose explained in his seminal work, Authors and
Owners,in the seventeenth century,
[t]here may have been some feeling that authors should have
the right to control the first publication of their writings. But in
England at any rate no clearly defined set of authorial rights
existed .... Indeed, the very concept of "author" was still
incompletely developed. Not only was the modern notion of the
author as an autonomous creator, the producer and first
proprietor of original works, not yet formed, but even the
Renaissance notion of the author as an individuated authority
was often problematic.3'
Rose argued that the conceptual foundations for the author's
copyright were established in British law in the eighteenth century.
Locke's theory of property as arising from the mixing of the
individual's labor with nature united easily with an emerging
aesthetic of books as products of a creative mind and as
manifestations of an individual's personality." The author as creative
genius and the author as proprietor simultaneously described the
nature of literature and justified the existence of legal rights in
ideas.33 The author-proprietor proved to be a very durable synthesis
that persists in American copyright law today.
Rose and others have shown how writers from Defoe to Dickens
joined forces with publishers in advocating the moral, cultural, and
economic benefits of protecting the authors' proprietary rights in
their works. 3 From the eighteenth century through the twentieth,
celebrity authors have been effective lobbyists for a set of copyright
rules that benefited corporate publishers as much as individual
scribblers.35 The early approach to employee authors is no exception.
One of the authors who figured prominently in the nineteenth-
century cases involving theater employees, Dion Boucicault,

31. ROSE, supra note 3, at 25.

32. Id. at 114.
33. Id. at 6.
34. See generallyid. and sources cited supra notes 3 and 4.
35. The celebrity author copyright advocates include Daniel Defoe, Alexander Pope, and
William Wordsworth. See ROSE, supra note 3, at 34-37, 60-65, 110-12; Dion Boucicault, see
infra text accompanying notes 105-115. A modern example of the successful melding of
celebrity with copyright advocacy is the Sonny Bono Copyright Extension Act, Pub. L. No.
105-298, 11 Stat 2827 (codified in scattered sections of 17 U.S.C.) named for the late
entertainer by his widow and other legislative sponsors whose Hollywood constituents perhaps
care more that Mickey Mouse never enters the public domain than the music and lyrics to I
Got You, Babe.
 Yale Journal of Law & the Humanities [Vol. 15:1

successfully lobbied for a revision of American copyright law to

provide that the copyright in a play allowed the author to restrict
performances as well as publication.36
Telescoped summaries of the nineteenth-century ethos of
copyright posit an unassailable hegemony of the paradigm of the
individual author. The dominant ideology of copyright in both law
and literature is the position forcefully advocated by Dickens and
Wordsworth and usually described as "Romantic Authorship." That
is, copyright exists to reward the genius of the individual author who
creates great literature out of his own imagination and hard work
and little else. This vision of authorship is part of the nineteenth-
century Romanticism that lionized individuality, creativity, and the
unique genius that fired both. But the Romantic Author may not
have been quite as hegemonic even in the nineteenth century as is
sometimes suggested. Paul Saint-Amour's study of nineteenth-
century plagiarism apologists reveals that even in the hey-day of
Romanticism, there were many who advocated a broader role for the
public domain and for a continued appreciation of collective
creation.37 Depending on how broadly Saint-Amour's thesis applies,
the notion that employers might claim to own the creative works of
their employees was not as revolutionary as one might think.3"

36. See infra notes 73-75 and 101-118.

37. See SAINT-AMOUR, supranote 3, at 124-63.
38. The dichotomy between authors' rights and corporate rights in intellectual property is
not a uniquely Anglo-American phenomenon and has therefore received attention from
continental European courts and scholars as well. In France and Germany, in contrast to
Britain and the United States, the conceptual starting point for protecting intellectual property
rights in writings is the inalienable and personal right of the natural person who is the actual
creator of the work. In the common law countries, the starting point is protection of the value
of the work. Given the focus on authors' rights in civil law regimes, the development of an
allocation rule of employer ownership would be particularly vexing. In Germany and France,
the law limits the ability of legal entities to be copyright owners more strictly than the law in
the United States and the United Kingdom. For the most part, the principle of employer
ownership seems to have been achieved through notions of implied contract or implied
assignment. But what is significant for present purposes is that scholarship on civil law authors'
rights regimes argues that "the present-day opposed approaches of the civil law and common
law systems on the question of corporate ownership of rights are of a comparatively recent
date and not rooted in the origins of the two systems." Gillian Davies, The Convergence of
Copyright and Authors' Rights-Reality or Chimera?,26 INT'L REV. OF INDUS. PROP. AND
COPYRIGHT L. 964, 972 (1995); see also Thomas K. Dreier, Authorship and New Technologies
from the Viewpoint of Civil Law Traditions,26 INT'L REV. OF INDUS. PROP. AND COPYRIGHT
L 989, 989-94 (1995).
The concept of the individual author's inalienable moral rights developed in France and
Germany in the nineteenth century, just at the same time that corporate ownership of
copyrights developed in the United States. See GILLIAN DAVIES, COPYRIGHT AND THE
PUBLIC INTEREST chs. 6-7 (1994). More significantly, there have always been counter examples
in both regimes-of employer ownership in the United States even in the antebellum period
and of legal entity ownership in Germany and France.
In Britain, the law was relatively similar, including the rise of contract concepts in the mid-
nineteenth century and the vagueness about what kinds of circumstances would give an
2003] Fisk

It is important to remember that what courts said the default rules

were and what employers thought they were may have differed. As
Zorina Khan has demonstrated, between forty-four and forty-nine
percent of copyright registrations between 1790 and 1800 were by a
person other than the actual author. 9 (At that time, one had to
register in order to obtain a copyright; today, registration is no
longer required.) A significant number of registrations were for
works other than the kinds of books that typically would have an
individual author; they included maps and charts, dictionaries, and
directories.' ° It is reasonable to suppose that among these, some
employers were registering the copyright to works that persons in
their employ had created, at least in part. It is unclear whether all
those registrations were pursuant to an express assignment by the
employed author. Absent an express assignment, however, the scant
case law suggests employers might not have been able to obtain a
copyright for a work prepared by an employee. In Binns v.
Woodruff, Justice Washington, sitting as a circuit judge, held that
one who employed others to create an elaborately decorated print of
the Declaration of Independence could not obtain a copyright
because he had neither designed, drawn, nor engraved the work and
thus was not an author as required by the statute.
In the face of a dearth of legal or business archival evidence of
how firms handled employee copyrights, two somewhat
contradictory facts appear. First, compared with today, relatively few
kinds of works were subject to copyright (the first Copyright Act of
1790 provided for copyright only in maps, charts, and books) 42 and
there were relatively few copyright registrations in the antebellum

"implied" contractual right to the employer. EATON S. DRONE, A TREATISE ON THE LAW OF
PROPERTY IN INTELLECTUAL PRODUCTIONS IN GREAT BRITAIN AND THE UNITED STATES
EMBRACING COPYRIGHT IN WORKS OF LITERATURE AND ART, AND PLAYRIGHT IN
DRAMATIC AND MUSICAL COMPOSITIONS 243-53 (Boston, Little Brown, 1879). 1 discuss
Drone's treatise infra beginning at note 162. But in countries with a stronger emphasis on
authors' rights, the story seems to have been somewhat different. In France and Germany, to
varying degrees, employee-authors were entitled to the copyright, and employers could obtain
the copyright only by express or implied assignment. As the moral rights concept developed in
both countries over the course of the nineteenth century, the employee's rights increased,for
moral rights are inalienable. Yet, at the same time, courts in France, Britain, and Germany
employed similar implied contract concepts to transfer, as a matter of law, or at least without
any express agreementbetween employer and employee, whatever rights could be transferred
from the author to another. See generallyDavies,supra, at chs. 2-3.
39. KHAN, supra note 7, at ch. 8, tbl. 1.
40. Id. In the period from 1790 to 1800, for example, 17% of the registrations were for
dictionaries, directories, atlases, maps, and works of "commerce." Among books that likely
had individual authors, 14% were for what Khan calls "social and political" books and another
17% were for textbooks. Eleven percent were for music, poetry, and plays, and 5.7% were for
law books. Id
41. Binns v. Woodruff, 3 F. Cas. 421 (C.C.D. Pa. 1821) (No. 1424).
42. Act of May 31, 1790, ch. 15, 1 Stat. 124.
 Yale Journal of Law & the Humanities [Vol. 15:1

period. Second, copyright played a relatively minor role even in

those few employment settings in which people were paid by others
to create material subject to copyright protection. Although the case
law did not recognize an employer's right to employee works until
the late nineteenth century, persons not the authors frequently
registered copyrights, and it is reasonable to infer that they were
registering the copyright in works created by their employees. Thus,
employers may have contracted around the default rule or may
simply have assumed that the default was precisely the opposite of
what it was.43

B. Authors, Owners, and,American Law Publishing Wheaton v.

Peters and the EarlyRise of Employee A uthorship
One of the most pressing needs of the early American legal
community was to develop an adequate system for the reporting and
publication of the new American law. It turned out to be a struggle
to find enough competent people to report the decisions of the
various state and federal courts. Even more difficult was finding
printers willing to print the reporters' work at a price that lawyers
would pay and to develop a plan for distributing the books once
printed. A system, if one can call it that, emerged whereby lawyers
interested in the work of the court obtained some sort of
appointment from the judges (and, in some states, the legislature) as
the official reporter of the court, and it was up to the lawyer to find a
printer who would print the reports at a price and quantity to fulfill
the reporter's obligation to the court and to enable the reporter to
sell enough to make a profit to support himself. The court reporters
in these years were not exactly employees of the court; neither the
court nor the government paid a salary. They were independent
contractors in the modern parlance, although in some jurisdictions
they were also office holders.'

43. Doctrine and practice were not necessarily inconsistent, in the sense that courts
acknowledged the legitimacy of express contracts assigning the copyright to the employer.
However, without knowing whether early copyright registrations by employers were pursuant
to an express agreement, it is difficult to know whether the court's description of the default
rule was consistent with the practice of employer registrations or not.
44. On early American law publishing, see EDWIN C. SURRENCY: A HISTORY OF
AMERICAN LAW PUBLISHING (1990). On the law reports and law reporters of particular courts
and states, see Ann Fidler, "Till You Understand Them in Their PrincipalFeatures"
Observations on Form and Function in Nineteenth-Century American Law Books, 92
BIBLIOGRAPHY SOC'Y AM. 427 (1998); Joel Fishman, The Digests of Pennsylvania, 90 LAW
LIBR. J. 481 (1998); Joel Fishman, The Reports of the Supreme Court of Pennsylvania,87 LAW
LIBR. J. 643 (1995); Craig Joyce, The Rise of the Supreme Court Reporter. An Institutional
Perspective on MarshallCourt Ascendancy, 83 MICH. L. REV. 1291 (1985); John H. Langbein,
ChancellorKent and the History of Legal Literature,93 COLUM. L. REV. 547 (1993); Kurt X.
Metzmeier, Blazing Trails in a New Kentucky Wilderness: Early Kentucky Case Law Digests,
20031 Fisk

The first American copyright case in which the author was hired to
write was the famous dispute between the third and fourth reporters
for the U.S. Supreme Court, Henry Wheaton and Richard Peters. 5
More accurately, Wheaton v. Peters was a dispute between the
publishers of the reporters, for just as in the famous eighteenth-
century English copyright disputes that established the author's
copyright, Wheaton the author had assigned his copyright to his
publisher, and it was the publisher who litigated the suit. Wheaton v.
Peters was also the Supreme Court's first intellectual property
decision.46 Although it is famous for its rejection of the notion that
courts could confer common law copyrights in circumstances where
the statutory requirements were not met, a less well-known aspect of
the opinion was its assumption that reporters of judicial decisions
were entitled to copyright their work.
Henry Wheaton was the reporter of decisions of the Supreme
Court from 1816 to 1827, the third in the Court's history.47 After
Richard Peters became the reporter in 1827, he published and sold a
volume called CondensedReports of Cases in the Supreme Court of
the United States containing all the Court's decisions from its
founding until 1827, including Wheaton's. Wheaton sued Peters for
copyright infringement, and Peters argued that Wheaton (actually,
his publisher) had not complied with the statutory procedures for
registering a copyright, and that the copyright was therefore void.48
The Court remanded for factual findings on the adequacy of
Wheaton's efforts to register the copyright.49 The Court's opinion
assumed that if his efforts were sufficient, he would have a valid
copyright, notwithstanding his employment or appointment as
reporter for the Court." His status-whether employee or appointed
officer of the Court or of the United States--was, in the Court's
view, irrelevant to his copyright claim. Nor did the Court mention
whether Wheaton's claim to the copyright was based on a contract,
on the Reporters Act of 1817 (which said nothing about copyright),51

93 LAw LIBR.J. 1 (2001); and Edwin C. Surrency, Law Reports in the UnitedStates,25 AM. J.
LEGISLATIVE HIST. 48 (1981).
45. Wheaton v. Peters, 33 U.S. 591 (1834).
46. My account of Wheaton v. Petersdraws heavily on EDWARD WHITE, THE MARSHALL
COURT AND CULTURAL CHANGE, 1815-1835, at ch. 6 (1988) and Joyce, supranote 44.
47. Alexander Dallas and William Cranch, Wheaton's predecessors, reported the Court's
decision before 1816. After serving, like them, in an uncompensated and unofficial capacity
beginning in 1816, Wheaton became the Court's first official reporter in 1817 after Congress
enacted a statute providing for an official reporter of the Supreme Court's decisions. See
WHITE, supra note 46, at 389-90; Joyce, supra note 44, at 1343.
48. 33 U.S. at 634-35.
49. Id at 667.
50. Id. at 668.
51. See infra note 54.
 Yale Journal of Law & the Humanities [Vol. 15:1

or on other terms of his appointment as the reporter. His right to the

copyright was simply assumed, provided he had complied with the
statutory registration procedures.
Wheaton had himself assumed he owned the copyright in his
reports.5 2 The first Term during which Wheaton was the reporter
ended in March 1816, and by early May, Wheaton had the volume
ready to publish. But he could find no publisher willing to print it.
He finally was forced to give the copyright to the Philadelphia
bookseller Matthew Carey in order to persuade Carey to print it, and
Carey paid only $1200 in notes for all the rights. 3 Wheaton was not
unreasonable in thinking he owned the copyright to the volume. In
the first place, Congress had resisted his requests to vote him a
formal title and salary as reporter,54 and, in the second place, his first
volume contained 487 pages of abstracts, arguments, and opinions
and forty-six pages of notes; the abstracts, arguments, and notes
were, of course, largely Wheaton's writing rather than the Justices'. 5
Yet in another sense, his employment status had the potential,
realized in later cases involving reporters of decisions,56 both to
strengthen and to weaken his claim to the copyright. In particular,
the Court stated "no reporter has or can have any copyright in the
written opinions delivered by this court; and.., the judges thereof
cannot confer on any reporter any such right."57 Wheaton could
copyright whatever he added: headnotes, a summary of the decision,
arguments of counsel, pagination, and perhaps even his report of
opinions rendered orally. He simply could not copyright the Justices'
own work.
As to the parts of the reports that were not opinions, the Court
surely had no doubt that Wheaton's work was "authorship" in the
sense that would ordinarily entitle him to copyright protection.
Wheaton's industry and erudition in writing the marginal notes and
the appendices on admiralty and prize law were plain to all. The

52. So, indeed, did the Justices and both the members of Congress who supported the
Reporters Act of 1817 and those who opposed it on the ground that it was unnecessary to pay
a salary to the Reporter. See WHITE, supranote 46, at 389-90.
53. Joyce, supra note 44, at 1326.
54. Id at 1325. The Reporters' Act of Mar. 3, 1817, ch. 63, 3 Stat. 376 provided a salary of
$1000 per annum, but Wheaton had been employed before that to prepare the first volume at
no salary. As to every year after 1817, Wheaton argued that the $1000 was "unequal to the
labour and time" and was, in any event, compensation for the eighty volumes that the statute
obliged him to deliver to the government. 33 U.S. at 614.
55. Joyce, supra note 44, at 1327. In fact, some were anonymously written by Justice Story,
who gave them to Wheaton to assist him in publishing the Reports. Id; WHITE, supra note 46,
at 391, 393, 402.
56. See infra text accompanying notes 64-67.
57. Wheaton, 33 U.S. at 668.
58. See WHITE, supra note 46, at 390-91.
2003] Fisk

Court recognized these as his valuable work. And, inasmuch as he

was not being compensated for writing them, it would be
inconceivable to say that the copyright was not his. Besides sympathy
for him, there would also be the instrumental concerns-if he could
not sell the copyright to the printer, there would be no printer willing
to print, and the opinions would never be disseminated.59 It might
also have mattered, as the Justices no doubt knew, that the very
limited market for law books made the risk of copying seem quite
remote. 6 And finally, there would be the matter of the Justices'
respect and affection for Wheaton. He was a personal friend of
Story's; indeed, they were roommates at the Washington boarding
house where all the Justices stayed during the Term.6'
It was obvious why Wheaton could not copyright the Justices'
opinions without their consent, but the Court went further: The
Justices could not confer a copyright on the reporter. 2 The Court did
not explain why, although later courts explained that judges could
not copyright their written opinions because they were public
servants.63 But so were reporters. What was the difference? Two
rationales are possible, though neither was offered in the Court's
opinion in Wheaton v. Peters.One was that law reports, like judges'
opinions, could not be copyrighted because they were either not a
proper subject of property or should be the "property" of all. In this

59. Wheaton v. Peters seems an apt example of why the initial allocation of intellectual
property rights as between employer and employee can matter. First of all, the early
nineteenth-century judicial mind did not equate allocation of "authorship" as the legal
equivalent of allocating rights in Blackacre. Second, the Justices had institutional reasons to
insist that Wheaton was the "author" and therefore the initial copyright owner of his Reports.
The Justices thought it mattered a great deal whether they, the people of the United States, or
Wheaton had the initial entitlement to copyright, for only Wheaton was in a position to
negotiate with printers like Matthew Carey to publish and distribute the Reports.
60. See WHITE, supra note 46, at 389.
61. See Joyce, supra note 44, at 1313, 1322. On the effect of the boarding house residence
on Marshall Court practice, see WHITE, supra note 46, at 160-61, 190-91.
62. 33 U.S. at 668.
63. Today, there can be no copyright at all in works produced by U.S. government
employees. 17 U.S.C. §§ 101, 105. Section 105 provides that copyright protection "is not
available for any work of the United States Government," and § 101 defines such works to
include "a work prepared by an officer or employee of the United States Government as part
of that person's official duties." The legislative history of the 1976 revision of § 105 states that
the concept should be construed "in the same way" as works made for hire. H.R. REP. No. 94-
1476, at 59 (1976). The purpose of § 105, according to the legislative history of the 1976 Act,
was "to place all work of the United States Government, published or unpublished, in the
public domain." Id. The predecessor to § 105 first appeared in the Printing Act of 1895, ch. 23,
28 Stat. 601 (1895), and was recodified and retained unchanged in the 1909 Copyright Act. Act
of Mar. 4, 1909, ch. 320, 35 Stat. 1075. See H.R. Rep. No. 60-1 (to accompany house bill H.R.
28192, Feb. 1909). Even prior to 1895, there was some recognition that copyrighting federal
government materials was problematic. See Robert M. Gellman, Twin Evils: Government
Copyrightand Copyright-Like ControlsOver Government Information,45 SYRACUSE L. REV.
999 (1995).
 Yale Journal of Law & the Humanities [Vol. 15:1

view, the nature of the written product as law makes it unfit for
copyright-the law must be in the public domain so that it is free for
all to use, quote, reproduce, and disseminate widely.
A second rationale might be that judges could not obtain copyright
and thus could not assign it to reporters because they were public
servants and their "employer," the U.S. government, could be the
only proper owner of the copyright. As public servants, the judges
could not acquire the copyright in the work they were employed to
do--the preparation of opinions-and thus had no rights to confer
upon the reporter. If that were the rule the Court adopted, then it
might mean more generally that employment to create copyrighted
works should be deemed to constitute an assignment of the copyright
to the employer.
The one case that did adopt such a rule was based both on an
express contract and on a statute providing that the reporter of
decisions of the New York Court of Appeals would not obtain a
copyright to his notes, to his reports, or to the references he
compiled. The view that judges, as public servants, could have no
copyright in their opinions to assign to the reporter because the
judges' employer-4he state government-should own the copyright
was thus not a default rule in the case.' Indeed, the underlying
contract between the reporter and the New York Secretary of State
contained an assignment of the copyright to the Secretary of State,
which suggests that both the reporter and the Secretary of State may
have believed that the reporter would, absent the contract, have a
copyright to assign.
No nineteenth-century case read Wheaton v. Peters to hold that
judges, as "employees" of the government, could not confer a
copyright on the reporter because the copyright was not the judges'
to confer. A few cases adopted at most a hybrid, but did so only
much later. For example, in Banks v. Manchester, the Supreme
Court held that the reporter of Ohio could not enjoin the copying of
the official reports:
In no proper sense can the judge who, in his judicial capacity,
prepares the opinion or decision, the statement of the case, and
the syllabus, or head-note, be regarded as their author or their
proprietor.., so as to be able to confer any title by assignment
on the states, sufficient to authorize it to take a copyright for
such matter .... Judges, as is well understood,receive from the
public treasury a stated annual salary, fixed by law, and can

64. Little v. Gould, 15 F. Cas. 612, 613 (C.C.N.D.N.Y. 1852) (No. 8395). Other phases of
the litigation over the copyright to the reports were published as well. Little v. Hall, 59 U.S.
165 (1855); Little v. Gould, 15 F. Cas. 604 (C.C.N.D.N.Y. 1851) (No. 8394).
2003] Fisk

themselves have no pecuniary interest or proprietorship, as

65
againstthe publicat large,in the fruits of theirjudiciallabors.
Although the second, italicized part of the passage suggests
employment as a public servant rendered one unable to obtain a
copyright, the first part of the passage suggests that even the
government cannot obtain one. It suggests that opinions prepared by
judges are essentially and inherently common property and can be
copied by anyone, a rule that was codified as regards U.S.
government documents in the Printers Act of 1895.' Moreover, by
1888, courts had begun to see and enforce employees' agreements to
assign copyrights. From there it was not a large step to conclude that
the contract of employment itself might in particular cases operate as
an assignment agreement.
The predominance of disputes involving reporters of appellate
decisions in the early cases played a significant role in shaping the
law. Although judges no doubt understood them as public servants,
they were not the judges' own servants, nor were they "servants" in
the conventional sense. They were men of stature and education,
trained in the law, whose diligence, analytic skills, and attention to
detail were crucial in the dissemination of the law. Their work was
also essential to the edification of the bench and bar. They were far
too important for judges to see them as servants or to compromise
their independence by restricting their claim to copyright. In many
cases, their relations with their "employer"--the court or the
government--more closely resembled what today would be deemed
an independent contractor relationship. In the early nineteenth
century, master-servant law applied only to masters and servants, not
to every person who worked for someone else, and there was no
unified body of law that applied to all who worked for others.67 By
the late nineteenth century, courts began to apply master-servant
principles to all kinds of employed persons. They thus effaced the
distinctions of status and independence that previously had accorded
different rights to different kinds of persons and replaced them with
a uniform body of law that translated rights and obligations
previously determined as a matter of status into implied contractual
terms. But since the early employee copyright cases had involved
persons who would not be governed by the hierarchical rules of

65. 128 U.S. 244, 253 (1888) (emphasis added).

66. See Gellman, supra note 63, at 1002-05,1023-24 & n.102.
67. See DAVID GIBBONS, RUDIMENTARY TREATISE ON THE LAW OF CONTRACTS FOR
WORKS AND SERVICES 54, 109 (London, John Weale 1857) (contrasting obligations owed by
contractors and those owed by servants); HORACE GAY WOOD, A TREATISE ON THE LAW OF
MASTER AND SERVANT 2-3 (Albany, J.D. Parsons, Jr. 1877). See generallyCHRISTOPHER L.
TOMLINS, LAW, LABOR, AND IDEOLOGY IN THE EARLY AMERICAN REPUBLIC (1993).
 Yale Journal of Law & the Humanities [Vol. 15:1

master and servant, the courts did not apply the assumptions of
master and servant law in deciding the respective rights of the hired
person and the hiring party. As a result, the relatively pro-employee
rules became settled doctrine, only to be changed at the beginning of
the twentieth century once the distinction between employees and
independent contractors emerged.
Judges probably also understood more clearly in the case of
reporters than in the case of almost any other occupation how crucial
allocation of copyright ownership was to the economics of law
publishing. Granting the reporter the copyright was in lieu of paying
a more generous compensation from the public fisc. In an era in
which government was quite limited in its ability to raise revenue to
fund a judicial bureaucracy, judges doubtless were reluctant to
interfere with the scheme by which reporters could be compensated
from sources other than a full salary paid from the treasury. Perhaps
judges understood that the willingness of competent, educated men
to take on the task depended on their ability to make a profit. The
speedy publication and distribution of reports necessary to the law's
development may also have benefited from the reporter's industry
and been spurred on by the profit motive.
Although law publishing cases dominated the early development
of employee copyright law, one other decision from this era took a
similar position. Judicial solicitude for the rights of employee authors
was thus not entirely the result of concerns about law book
publishing. Pierpont v. Fowler, a case between an author who had
been commissioned to write school books and the person who hired
him, presented the question of whether the author or the employer
had the right to obtain the renewal term.' The court concluded that
the renewal rights did not pass with the contract assigning the initial
copyright to the employer.69 In explaining the result, the court
emphasized the importance of authors' rights:
It was the genius which conceived and the toil which compiled
the book that is to be rewarded by even the first copyright, and
no one ever dreamed that an assignee could alone take out the
second or extended term, unless he has paid for it, clearly
contracted for it .... [I]f such a hirer of others was entitled
before to take out a copyright, how does this act encourage and
aid genius? It rather aids those kinds of patrons, who fatten on
the labors of genius."

68. 19. F. Cas. 652, 14 Copyright Dec. 2075 (C.C.D.Mass 1846).

69. 14 Copyright Dec. at 2088-89.
70. 14 Copyright Dec. at 2087-88.
2003] Fisk

The court appeared to be more concerned with protecting authors

from bad bargains than with facilitating the development of the
publishing business.7'

C. Commissioned Works and the RelationshipBetween Playwrights

and Theaters
If the first cases involving law publishing are the antecedents of the
branch of work-for-hire doctrine involving employees, the other
branch of modern work-for-hire doctrine--commissioned works-
had its origins in a different line of cases that began in this early
period as well. These were the cases involving playwrights and
composers commissioned by theater managers or owners to write a
play or adapt or compose a musical score for performance at the
theater. During this period, the playwright or composer invariably
acquired the copyright unless the parties agreed otherwise in writing.
There is an interesting irony in the courts' insistence that the
playwright owned the copyright rather than the theater owner, for in
the 1840s and 1850s the playwright had yet to obtain the status of
genius artist that people later accorded to Ibsen, Shaw, or O'Neill.
Theater in the 1840s was hardly considered art. An evening's bill
might last four to six hours and include a main piece, a curtain-
raising farce, and a closing burletta. The audience wanted to see
comedy, melodrama, and extravaganzas with spectacular effects.72 In
that context, the playwright was a sort of "handyman to the
company. He existed to make their performance possible, rather
than they to interpret his work to an audience." 3
Yet this was beginning to change around 1860, and one of the
moving forces in that transformation was both an important litigant
and one of the most prominent playwrights, actors, and theater
managers of the era: Dion Boucicault. What is particularly
interesting is that the ambiguous roles of playwright, actor, and
theater manager in that era were reflected in the ambiguous
approaches of the courts to the allocation of intellectual property

71. The restriction of the assignment of renewal rights, so as to protect authors from
improvident bargains made when the value of a copyrighted work was unknown, was a
pervasive concern in copyright law into the twentieth century. See infra note 224. Thus, one
ought not read the case as being just about employment, but also about the concern for
protecting all authors--employees or not.
72. On the evolving nature of melodrama in the mid-nineteenth century, see BRUCE A.
MCCONACHIE, MELODRAMATIC FORMATIONS: AMERICAN THEATRE AND SOCIETY, 1820-
1870 xii-xiii, 242-43 (1992). On Dion Boucicault's use of sensational melodrama, see James
Hurt, Dion Boucicault's Comic Myths, in WHEN THEY WEREN'T DOING SHAKESPEARE:
ESSAYS ON NINETEENTH-CENTURY BRITISH AND AMERICAN THEATRE 253 (Judith L. Fisher
& Stephen Watt eds., 1989).
73. ROBERT HOGAN, DION BOUCICAULT 30 (1969) (quoting GEORGE ROWELL, THE
VICTORIAN THEATRE, A SURVEY 1 (1956)).
 Yale Journal of Law & the Humanities [Vol. 15:1

rights in plays. Though the playwright may have worked for the
company or the theater, some playwrights such as Boucicault were
beginning to enjoy a certain celebrity status and to assert their
entitlement to copyright their works (even when they plagiarized
from others and were paid to do so by a theater). If, as in the case of
Laura Keene discussed below, the playwright, the actor, and the
theater manager were one and the same, and the competing claimant
was an actor in her company, the courts saw the entitlement to
copyright very differently.
The earliest case involving a commissioned work was A twill v.
Ferrett,a theater case which stated a general rule that the writer of
an opera was the author, even if the work was commissioned by
another. 4 The opinion also stated, however, that since the theater
manager who had commissioned the work had been so involved in
modifying the opera, he was entitled to a copyright in the version
performed in his theater. The plaintiff, Atwill, sought to produce in
New York an opera, The Bohemian Girl, that had been composed
and performed in Europe. He commissioned an unnamed third party
to compose an arrangement of the opera, and Atwill copyrighted the
new arrangements in his own name. The defendants copied and sold
the plaintiff's arrangements and, in demurring to the infringement
suit, argued that the plaintiff's copyright was void as he was not the
composer. The plaintiff's counsel argued that, under English law,
one could secure the copyright in works he had commissioned others
to create. The court rejected the contention. In the court's view, the
English cases
recognize the right of authorship, although the materials of the
composition were procured by another, and also an equitable
title in one person to the labors of another, when the relations
of the parties are such that the former is entitled to an
assignment of the production. But, to constitute one author, he
must by his own intellectual labor applied to the materials of his
composition, produce an arrangement or compilation new in
itself.75
The cases cited, according to the court, related to "new
productions arranged or compiled from materials before known, or
obtained by others for the author, and not to the appropriation by
copyright of those materials in the same state in which they are
furnished."76 The court ultimately determined that the plaintiff's

74. 2 F. Cas. 195 (1846).

75. Id.at 197-98.
76. Id.at 198.
2003] Fisk

allegations of having added to and altered the music were legally

sufficient to constitute him the author. It cautioned, however, that if
the plaintiff's title had rested only on his having commissioned the
composition, his claim would have been defective. In short, although
the court was prepared to recognize as a matter of equity that an
author might have a duty to assign the copyright in a work he had
been commissioned to create, the law of copyright recognized no
principle that employment or a commission to create by itself
entitled the employer to the copyright. The court simply asserted,
without defending the proposition, that only intellectual labor could
make one an author.
The irony of Atwill v. Ferrett is palpable. Notwithstanding the
court's insistence that only intellectual labor could make one an
author, both the plaintiff and the unnamed third-party author of the
arrangement of the opera were pirating it from the European
composer and librettist. They were not alone: American book
publishers and theater managers were notorious in their liberal
reprinting of works that appeared abroad. As described below, Dion
Boucicault, who fought so hard for improved American copyright
protection for authors, freely borrowed from others. Yet what is
significant about Atwill is the clear statement of law: employers
acquired a claim to the copyright based only on their intellectual
contributions, not by virtue of having paid the employee to create.
The same principle of employee ownership was asserted in another
case fifteen years later, in a dispute between the copyright owner of a
book and newspaper publishers who allegedly printed the entirety of
the book in the papers. In De Witt v. Brooks, the plaintiff registered
a copyright in a book about the life and exploits of a pirate.77 Part of
the book was allegedly written by the pirate himself, part was the
pirate's alleged confession to a Deputy U.S. Marshal, and part was a
description of the "phrenological character" of the pirate "as
described by L.N. Fowler, Office U.S. Marshal." 8 Although the
court's opinion has been lost, the headnotes state that the court held
that the plaintiff could not maintain the action for infringement
because he was not the author: "The literary man who writes the
book and prepares it for publication is the author, and the copyright
is intended to protect him and not the person who employed him."79

77. 7 F. Cas. 575 (C.C.S.D.N.Y. 1861). The report in Federal Cases contains only the
headnotes. 13 Copyright Dec. 756 contains the declarations of the plaintiff and the defendants
and a note indicating that the opinion of the court has been lost.
78. 13 Copyright Dec. at 757.
79. 7 F. Cas. at 575.
 Yale Journal of Law & the Humanities [Vol. 15:1

D. Cartographyand Copyright."The Making andPublishingof Maps

and Charts
Unlike other industries described in this paper, there appears to
have been little reported litigation over the copyrights to maps.
Nonetheless, map-making is an important counterpoint to law
publishing and theater in a study of copyright. The available
evidence suggests that employers may have obtained the copyrights
to the works of their employees even if the legal rules described
above would suggest they were not entitled to them.
The first American copyright act mentioned only books, maps, and
charts among the works subject to copyright. It was not established
until the mid-nineteenth century that music and art could be
copyrighted. Photography, lithography, and other media followed in
the late nineteenth and early twentieth centuries." Maps and charts
were in great demand in early America because of the rapid
expansion of white settlements into new regions, the development of
roads, canals, and railroads, and the absence of a long-established
collection of maps for the white population. Especially noteworthy
for my purposes, the making of all but the most rudimentary maps
required the efforts of many people, including surveyors, draftsmen,
and printers.
My study of the cartography literature suggests that this is a field
where, from the beginning, the person or firm who managed to
organize the collective effort of the surveyors, draftsmen, and
printers was the one who claimed the copyright. The secondary
literature does not reveal, however, whether the person or firm who
claimed the copyright contracted with the various contributors to
assign whatever rights they might have, or whether they simply
assumed they owned it.
For the most part, mid-nineteenth-century cartography seems to
have been characterized by small enterprises: partnerships, sole
proprietors, and, later, small corporations." The publisher-often
one or two individuals, sometimes a firm--usually claimed the
copyright. 2 What the publisher did to merit such a claim varied. A

80. See 4 Stat. 436 (1831) (music); 11 Stat. 138 (1856) (dramas); 13 Star 540 (1865)
(photographs); 16 Stat 212 (1870) (painting, drawing, and sculpture); 31 Stat 488 (1912)
(motion pictures).
81. See WALTER W. RISTOW, AMERICAN MAPS AND MAPMAKERS: COMMERCIAL
CARTOGRAPHY IN THE NINETEENTH CENTURY, 20-23 (1985); JOHN RENNIE SHORT,
REPRESENTING THE REPUBLIC: MAPPING THE UNITED STATES, 1600-1900 (2001).
82. See, for example, the discussion of Henry Francis Walling and his "Map
Establishment" in New York City. RISTOW, supra note 81, at 332. Some of his maps stated that
the surveys had been made "under the direction of H.F. Walling"; some have the imprint,
"Engraved, Printed, Colored Mounted at H.F. Walling's Map Establishment"; and some have
the imprint "Walling & Rice, Publishers." Id. See also Judith A. Tyner, Images of the
2003] Fisk

single person seldom did all the actual work of surveying, drafting,
and design. Although one person could map a town or village with a
compass and odometer, larger maps would require the work of more
than one surveyor. Maps frequently gave credit to the engraver or
lithographer, but not to the draftsmen or surveyors. In some cases,
an individual who sought to publish a map did the surveying and
drafting himself and contracted with an engraver, lithographer, or
printer to manufacture it. 3 In others, a publisher with no
professional training or experience in cartography or surveying
staked his claim to the status of publisher based on his knowledge of
map reproduction and printing.' 4
Even before some map publishers grew quite large-Rand
McNally was publishing a very large number of maps by the
1870s--the practice of not crediting the contributions of all but the
supervising engraver was well established. Even during its early
years, Rand-McNally did not identify the compilers, cartographers,
or draftsmen who helped create its maps. At most the published
maps credited the supervising engraver. An 1876 map said, for
example," engraved under the direction of C.H. Waite."
Immediately below that line, however, the map's title cartouche
read, "Drawn, Engraved and Printed in Colors (under Letters
Patent) by RAND MCNALLY & CO. Chicago."s"
Cartography would seem to be a promising area for the study of
copyright and the social construction of knowledge in the work
relationship. The creative process is inherently collaborative and
usually builds-explicitly or tacitly--on past work. Cartography has
been thoughtfully analyzed from this perspective, but the full
implications of their insights have been suggested rather than
studied.' 6 J. Brian Harley has written that,

Southwest in Nineteenth-Century American Atlases, in THE MAPPING OF THE AMERICAN

SOUTHWEST 57, 60-77 (Dennis Reinhartz & Charles C. Colley eds., 1987) (discussing various
atlases from 1817 to the 1880s).
83. Henry Francis Walling used this method, for example. See RISTOW, supra note 81, at
327.
84. Robert Pearsall Smith is an example. See RISTOW, supra note 81, at 339-53. Another
prominent example was Matthew Carey, one of the leading publishers in Philadelphia and,
hardly coincidentally, the publisher of Wheaton's Reports of the Supreme Court decisions. See
supra text accompanying notes 44-62.
85. RISTOW, supra note 81, at 471-72. The patents were presumably for the color printing
process. A similar phenomenon regarding authorship-usually, but not always, the publisher
claimed it--may be seen in the maps of the American Southwest discussed in Tyner, supra
note 82, at 57, 60-77.
86. See, e.g., J.B. HARLEY, THE NEW NATURE OF MAPS: ESSAYS IN THE HISTORY OF
CARTOGRAPHY 35-36 (Paul Laxton ed., 2001); SHORT, supra note 81, at 34 (arguing that "the
illusion of single authorship" and that "maps resulted solely from the gaze of the Western
observer" are two misconceptions about early American maps, but that maps were often
 Yale Journal of Law & the Humanities [Vol. 15:1

far from holding up a simple mirror of nature that is true or

false, maps redescribe the world-like any other document-in
terms of power and of cultural practices, preferences, and
priorities. What we read on a map is as much related to an
invisible social world and to ideology as it is to phenomena seen
and measured in the landscape."
He also noted that, "more than many other texts, maps are thus
mediated by a series of technical activities, each performed by a
different 'author."'" Cartographers evidently debate the extent to
which copyright should restrict later cartographers from using earlier
maps." Yet, the dearth of reported litigation suggests that proof of
infringement was either so difficult or so easy as to make few cases
uncertain enough to require litigation to a published disposition.
The paucity of reported decisions involving disputed claims to
copyrights in maps or charts suggests that the question of map
copyright ownership appears to have been resolved significantly
through self-help.' The evidence that does exist is somewhat
contradictory. On the one hand, there is the case of Abraham
Bradley, Jr., Assistant Postmaster General in the late eighteenth
century, who obtained one of the earliest copyrights for a map in
U.S. history.91 Bradley obtained information from postmasters in
various parts of the country, and it was their work--the fact that his
map was based on new information rather than on previously
published maps-that distinguished it from maps published before.
Significantly, although Bradley obtained the information on which
his map was based from government officials in connection with his
official duties, he copyrighted it in his own name and published it
privately.
In some respects, he was like the reporters of decisions, in that he
was copyrighting a work produced during the course of his
government employment, and he was even more of an "employee"

composites building on earlier maps, the notes and sketches of travelers, and the reports of
indigenous peoples).
87. HARLEY, supra note 86, at 35-36.
88. Id. at 38.
89. See, e.g., id. at 199 (referring to "the perennial copyright problem," but suggesting that
a more pressing ethical issue for cartographers should be the political and social judgments
they make when they draw maps, such as when cartographers of South Africa made the white
urban areas the center of city maps and the black townships all but invisible).
90. The editors' note following one case stated that "[t]he publication of a map made from
materials collected while in the service of the government as draughtsman belongs to the
government." Chapman v. Ferry,12 F. 693, 696 (C.C.D. Ore. 1882). However, the report of the
case indicates it concerned available remedies for copyright infringement rather than
ownership of the copyright as between employer and employee. Id. at 695.
91. This 1796 map was obviously the product of work by postal employees. My account
draws primarily on RiSTOW, supranote 81, at 20, 70.
2003] Fisk

(and less an "independent contractor") than they were. It may be

that the Post Office concluded, just as the courts had, that allowing
government employees to copyright their work would be most likely
to ensure that the map would be published and distributed widely to
those who might benefit from it. Although Bradley's chart became
the official map of the United States Post Office around 1825, the
nature of his understanding with the Post Office about ownership
and use of the map is not known. Bradley had practiced law before
his appointment to the Post Office Department, which may explain
his interest in asserting his legal rights to the map. His legal
experience is also likely to have given him greater sophistication in
negotiating over copyright than most employee authors had at that
time.92
Although Bradley's exploits suggest a principle of employee
ownership akin to those enjoyed by reporters and playwrights, a few
cases and an attorney general opinion suggest a different rule
regarding maps prepared by government employees. These cases
might be read either as exceptions to the early rule of employer
ownership or as evidence of the coming of the transition period,
since they were all decided in the late 1850s. 3

92. Other persons commissioned by the government to compile maps may also have
claimed the copyright. The commander of the U.S. Army operations in Texas in 1849, General
William Jenkins Worth, commissioned the Federal Indian Agent for Texas, Robert Simpson
Neighbors, to determine whether troops could pass through Texas between the Pecos and El
Paso. Neighbors recruited a former Ranger, John Salmon "Rip" Ford, to accompany him. Both
Ford and Neighbors prepared reports; Ford's report was incorporated into a pamphlet
published in 1849 by a draftsman employed by the General Land Office named Robert
Creuzbaur, who evidently sought to profit from sales of the pamphlet to the large number of
people crossing Texas on their way to the California gold rush in 1849. The pamphlet was
privately published in Austin and in New York in 1849. See Robert Sidney Martin, United
States Army Mapping in Texas, 1848-50, in THE MAPPING OF THE AMERICAN SOUTHWEST,
supra note 82, at 37, 42-43 & n.16. This article reveals nothing about the copyright on the map
or about the nature of the employment relation (if any) between Ford, Neighbors, and the U.S.
Army. I infer it was privately published because the citation is to a New York publisher and to
Creuzbaur for the Texas publication. However, the maps prepared by Neighbors and Ford
probably were incorporated, along with maps and surveys conducted by other Army personnel
in 1848-1850, into a larger map that was presented to the Senate and printed by the War
Department in 1850. Id. at 55-56 & n.50. Thus, the private publication and perhaps sale of the
earlier version might not have been objectionable, since the U.S. Army obtained what it
wanted in the process.
93. No provision in the early Copyright Acts addressed government maps. In Atwill v.
Ferrett,the court mentioned in passing that "the title to road-books, maps & c., rests upon
[the] principle" that one who, by his own intellectual labor, produces a new composition may
claim the copyright even if others procured the materials for him. 2 F. Cas. 195, 198
(C.C.S.D.N.Y. 1846). The 1909 Copyright Act, repeating the prior rule established in the
Printing Act of 1895, prohibited copyright in "any publication of the United States
Government." 17 U.S.C. § 8 (1909). See supranote 63. However, if a cartographer copyrighted
the map before publication by the government, he might have retained the copyright, as § 8
also provided that "the publication or republication by the Government, either separately or in
a public document, of any material in which copyright is subsisting shall not be taken to cause
any abridgement of the copyright." Id.
 Yale Journal of Law & the Humanities [Vol. 15:1

One 1858 case concerned a map made by a cartographer who had

gathered his materials either while conducting surveys in the employ
of the Commonwealth of Pennsylvania or while employed as a
draughtsman "of a party working at her cost."94 The court held that
all rights to the map were the property of the state, as all the results
of his labor while in the state's employ were state property.95 An
1856 opinion letter written by Attorney General Cushing to
Jefferson Davis, then serving as Secretary of War, concerned a man
who had contracted with the government to engrave plates for a
nautical chart, the original manuscript of which was the property of
the United States. Attorney General Cushing opined that the
engraver had no right to make prints from the engravings: "The
printer, who is hired to print any of the works of Irving, Prescott, or
Bancroft, might, with as much sense and reason, claim a copyright in
them against the author, as Seibert can set up such claim here against
the United States."96
' But, unlike Bradley, he appeared to be copying

an existing chart in a new medium rather than designing a new map

from survey data; the different levels of originality in the
mapmakers' work might, therefore, explain the result as much as
their employment status.
A case decided in 1857, concerning a field related to cartography,
suggests the likelihood that surveyors, engravers, and printers may
have explicitly agreed that copyrights would be owned by the
publisher. In Heine v. Appleton, an artist named William Heine
accompanied an expedition to Japan and the China Sea commanded
by Commodore Perry. 9 The U.S. Government had funded the
expedition, and Heine's drawings were included in the expedition
report to the Secretary of the Navy, which was published by order of
Congress. When Heine later sought to enjoin publication of books
containing his illustrations from the expeditions, the court denied the
injunction on two alternative grounds. First, the court found that
Heine and Perry had expressly agreed that Heine's drawings would
become the property of the U.S. government.98 Second, the court
found that, after the expedition, the defendants (publishers of the
books) had paid Heine to modify several of the original drawings to
make them suitable for inclusion in the books. At that time Heine
had not claimed the copyright in the drawings. "It would be
inequitable now to permit him, when he has been paid to aid in their

94. Pennsylvania v. Desilver, 3 Phila. 31, 14 Copyright Dec. 2039 (C.P. 1858).
95. Id.14 Copyright L.Dec. (CCH) at 2039.
96. 7 Op. Att'y Gen. 656 (1856), 15 Copyright Dec. 3049.
97. Heine v. Appleton, 11 F. Cas. 1031 (C.C.S.D.N.Y. 1857).
98. Id. at 1032.
2003] Fisk

publication and sale, and has thus aided in their publication, with a
view to their sale, to stop their sale, even if he had a valid copyright
in them.""
Heine cannot be read as an early work-for-hire case in the modern
sense of works-for-hire. Neither of Heine's alternative holdings
recognizes that the employer would own the copyright to employee
drawings simply by virtue of employment. One holding relied on an
express agreement and emphasized that Heine persuaded
Commodore Perry to allow him to accompany the expedition only
by promising that the government would own his sketches and
drawings.1" The other relied on the notion that Heine was estopped
from seeking an injunction against publication of drawings because
he had been paid to modify them for publication and had failed to
claim a copyright.
At the very least, the example of cartography suggests that even
during the mid-nineteenth century (and certainly later), the
attribution of authorship and the ownership of copyright of works
produced for commercial purposes depended on the nature of the
work relationship and the creative process that occurred within it.
The more likely that there were multiple creators, the more likely
that copyright could be claimed by the coordinator of the creative
effort rather than by other contributors. Had the cases that ended up
in litigation forced courts to contemplate such a creative process,
they might have crafted a different sort of default rule for all other
cases involving the employment of creative people. Later in the
century, when cases that involved collaborative creative processes in
the workplace were litigated, courts struggled to decide whether

99. Id at 1033.
100. The opinion explained the negotiations as follows:
Previous to the sailing of the expedition to Japan, the plaintiff applied to Commodore
Perry, to be employed as an artist, and to accompany the expedition, as such ....
[Flinally the commodore consented to receive him in the capacity of a master's mate, on
condition that he should sign the shipping articles as such master's mate, and do
whatever duties might be required of him, and be subject to all the rules and regulations
of the squadron. When the commodore consented that the plaintiff might join the
expedition, he informed him that all the sketches and drawings which should be made by
any one attached to the expedition were to be the exclusive property of the government
of the United States, and that no one could appropriate to his own use any sketch or
drawing that might be made. To this the plaintiff gave his assent .... Upon the return of
the expedition to his country, the sketches and drawings which the plaintiff made, were,
with his assent, incorporated in the report made by the commodore to the secretary of
the navy, and were placed at the disposal of congress; ... Under these circumstances, the
plaintiff was not such author of the prints and engravings in question, as to be able to
acquire an exclusive right to the same as author or proprietor, by virtue of the certificate
of copyright which he obtained. The sketches and drawings were made for the
government, to be at their disposal.
Id. at 1032-33.
 Yale Journal of Law & the Humanities [Vol. 15:1

these early precedents, which clearly articulated a rule of employee

ownership, were distinguishable.
The change in default rules between the early nineteenth century
and the early twentieth may be explained, in part, by a rise in the
number of cases involving employees who participated in
collaborative creative processes. The more the courts saw cases in
which a number of people had contributed to the work, the more
logical it was to accord the copyright to the representative of the
collective--that is, the employer. On the other hand, the increasing
presence of collaborative creation surely was not the only factor. As
explained below, cases involving individual authors working alone
(although in the employ of another) continued to dominate the field
of employee-employer copyright litigation throughout the
nineteenth century.
Prior to the Civil War, no court recognized that an employer was
entitled to copyright the works of its employees simply by virtue of
the employment; indeed, courts assumed just the opposite. It took an
express agreement to assign the copyright to the employer, as in
Heine v. Appleton, to persuade a court to conclude that an employer
owned the copyright. The strikingly pro-employee law was a function
of the elevated social status and independence of the "employees"
who created the work, the judges' assumptions about the nature of
the creative process, instrumental concerns about facilitating book
distribution, and the powerful rhetoric of authorship as highly
personal intellectual labor.

III. THE PERIOD OF TRANSITION: 1860-1900

In the postbellum period, courts began to recognize the rights of
employers in the copyrighted works of their employees. At the
beginning, employer ownership was based on the existence of an
express contract granting the employer the copyright or, in one case,
on an equitable principle, not a copyright rule. By the end of the
century, however, courts increasingly began to describe the
employment relationship as a contract by which the employer
acquired the rights to all of the employee's work, including the
copyrights. Furthermore, by the end of the century, corporations as
employers came to the fore. In the earliest such cases, the personal
involvement of the corporation's president was identified as the basis
for corporate copyright ownership, but in the twentieth century,
courts came to understand that a corporation--the quintessential
"corporate" (as in collective) author--should own the rights to the
work created by all of the persons who worked for the corporation.
2003] Fisk

A. Civil War Theater and the Beginning of the End of Employee

Ownership
A spate of theater cases decided in 1860, 1861, and 1862 began the
transition from the old rule of authorial ownership to the modern
rule of employer ownership. All involved actors-cum-theater
managers who contracted with theater owners to write a play to be
performed in the theater. The cases stated a rule that the
employee/actor/stage manager owned the copyright and the
employer at most had a license to perform it in that particular
theater for that particular run. Yet the courts suggested the
possibility for greater rights for theater manager/employers by
emphasizing the contributions that they had made to the changes in
the plays. Because these cases involved some of the most acclaimed
actors of their day, it is unclear how much light they shed on how
courts would (or did) handle anonymous, unreported cases involving
litigants who were not celebrities.
The first two cases involved The Octoroon, written by the famous
playwright and actor, Dion Boucicault. These cases should be
understood in the context of their facts. Boucicault was famous and,
at the time of the litigation, had already had plays produced in New
York theaters. He had made his debut on the American stage in
Boston and New York in 1854, he and his wife appeared in
numerous plays he had written in the late 1850s, and he received
significant attention for his successful play, The Poor of New York,
which opened in 1857 at Wallack's theater in New York."' Thus, at
the time the pair of Octoroon cases were decided, Boucicault was
hardly the ordinary, anonymous employee. He was something of a
celebrity in American theater, and he was a recognized playwright
(even if some took a dim view of the intellectual merit of
12
Boucicault's work). 0
Boucicault had been active in a successful effort to secure
improved copyright protection for American playwrights. Along
with others, he had lobbied Congress to amend the law to provide
for copyright protection in presentations of plays as well as in their
texts; Congress enacted such a law in 1856.2°3 Ironically, as one of
Boucicault's biographers tartly observed, Boucicault fought for
copyright revisions and litigated to protect his rights, yet "he had not

101. See generalIyRICHARD FAWKES, DION BOUCICAULT (1979); HOGAN, supra note 73.
102. Boucicault ranked only behind Shakespeare as the most frequently performed
playwright in Philadelphia theaters between 1856 and 1878. The numbers were: Shakespeare,
2314; Boucicault, 1587; Tom Taylor, 934; J.B. Buckstone, 839; John Brougham, 829; and J.M.
Morton, 652. BRUCE A. MCCONACHIE, MELODRAMATIC FORMATIONS: AMERICAN
THEATRE AND SOCIETY, 1820-1870, at 241 (1992).
103. 11 Stat. 138 (1856).
 Yale Journal of Law & the Humanities [Vol. 15:1

the least hesitation in continuing his piratical raids on French

dramatic literature and, indeed, any place that suggested a fertile
idea. Sometimes he quite generously acknowledged his source, but I
doubt that he usually paid any royalties. '""° He complained bitterly
about London theater managers who preferred to pay £25 for a
translation of a successful French play than £300 to £500 for a new
but untried work. By such practices, Boucicault protested, "the
English dramatist was obliged to relinquish the stage altogether or to
become a French copyist." ' 5 Or, as Boucicault himself labored to do,
the dramatist simply had to make sure that his pirated versions of
others' works were more successful than others' pirated versions of
his."°
Just as Boucicault was no ordinary employee, The Octoroon was
no ordinary work. 1 Its portrayal of life on a Louisiana plantation
created a sensation in American theater, in part because it dealt with
slavery in a way calculated to offend neither North nor South by
suggesting that slavery was an intolerable system while depicting
Southern slave owners as sympathetic characters. The portrayals of
the slaves (and the Indian, who was played by Boucicault) were at
once dismally racist and yet humane (at least by the white standards
of the time)."0 '
According to the facts recited in the opinion, Boucicault wrote the

104. HOGAN, supranote 73, at 33.

105. Id. (quoting Dion Boucicault, The Decline of the Drama, 125 N. AM. REV. 243
(September 1877)).
106. Boucicault was not alone in such "raids" or in complaining about the lack of copyright
protection. Others complained that Laura Keene preferred foreign plays to native ones
because she could obtain them free. Tice L. Miller, The Image of Fashionable Society in
American Comedy, 1840-1870, in WHEN THEY WEREN'T DOING SHAKESPEARE: ESSAYS ON
NINETEENTH-CENTURY BRITISH AND AMERICAN THEATRE 244 (Judith L. Fisher & Stephen
Watt eds., 1989)[hereinafter WHEN THEY WEREN'T DOING SHAKESPEARE].
107. It was Boucicault's first success in American theater; he evidently had made a careful
study of American theater and American taste and in the Octoroon he finally hit gold. James
Hurt, Yn WHEN THEY WEREN'T DOING SHAKESPEARE, supra note 106, at 255.
108. See Dion Boucicault, THE Octoroon, in SELECTED PLAYS BY DION BOUCICAULT
135-90 (Andrew Parker ed., 1987). The story of The Octoroon is worth telling. A Louisiana
plantation is to be sold to satisfy debts incurred by its late, spendthrift former owner, Judge
Peyton, on account of the mismanagement of the scheming yankee overseer M'Closky. The
slaves are to be sold along with the plantation. The plantation had been left to the judge's son,
George, a charming playboy recently returned from Paris. Zoe, the judge's daughter by a slave,
is to be sold along with the slaves because the judge's letter manumitting her is declared invalid
on account of a legal technicality. Zoe had been raised as a daughter by Mrs. Peyton.
Naturally, George and Zoe fall in love. M'Closky arranges to intercept a letter and check
containing enough money to pay off the debts, but the faithful slave, Paul, gets the letter first.
M'Closky kills Paul and blames the Indian Wahnotee (played by Boucicault) for the murder. A
camera happened to photograph the murder and later the photograph is discovered. The
Indian Wahnotee avenges the death of Paul by murdering M'Closky. Unfortunately, M'Closky
is discovered too late, because Zoe, who had already been sold to M'Closky at the slave
auction, swallows poison to avoid her fate, and in the closing scene she dies in George's arms.
2003] Fisk

play in 1859 while "employed as an actor and stage manager" of the

Winter Garden Theater in New York City. William Stuart
(sometimes spelled Stewart) was the lessee and manager of the
theater. A couple of years before, Stuart and Boucicault had opened
the Washington Theater in Washington, D.C., but the theater was
not a success."' So the Winter Garden was not their first
collaboration, and, obviously, Boucicault was not an "employee" in
the sense of being Stuart's subordinate."'
Boucicault and Stuart orally agreed that the former would, as one
court found, "write a play representing life on the Mississippi, and
that it should be performed at Stewart's theater so long as it should
continue to draw good audiences."' Boucicault quit the Winter
Garden after six nights because he and his wife, Agnes Robertson,
who also acted in the play, were in a dispute with Stuart over their
salaries. The day before quitting, Boucicault registered a copyright
on the play in his own name. When Stuart continued to run the play
without him, Boucicault brought suit in Massachusetts for a
preliminary injunction,"2 and in New York for damages."' Although
both courts found that Boucicault was Stuart's employee at the time
he wrote the play, neither thought that fact entitled Stuart to the
copyright.
The Massachusetts court rejected Stuart's defense that Boucicault
was employed to write the play. The court determined that he was
employed as an actor and stage manager, but not as an author, and
that a simple agreement to write a play for his employer did not
entitle the employer to the copyright." 4 The court's framing of the
case left open the possibility that an employer might obtain the
copyright of works by one employed for no purpose other than to
write.
In the suit for damages two years later, the New York court did
not emphasize the difference between employment as an actor and
employment as a writer. Rather, it suggested that only an express
contract would entitle the employer to the copyright in any of its
employees' works because of the special nature of authorship.
Boucicault's agreement to write and perform in the play, and to

109. HOGAN, supra note 73, at 40.

110. Boucicault indeed is credited in some books as being the producer of the play;
certainly one may think of him and Stuart more as partners than as employer and employee in
the modern sense. See, e.g., MARY CAROLINE CRAWFORD, THE ROMANCE OF AMERICAN
THEATRE 369 (1913).
111. Roberts v. Myers, 20 F. Cas. 898, 899 (C.C.D. Mass. 1860).
112. Id.at 898.
113. Boucicault v. Fox, 3 F. Cas. 977 (C.C.S.D.N.Y. 1862).
114. Roberts, 20 F. Cas. at 899.
 Yale Journal of Law & the Humanities [Vol. 15:1

share half the profits from it, did not confer upon Stuart title to the
drama:
The title to literary property is in the author whose intellect has
given birth to the thoughts and wrought them into the
composition, unless he has transferred that title, by contract, to
another. In the present case, no such contract is proved. The
most that could possibly be said, in regard to the right of Stuart,
or his trustee, in the play, is, that the arrangement entitled them
to have it performed at the Winter Garden as long as it would
5
run."

Note how the paternity metaphor-the work is the "brat of the

brain," in Daniel Defoe's terms, or "the author whose intellect has
given birth to the thoughts" here-does the rhetorical work of
justifying employee ownership. In the earlier theater case, Atwill v.
Ferrett,the court had invoked the property metaphor to achieve the
same result: "to constitute one an author, he must by his own
intellectual labor applied to the materials of his composition,
produce an arrangement or compilation new in itself."116 ' In both

cases, the crucial reasoning began with a metaphor of the nature of

authorship that by itself was supposed to justify copyright-either
the rights that flow from Locke's theory of property or the rights of
parents in their children-and then used that metaphor to justify
employee ownership of the copyright.
The court ruled that only an express agreement to assign the
copyright would suffice to transfer it from employee to employer. It
grounded the rule in the particularly personal nature of authorship:
"A man's intellectual productions are peculiarly his own, and,
although they may have been brought forth by the author while in
the general employment of another, yet he will not be deemed to
have parted with his right and transferred it to his employer."' '7 If
authorship is the "bringing forth" of new life, a new creation, and the
creation is "peculiarly" one's own and irreducibly personal, a legal
rule should not presume one has given away one's property (or one's
children) in the absence of a clear intent to do so. Thus, the court
articulated a clear statement rule, explaining that "producers" who
employed "authors in particular literary enterprises" could easily,
when they felt it in their interests, contract for employer ownership
of the copyright, and that a contract transferring the copyright would

115. Boucicault, 3 F. Cas. at 980.

116. A twill, 2 F. Cas. at 198. See supra text accompanying note 75.
117. Id
2003] Fisk

be enforceable." 8 The personal nature of authorship, in the court's

view, justified a default rule of employee ownership and required a
clear agreement to transfer the copyright to anyone else.
The third case in the trilogy of actor/writer cases from the Civil
War period was the 1861 case Keene v. Wheatley, which concerned
the famous play Our American Cousin."9 The decision articulated a
very different and more employer-friendly view of copyright
ownership. Keene drew on a variety of precedents having to do with
ownership of inventions'2" and trade secrets,' 2 ' and with the doctrine
of Lumley v. Wagner, which held that an opera singer who breached
a contract to perform at one theater could be enjoined from
performing elsewhere for the term of her contract.'22 Significantly, it
did not rely on copyright doctrine or on any of the copyright cases
that preceded it.'
According to the statement of facts, Joshua Silsbee, an American
actor, had performed in Our American Cousin in Benjamin
Webster's famous London theater company. Silsbee returned with a
copy of the play to the United States where he died, bequeathing the
manuscript to his widow, who later sold it to two Philadelphia
theater producers, Wheatley and John Clark. Meanwhile, the British
author of the play, Tom Taylor, sold the American rights to Laura
Keene, the proprietor of a theater in New York, for $1000. With the
assistance of Joseph Jefferson, an actor in her company, Keene

118. Id
119. 14 F. Cas. 180 (C.C.E.D. Pa. 1861). The play is about a gullible English baronet whose
financial advisor swindles him out of the family fortune. To save the fortune, the baronet's
lovely daughter is engaged to be married to the unscrupulous advisor. An American relative
turns up on the scene and eventually everything is straightened out. TOM TAYLOR, OUR
AMERICAN COUSIN: THE PLAY THAT CHANGED HISTORY (Welford Dunaway Taylor ed.,
1990).
120. See 14 F. Cas. at 187 ("Where an inventor, in the course of his experimental essays,
employes an assistant who suggests, and adapts, a subordinate improvement, it is, in law, an
incident, or part, of the employer's main invention.").
121. See id. at 186-87 ("A calico printer discharged his head colorman, who sued him in
trover for a book of entries of processes of mixing the colors used in his business .... The
decision was that he could not recovery.... [T]hough there might be inventions of the plaintiff
in it, yet they were the property of the master.") (citing Makepeace v. Jackson, 4 Taunt. 770).
The principle of employer ownership of a book of secret dye recipes and samples was later
adopted in Pennsylvania in a pair of cases, Dempsey v. Dobson, 39 A. 493 (Pa. 1898), and
Dempsey v. Dobson, 34 A. 459 (Pa. 1896). I discuss them in Fisk, Law and the Employee
Inventor,supra note 27, at 1154-56.
122. See 14 F. Cas. at 186 (citing Lumley v. Wagner, 42 Eng. Rep. 687 (Ch. 1852)). In a
related case, a court held that a theater manager could be enjoined from soliciting the singer to
break her exclusive engagement. Lumley v. Gye, 118 Eng. Rep. 749, 755 (Q.B. 1853). The
Lumley doctrine is discussed in Lea S. VanderVelde, The Gendered Origins of the Lumley
Doctrine:BindingMen's Consciencesand Women's Fidelity,101 YALE L.J. 775 (1992).
123. 14 F. Cas. at 188 ("[T]he equitable jurisdiction which we are now considering is
exercisable on grounds which are independent of proprietary right in the party injured, or in
any other person.").
38 Yale Journal of Law & the Humanities [Vol. 15:1

adapted it for performance in her theater.'24 After the play's wildly

successful run at Keene's Theater in New York, Keene and her
theater company took it to Ford's Theater in Washington, D.C. 125
Due to OurAmerican Cousin's huge success in New York-which
not only shaped taste in theater, but also created fashion '26-- theater
managers in other cities eagerly sought to mount their own
productions and share in its success. Jefferson, too, sought to share
the wealth and sold to Wheatley and John Clark the additions he had
developed with Keene's company. But Keene, who had a long and
successful career as an actress and theater manager, was a
determined adversary. In addition to acting, managing her theater,
and touring, Keene also had a play-brokering business. She
constantly sought out new material and revised it to suit her needs;
she wrote her own plays; she copyrighted all of the plays in her own
name; and she frequently licensed them to theater managers all over
the country.' It is thus not surprising that Keene litigated over the
rights to the play, for she obviously thought of plays as valuable
literary properties that she could license to others as well as perform
herself. She regarded the company's changes to Our American
Cousin as her property rather than as the creative property of
Jefferson and the others. She copyrighted the play shortly before the

124. Id.at 182.

125. It was during the play that John Wilkes Booth shot Abraham Lincoln. He was able to
plan the assassination because he knew the play and the theater. After Booth shot Lincoln, he
jumped from the state box down onto the stage, knocked Keene out of his way, and ran out the
stage door. Keene went up to the box and held Lincoln's head for some of the time until he
was carried to the boarding house across the street, where he died. She stayed with Mary Todd
Lincoln at the boarding house until morning. Evidently Keene's blood-stained dress became an
artifact much sought-after by souvenir hunters. Keene herself was briefly arrested when she
and her company left Washington, presumably on suspicion that she had been involved in
Booth's plot. She was promptly released and never prosecuted. VERNANNE BRYAN, LAURA
KEENE: A BRITISH ACTRESS ON THE AMERICAN STAGE, 1826-1873, at 136-42 (1993).
An interesting coincidence is that Lincoln saw the play on the closing night at Ford's
Theater the play scheduled to open the next night was The Octoroon. See TAYLOR, supra
note 119, at 30 (showing a photograph of Ford's Theater draped in black following the
assassination, and in the foreground a billboard announcing The Octoroon, which was
scheduled to open). The enormous popularity of both plays makes the coincidence not
improbable. Boucicault and Keene had also collaborated on other projects. He wrote the
famous play The Colleen Bawn for her, which premiered in March 1860. CRAWFORD, supra
note 110, at 370-71.
126. However, perhaps it is a bit much to call it "the play that changed history," as one
recent work does. TAYLOR, supra note 119.
127. She offered to license one of "her plays" to a Boston theater manager as follows:
It is now the sensation of London and Paris. It has been carefully rewritten by myself to
suit America. All my stage knowledge has been employed to give the two grand scenic
effects the elements of such a success as will create a genuine sensation. The story is
simple, yet thrilling, the piece full of good parts, incidents and fun.... It must be played
immediately in your city and I give your house the preference therefore an early reply
will oblige.
BRYAN, supra note 125, at 115-16.
2003] Fisk

premiere, and, according to her biographer, she dreamed up the title

when she filed for copyright because Tom Taylor, the play's original
author, had not titled it.'2s
The work of those in Keene's company, including Jefferson, who
played the American cousin, Asa Trenchard, may have contributed
to the success of the play. Keene allowed an unprecedented two
weeks for rehearsal of the play before it opened in New York, and it
must have been during those two weeks that the play evolved to the
point that she and others would dispute ownership. The two weeks
"gave the cast members the unusual privilege of 'developing their
characters."" 29 Jefferson developed the American cousin character
he played by significantly altering a stock character in nineteenth
century theater (the "Yankee"), transforming the role from a "droll,
awkward, bumpkin" or "broad and extravagant boor" into someone
of "stature," a "quiet, easy, natural, and at the same time excessively
droll fellow."' 3 ° Edward Sothern created many distinctive aspects of
his Lord Dundreary character: He evidently devised the drooping
mustache that became the hallmark of an eccentric comic as well as a
variety of verbal and physical tics that became staples of nineteenth-
century comedy."' The cast, which included and was directed by
Keene, collectively revised the script. Keene designed elaborate new
sets.' All the work paid off. Sothern and Jefferson became stars and
Keene's own fame and wealth grew.'33
The court concluded that, because the British author was not a
U.S. resident, Keene did not have a valid copyright claim to the
original version of the play. 1"4 Moreover, she could not obtain a
copyright for the additions, even though they were done by

128. Id at 83. Keene and Jefferson disagreed about the extent of their own contributions
in revising Taylor's version of the play for the American stage. Keene's biographer emphasizes
Keene's rewriting. Id. Jefferson's autobiography, not surprisingly, emphasizes his. He suggests
that Keene initially "thought little of the play, which remained neglected upon her desk for
some time" and that it was only because her business manager "chanced" upon it and gave it to
Jefferson, who saw "the chance of making a strong character of the leading part," that the play
was recommended for production. JOSEPH JEFFERSON, THE AUTOBIOGRAPHY OF JOSEPH
JEFFERSON 193-94 (New York, Century Co. 1889).
129. BRYAN, supra note 125, at 83.
130. Id at 84 (quoting Frank Leslie, ILLUSTRATED NEWSPAPER, Mar. 5, 1859).
131. See CRAWFORD, supra note 110, at 334. The book also contains a photograph of
Sothern as Dundreary.
132. Id
133. Id at 327-33. The play was so popular that she gave benefit performances to raise
money for good causes, and eventually she established the tradition of a charity matinee.
Society ladies who previously had snubbed Keene began to flock to her theater for the charity
performances, and Keene began to find the social acceptance that eluded most actresses and
theater people. She also found prosperity. Jefferson quipped that during the run, Keene began
"to twinkle with little brilliants until at the end of three months she was ablaze with
diamonds." BRYAN, supranote 125, at 84 (quoting JEFFERSON, supra note 128, at 194).
134. 14 F. Cas. at 185.
 Yale Journal of Law & the Humanities [Vol. 15:1

American residents (principally Joseph Jefferson), because they

were only accessions, not an independent work. Nor could she
challenge the defendants' use of the play obtained from Silsbee's
widow on any other grounds, for they owed her no duty not to use
it. 3' But the court found she could seek a remedy for the use of the
additions procured from Jefferson, which is what makes the case
particularly significant:
Mr. Jefferson, while in the general theatrical employment of the
complainant, engaged in the particular office of assisting in the
adaptation of this play; and made the additions in question in
the course of his willing performance of this duty. She
consequently became the proprietor of them as products of his
intellectual exertion in a particular service in her employment.'36
To reach this result, the court downplayed the relevance of
copyright law, looking rather to general principles of "equitable
jurisprudence," independent "of any question of literary
proprietorship."137' Rather
than emphasizing the role of the author,
the court relied instead on cases involving other sorts of employees:
a calico printer forced to relinquish a book of his dye recipes when
fired, 3' and the opera singer Johanna Wagner, who was enjoined
from performing for any other theater during the term of her
exclusive engagement.'39 The court further relied on general
principles governing "secrets of inventors, or improper disclosures of
knowledge '' ° acquired in professional relations or in those of service or
agency. iN

The court's treatment of these. cognate doctrines involving

ownership of workplace knowledge--the enticement rule of Lumley
v. Wagner, the emerging trade secret rule-was unprecedented.
Most of the cases it cited involving employees were of recent vintage
(none was more than a decade or two old), and Keene v. Wheatley
was the first court to put them all together in a published opinion
articulating a default rule of employer ownership. The court stated a
general principle of employer ownership of employee knowledge or
creative works at a time when very few courts had so ruled and when
there were some cases stating just the opposite.''

135. Id at 188.
136. Id at 187.
137. Id
138. Id at 186-87 (citing Makepeace v. Jackson, 4 Taunt. 770).
139. Id. at 186 (citing Lumley v. Wagner, 42 Eng. Rep. 687 (Ch. 1852)).
140. Id. at 187.
141. See Fisk, Law and the Employee Inventor, supra note 27, at 1148; Fisk, Working
Knowledge, supra note 27, at 483.
2003] Fisk

Keene's win in Keene v. Wheatley in 1861 was not the end of the
story. She spent years and instituted at least three suits trying
(unsuccessfully) to protect her exclusive right to Our American
Cousin. Six years later, Keene was again trying to fend off those who
sought to produce the play without her permission. "2 She learned
that Edwin Booth, brother of the infamous John Wilkes Booth,'43
had purchased the Winter Garden theater and named as manager
her adversary from Keene v. Wheatley, the co-defendant John Clark.
She sued Clark and, perhaps growing skeptical about the efficacy of
legal remedies, took her fight public. She wrote a letter addressed to
him, but intended for publication, in which she accused him of theft.
She asserted that it was her "enterprise, industry and expenditure"
that made the play a success:
You know well this is not a case of copyright it was my personal,
private property, and should have been held sacred to me by
every respectable member of our profession .... I am so
identified with that play that I consider it one of my most
valuable possessions. Why seek to deprive a woman of her
honestly acquired property? Why not take the Beteran or
Rosedale or any other play owned by a man?'"

142. Keene v. Clark, 2 Abb. Pr. (n.s.) 341 (N.Y. Super. Ct. 1867). In Keene v. Clark, Keene
sought to prove that her entitlement to Our American Cousin was res judicata by virtue of the
judgment in Keene v. Wheatley. (She could assert res judicata because Clark was a defendant
in both suits; presumably her theory was that Booth was in privity with Clark and thus could be
bound by the judgment in the first suit even though he was not a party.) The trial court refused
to admit the pleadings and decree, but on appeal the Superior Court held that the record
should be admitted, and remanded for a trial at which Keene would prevail if she proved that
she had not printed or otherwise dedicated to the public the manuscript to her version of Our
American Cousin, and that Clark had obtained a copy of it from a "surreptitious source" and
intended to produce it in New York. Id. at 341. Neither the published opinion nor Keene's
biography indicates what happened after the Superior Court sent the case back for trial.
Presumably it settled, though on what terms remains unknown.
143. On Edwin Booth's life and relationship to John Wilkes Booth, see CRAWFORD, supra
note 110, at ch. XIII.
144. The letter read:
Sir-I see by your advertisement in the Herald that you purpose playing Our American
Cousin. No one in our profession is better aware than yourself that in all honor, honesty
and fair dealing the said play is my sole property. I gave the author a large sum for it
when it was an unacted manuscript. By my enterprise, industry and expenditure, I made
the play a great success. You then being one of the managers of the Arch street theatre,
Philadelphia, determined to act the play. You obtained a copy of it, which was proven on
trial to have been stolen from the author, Tom Taylor. I brought a suit against you then
and it was proved you had obtained all the original matter which had been written by my
stage manager and myself from an employee of my theatre. Judge Cadwalader decided
against you, and under whose decision you were compelled to play damages. You
ultimately apologized for your share of the transaction and wished to shift all the blame
on others. You had made many thousands of dollars, however, while the suit lasted.
Since then you have played it in Washington, Baltimore, etc., knowing full well that my
professional engagements would prevent restraining you by law from so doing. In these
places you have again made large sums from its production. Should not the money you
 Yale Journal of Law & the Humanities [Vol. 15:1

Keene was absolutely right that this was "not a case of copyright."
Unlike the cases involving The Octoroon, the ruling in her favor was
based on "equitable principles," not copyright, and thus it was
nominally consistent with the Boucicault rulings. But at another
level, it was not at all consistent. Except for the fact that she had
been involved in adapting OurAmerican Cousin, and no one alleged
that Stuart had been involved in the creation of The Octoroon, there
was no more reason for her to receive copyright protection than
Stuart, so long as one adheres to the intellectual labor theory of
authorship.
The court did not attempt to distinguish the cases, so why the
difference? Perhaps it was Keene's own involvement in adapting Our
American Cousin that led the court to recognize her entitlement.
Keene's lawyer may have made in litigation the same claim that
Keene herself later made in print: the play was her "personal, private
property," and the product of her "enterprise, industry and
expenditure." Characterizing the play as Keene's own work made
the claim of employer ownership more compelling because it did not
contradict the extant justification for employee ownership of the
copyrights to works that were entirely the product of their own
efforts (even though they were employed to exert that effort).
Finally, there was her celebrity, the notoriety of the play, and the
fact that it was well known that neither she nor Jefferson was really
the "author" in the copyright sense. These facts, as well as her

have already made from my property content you? You know well this is not a case of
copyright it was my personal, private property, and should have been held sacred to me
by every respectable member of our profession; but now, while I am absent from New
York, you take advantage of that absence to play Our American Cousin. I am so
identified with that play that I consider it one of my most valuable possessions. Why seek
to deprive a woman of her honestly acquired property? Why not take the Beteran or
Rosedale or any other play owned by a man?
The bad taste of seeking to deprive me of the use of this play is only equaled by your
ever appearing in a comedy which ought to have only a memory of shame and horror for
you and every member of your family. You cannot lift the cloud which has fallen upon
our whole profession by acts which set at naught all regard for principle and right.
Laura Keene
BRYAN, supra note 125, at 144-45 (quoting BEN GRAF HENNEKE, LAURA KEENE, A
BIOGRAPHY 223-24 (1990)). The letter and its subsequent history underscore the importance
of claiming the moral high ground and the rhetorical force of staking that claim in authorship.
Keene's letter used a number of devices to do so. Obviously, one was the characterization of
the play as her property that he had stolen. Interestingly, in the ensuing litigation, Keene v.
Clark,attorneys for both parties stipulated that she did not in fact write the letter accusing him
of theft, since, as a lady, she could not possibly have written such a scurrilous attack. BRYAN,
supra note 125, at 145. Perhaps one ought not take this stipulation at face value. Lawyers for
both litigants might have wanted the letter excluded from the record for other reasons. A
second tactic was to remind the world that a member of the Booth family, above all people,
should not have any connection with the play that Lincoln was watching when John Wilkes
Booth killed him. Third, invoking chivalry and feminine prerogative, she accused him of
stealing from her because she was a woman.
2003] Fisk

aggressiveness in claiming the fruits of her "industry and enterprise"

in hiring Jefferson and promoting the play, combined to make her
claim to ownership of the play more compelling than the theater
managers' claims had been. The old rhetoric of authorship, thus, did
not fit. The court, free from the adverse precedent of Atwill and
Wheaton, created a new rhetoric of employer prerogative based on
industry, enterprise, and collaborative creation.

B. Law Publishingand EarlyRecognition ofEmployer Ownership

The possibility of distinguishing the Keene cases from the
Boucicault cases on the difference between copyright rules and
equitable rules did not last even a decade. In 1869, a federal court in
Massachusetts stated in dicta, without acknowledging its apparent
departure from Roberts v. Myers, that an employer would have the
copyright in the literary products of its employees. Lawrence v.
Dana,which became a very influential decision, was another chapter
in the long (and unsuccessful) copyright battles fought by Henry
Wheaton, the erstwhile Supreme Court reporter. Wheaton authored
a respected treatise on international law and, after his death, his wife
Catharine contracted with William 45Lawrence to prepare an updated
edition of her late husband's work.
In Lawrence, as in previous cases, the existence (or absence) of an
express contract was crucial. Catharine Wheaton had expressly
contracted with Lawrence that she would make no use of his notes in
any later editions without his written consent and that he would have
the right to make any use of them he wished. However, what is
significant about the case is the dictum. The court remarked that,
absent such a contract, Catharine would have owned the copyright to
46
the notes.1
The principle of employer ownership as a default rule, recognized
in Keene and Lawrence v. Dana, could be introduced relatively
painlessly because in both cases the employee did not lose significant
rights. In Keene, Jefferson lost little because he was not a party to
the litigation; presumably, he had sold Wheatley and Clark his rights,
and they were the ones who would stand to lose if it turned out that
he had no rights to sell. In any event, he only authored certain
additions, not the whole play, and the defendants had an
independent source for everything in the play except his additions.
In Lawrence v. Dana, the employee-author lost nothing by
recognition of the principle because there was an express contract
protecting his claim to copyright.

145. 15 F. Cas. 26 (C.C.D. Mass. 1869).

146. Id at 50-51.
 Yale Journal of Law & the Humanities [Vol. 15:1

It is enormously important that the work-for-hire principle slipped

into the cases without the usual adversary process (either because
the actual employee author was not a party or because the principle
was stated in dicta), for these seminal cases did not actually force a
court to choose between the rights of the employee and those of the
employer. For a generation afterward, courts simply repeated the
principle as dicta giving it the patina of age, without ever actually
relying on it to decide a case. A string of cases decided in the 1880s
and early 1890s persisted in finding one reason or another to
recognize the copyright of the employee-author. Many
acknowledged employer ownership in dicta, but no courts actually
held that the fact of employment ipso facto entitled the employer to
the copyright. 47
The Supreme Court's 1888 opinion in Callaghan v. Myers, an
action alleging infringement of the Illinois Reports, also articulated a
default rule of employee ownership, making clear that the rationale
for the rule was that it reflected the likely intent of the parties.'48 The
plaintiff had purchased the copyright from the reporter of decisions.
He sued the publisher who was selling copies of the Reports,
including the reporter's headnotes, statements of the case, and
summaries of the arguments of counsel. The defendants argued that
the reporter could acquire no copyright in them. The lower court
found the contention would have "great force" "if an adequate
compensation was paid by the state to the reporter for the work
done by him in preparing the volumes." But the court concluded:
I cannot find that view was taken of the case by the state and the
court in the appointment of the reporter at that time. On the
contrary, it seems to have been considered that the reporter was
entitled to any profits which might arise from the sale of these
volumes, and that they constituted part of the perquisites of his
office.'49
On review, the Supreme Court found no merit to the contention
that the reporter, as a public servant, could acquire no copyright:
Even though a reporter may be a sworn public officer,
appointed by the authority of the government which creates the
court of which he is made the reporter, and even though he may
be paid a fixed salary for his labors, yet, in the absence of any
inhibition forbidding him to take a copyright for that which is
the lawful subject of copyright in him, or reserving a copyright

147. See infra text accompanying notes 148-153 and 178.

148. 128 U.S. 617 (1888).
149. Myers v. Callaghan,5 F. 726 (C.C.N.D. Il. 1881).
2003] Fisk

to the government as the assignee of his work, he is not

deprived of the privilege of taking out a copyright which would
otherwise exist.'5'
Later, the Court added that whether the reporter was compensated
by salary (he was not), or even by the state's purchase of a certain
number of volumes, was irrelevant: "[I]n the view we take of the
case, the question of a salary or no salary has no bearing upon the
subject."'' . Rather, the Court said, there was "a tacit assent by the
' 52
government to his exercising such a privilege.'
In just twenty years, courts had reallocated copyright ownership
simply by rewriting the implied contract between employer and
employee to include a principle of employer ownership.'53 In other
words, once the courts began to think of copyright ownership as a
matter of "tacit assent"--rather than as a virtually inalienable right
associated with ownership of real property or in terms of the strong
tie connoted by the paternity metaphor--courts felt a different
default rule was appropriate.

C Employees andIndependent Contractors

A number of factors may explain the change in default rules.
Courts might have felt that a default rule of employer ownership was
more likely to reflect the intent of most parties and wanted to save
the parties the trouble of negotiating for employer ownership.
Alternatively, courts might have begun to see employers as
possessing a stronger moral claim and believed that any employee
who planned to assert copyright ownership ought to be forced to
disclose that intent and negotiate for it. Finally, as changing
assumptions about the nature of authorship strengthened the
rhetorical force of the employer's claim, a default rule of employer
ownership might have seemed more intrinsically appealing,
irrespective of whether the parties might negotiate around it.
Readers familiar with contemporary American doctrine on work-
for-hire would find the result quite unsurprising. In today's law,

150. 128 U.S. at 647.

151. Id. at 650.
152. Id. at 647 (emphasis added).
153. Two other cases from the 1880s contain significant dicta on the employer ownership
principle. Sarony v. Burrow-Giles Lithographic Co,, 17 F. 591 (C.C.S.D.N.Y. 1883), restated in
dicta the principle from Wheaton v. Peters that judges could not own the copyright in their
opinions and that, therefore, a reporter could not either. It did so, however, in a way that
emphasized the fact of the judges' employment rather than the public nature of statements of
law. See Wheaton, 33 U.S. at 668. Similarly, In re Gould & Co., 2 A. 886, 896 (Conn. 1885),
which preceded the Supreme Court's pro-employee decision in Callaghan, held that "[t]he
judges and the reporter are paid by the state; and the product of their mental labor is the
property of the state, and the state, as it might lawfully do, has taken to itself the copyright."
 Yale Journal of Law & the Humanities [Vol. 15:1

when the creator is an employee, the default rule is employer

ownership; when the creator is an independent contractor, the
default rule is employee ownership.'54 The reporter in Callaghan
would be characterized as an independent contractor because he was
paid by the job (rather than salaried) and, for all that appears, the
judges had no control over the manner, means, or timing of his
work.'55 But the distinction between employees and independent
contractors did not exist as clearly in American law in the nineteenth
century as it does today,'56 and, in any event, it was never mentioned
in the copyright cases. An English treatise on labor contracts, while
discussing the right of the "master" to the "produce of the servant's
labour and inventions," did not note any distinction between
servants and other employed persons. It stated:
A person who employs and pays an author to write a drama or
literary work, is not by virtue of the employment entitled to the
exclusive right of representation or copyright, because the

154. 17 U.S.C. § 101.

155. See Cmty. for Creative Non-Violence v. Reid, 490 U.S. 730 (1989) (making the hiring
party's right to control the creator's work the crucial determinant of whether the creator is an
employee or independent contractor within the meaning of copyright law).
156. The distinction between servants and contractors functioned to determine the liability
of the employer to third parties for torts committed by the servant or contractor. See 2
HORACE GAY WOOD, A TREATISE ON THE LAW OF MASTER AND SERVANT 601 (1877)
(defining a servant as "a person who, by contract or operation of law, is for a limited period
subject to the authority or control of another person in a particular trade, business or
occupation," and explaining that an employer is not liable for the acts of contractors "except
when he retains control over the manner or instrumentalities of the work"). Mechem's treatise
on the law of agency confessed that
[tihe line of demarcation between the relation of principal and agent, and that of master
and servant is exceedingly difficult to define. This difficulty arises largely from the fact
that the two relations are essentially similar. Indeed, there is much reason for saying that
the difference between them is one of degree only, and not of kind.
FLOYD R. MECHEM, A TREATISE ON THE LAW OF AGENCY 2 (Chicago, Callaghan & Co.
1889). Mechem ultimately concluded that agency
properly relates to transactions of business with third persons, and implies more or less of
discretion in the agent as to the time and manner of his performance. Service, on the
other hand, has reference to actions upon or about things. It deals chiefly with matters of
mere manual or mechanical execution, in which the servant acts under the direction and
control of the master.
Id. at 3. The distinction between "things" and "business" obviously no longer holds, although
the different degrees of supervision are today the essential distinction between employees and
independent contractors. But Mechem went on to note all the exceptions to the notion of
supervision as being determinative, and added that the agent typically works by the project
rather than for a fixed period. He then cautioned, however, that an agent's period of
employment might be fixed. Id. at 4. See generally WILLIAM EVANS, A TREATISE ON THE
LAW OF PRINCIPAL AND AGENT IN CONTRACT AND TORT (Chicago, Chicago Legal News Co.
1879) (containing a chapter on the liability of employers for injuries caused by the negligence
of fellow servants but little discussion otherwise of employment, and no discussion of the
distinction between agents and servants); FRANCIS WHARTON, A COMMENTARY ON THE LAW
OF AGENCY AND AGENTS § 321, at 227 (Philadelphia, Kay & Bro. 1876) (explaining that an
agent is paid by commission, whereas a servant is paid by wages).
2003] Fisk

statutes vest such right in the author, and require the transfer of
such a right to be in writing.'57
While it is true, as noted above, that most of the early cases
articulating a default rule of employee ownership involved workers
who today would likely be deemed independent contractors, courts
never mentioned whether the author was a servant or a contractor. 58'
Nor did courts observe the distinction when they applied the rule to
cases involving workers who today would be deemed employees. It
may very well have been the independence of the employee-author
that influenced the court to adopt a default rule of employee
ownership, and thus the intuition of early courts was similar to the
intuition underlying the modern law that independent contractors
should presumptively own the copyrights to their works. But none of
that was made explicit in the cases. Rather, the courts contended
simply that one must examine the "contract express or implied"
between the parties, without articulating the basis for interpreting
implied understandings.159

D. DoctrinalConfusion, DefaultRules, and the Role of Contracts

By the last two decades of the nineteenth century, the law of
employee copyrights was highly uncertain and the results of cases
were quite unpredictable. However, as illustrated by Callaghan v.
Myers, there remained a strong line of cases articulating pro-
employee rules." ° One might expect, therefore, that firms concerned
about copyright ownership would increasingly resort to express
contracts to achieve their ends. To some extent, the case law bears
this out: the reported cases from the 1890s involve more express
contracts than appeared in earlier cases. At the same time, as
Callaghan also illustrates, courts became more inclined to
characterize the rules regarding copyright ownership in contract
terms rather than in the terms of immutable rules and moral right
that had predominated in earlier cases. Courts thus facilitated
contracting around the default rules. The judicial endorsement of a
contractarian view of copyright ownership paved the way for
employers to gain control of copyrights more effectively. At the
same time, employers were more likely to see the need for, and to be

157. DAVID GIBBONS, RUDIMENTARY TREATISE ON THE LAW OF CONTRACTS FOR

WORKS AND SERVICES 109-10 (London, John Weale 2d ed. 1857).
158. See cases cited supranote 24.
159. WOOD, supra note 156, at 4.
160. See supra text accompanying notes 148-52.
 Yale Journal of Law & the Humanities [Vol. 15:1

able to effectively negotiate and draft, contracts allocating workplace

copyrights. 6 '
The uncertain status of the employee-author and the role of
contract concepts in creating a default rule of employer ownership
can be seen quite clearly in Eaton Drone's influential 1879 Treatise
on the Law of Propertyin IntellectualProductionsin GreatBritain
and the United States.'62 Drone's was the leading nineteenth-century
copyright treatise and was frequently cited by courts and lawyers.
His utterly ambiguous characterization of the state of the law, as
explained below, provided ammunition for employers and
employees on almost any position they might have wanted to take.
Drone began his analysis with the uncontroversial proposition that
a "literary production is primarily the property of the author who
has created it." He then elaborated on this idea: "When a person has
conceived the design of a work, and has employed others to execute
it, the creation of the work may be so far due to his mind as to make
him the author." But, cautioned Drone, "he is not an author who
'merely suggests the subject, and has no share in the design or
execution of the work."' 63 Drone then went on to explain how
authors could assign their common law rights in an unpublished
work to another and how the other would thus obtain the right to
apply for the statutory copyright. Analogizing from this principle,
Drone asserted:
Assuming that the law is rightly expounded above, to the effect
that copyright will vest ab initio in the owner of an unpublished
production, it follows that any person may secure statutory
copyright for a work which he has employed another to write.
The produce of labor may become the property of him who has
employed and paid the laborer. Literary labor is no exception to
this universal rule. When an author is employed on condition
that what he produces shall belong to the employer, the
absolute property in such production vests in the employer by
virtue of such employment and by operation of law.'"
Drone cited no cases in support of this proposition. Later, when he
came to discussing the cases, he moderated the rule considerably. As
to the law in Britain, he agreed that,

161. John Witt has argued that uncertain default rules enhance the ability of employers to
use express contracts to their preferred results. See John Fabian Witt, Rethinking the
Nineteenth-CenturyEmployment Contract,Again, 18 LAW & HIST. REV. 627, 654-56 (2000).
162. DRONE, supra note 38, at 236-61.
163. Id. at 236 (quoting Shepherd v. Conquest, 17 C.B. 427, 445).
164. Id.at 243.
2003] Fisk

it may be going too far to say that the law to this effect is
judicially settled. It is conceded that, when one person has
employed and paid another to write a work, with the mutual
understanding that it shall be the property of the employer, the
latter acquires an equitable title which will enable him in a court
of chancery to assert his rights in the published production
against either the person employed or others.165
However, said Drone, "Whether a complete legal title to the
copyright will vest ab initio in such employer without the necessity of
a written assignment, is a point on which 166
the law has not been
expressly declared by the courts of law.'
Drone then discussed an English case in which the court
articulated the rule that one employed to compose a musical score
for a play retained ownership to the copyright unless the
employment involved an express understanding that the composition
would belong to the theater owner.1 67 The case indicated that the law
in Britain, like the law in the United States during that era, found
employer ownership only on the basis of an express agreement.
Drone recognized as much when he came to discussing American
law:
The mere fact of employment does not make the employer the
absolute owner of the literary property created by the person
employed. Where there is no agreement or implied
understanding that what is produced shall belong to the
employer, it is clear that the latter acquires no title to the
copyright."
The employer acquires the copyright only if there is an "agreement
that he, and not the person employed, is to be the owner of the work
produced. But such agreement need not be express. It 69may be
implied from the terms and conditions of the employment."1
Thus, Drone's treatise could be, and was, read to support three
contradictory propositions: a default rule of employee ownership; a
default rule of employer ownership; and a rule that ownership
should be assessed in each case from the express or implied
understanding of the parties. Perhaps the confusion Drone reported
was due to the fact that default rules of copyright ownership were
doubly sticky; not only did parties infrequently contract around them

165. Id.at 248.

166. Id.
167. Id.(citing Hatton v. Kean, 7 C.B. n.s. 268).
168. DRONE, supra note 38, at 257 (emphasis added).
169. Id at 258.
 Yale Journal of Law & the Humanities [Vol. 15:1

(which is the usual sense in which scholars refer to sticky default

rules), but courts also made them hard to contract around."'
This fact is more than simply amusing; it invites us to think about
how industries premised on intellectual property rights could thrive
in a climate of legal uncertainty. If it is claimed that nineteenth-
century economic development was facilitated by the existence of
certain and definite property rules, surely contrary evidence from the
most influential copyright treatise of the era suggests that American
publishing and advertising somehow managed to grow without
definite rules.
The uncertainty of property rights and the doctrinal chaos
reported in Drone's treatise were perhaps ameliorated by the
importance of express contracts. Drone explicitly noted the
possibility of contracting around whatever default rule existed.
Moreover, the fact that most copyrighted materials in the era were
produced by people who would have entered into relatively short-
term contracts in which both parties contemplated the creation of
copyrighted works suggests that in many cases the parties probably
allocated ownership rights by contract. The extent to which parties
contracted around the default rule is unclear and probably varied
significantly by industry. 1 However, Rochelle Cooper Dreyfuss
offers two reasons to believe that it was not particularly easy to
contract around the uncertain defaults. First, the costs of transacting
might be high when the parties have to discuss something as touchy
as authorship. Employers might have been afraid to alienate
employees by demanding assignment of the copyright, preferring to
run the risk of litigation later. Employees may have lacked legal
sophistication to realize that it was necessary to contract for
copyright ownership. Second, the instability of the law may have
made enforcement of any contract they did reach highly uncertain. '
The experience of Dion Boucicault and William Stuart bears this
out. Why, when they agreed that Boucicault would write The
Octoroon for performance at the Winter Garden, did they not agree
who would own the copyright? Surely they both contemplated that
their association would produce a copyrighted work and both must
have hoped it would be valuable. Moreover, Boucicault was not

170. See Ian Ayres & Robert Gertner, Majoritarianvs. MinoritarianDefaults,51 STAN. L.
REV. 1591,1598 (1999).
171. In future work, I expect to report on an empirical study of the contracting practices of
firms that produced copyrighted information. My work in other areas of employee-generated
intellectual property, however, suggests that the default rule of employee ownership of trade
secrets remained quite sticky until the early twentieth century. See Fisk, Working Knowledge,
supra note 27.
172. See Rochelle Cooper Dreyfuss, Collaborative Research: Conflicts on Authorship,
Ownership,and Accountability,53 VAND. L. REV. 1162, 1172-79 (2000).
2003]

naive about copyright law. Perhaps they failed to negotiate explicitly

for copyright ownership for the same reasons that playwrights and
those who assist them often fail to today: There may be serious costs
to the relationship in raising those issues and/or in reducing them to
writing. The parties may each hope for a better result through
informal resolution, or they may be unsure at the outset what each of
their contributions will be and how, therefore, best to divide the
credit and profit.'73 Then as now, there may be many reasons why
even those most likely to anticipate the production of a copyrighted
work might not negotiate around the default rule.
And when there was no express contract, as Drone's treatise
recognized, the problems arose. The dominant contemporary
American legal habit of creating all sorts of implied contractual rules
defining the obligations of employer and employed was not yet
entrenched.'74 Contract surely reigned supreme, but the problem was
that Drone provided (because the cases provided) little guidance on
how to identify an implied understanding between employer and
employee for the ownership of copyrights. The concept of default
rules was less familiar than it is today in part because the notion of
employment as an entirely contractual relationship in which all
obligations are subject to negotiation was only just beginning to
emerge.
After publication of Drone's treatise, reported decisions continued
to reflect significant uncertainty about whether the default rule was
one of employer or employee ownership. As we have seen in
Callaghan, the dominant view was that the employee-author was
entitled to copyright. The Court's decision five years earlier in
Burrow-Giles Lithographic Co. v. Sarony further illustrated this
approach.'75 Burrow-Giles Lithographic Co. involved a claim to a
copyright of a photograph of Oscar Wilde. The Court held that the
photographer, not the man who hired him, had used sufficient

173. See Thompson v. Larson, 147 F.3d 195 (2d Cir. 1998) (deciding a dispute between the
heirs of the playwright and the dramaturg of Rent, the successful Broadway musical of the
1990s); Roberta Rosenthal Kwall, supra note 2 (discussing Thompson v. Larson); Paulette S.
Fox, Note, Preserving the CollaborativeSpirit of American Theater: The Need for a "Joint
Authorship Default Rule" in Light of the Rent Decision's Unanswered Question, 19
CARDOZO ARTS & ENT. L.J. 497 (2001) (same); see also Seth F. Gorman, Who Owns the
Movies? Joint Authorship Under the Copyright Act of 1976 After Childress v. Taylor and
Thompson v. Larson, 7 UCLA ENT. L. REV. 1 (1999); Jane C. Lee, Comment, Upstagingthe
Playwright: The Joint Authorship Entanglement Between Dramaturgsand Playwrights, 19
LOY. L.A. ENT. L. REV. 75 (1998).
174. Indeed, the rise of an objective theory of contract removed from the particulars of the
contracting party's status was a phenomenon of the end of the nineteenth century. It was
underway at the time of these cases, but hardly well settled. See P.S. ATIYAH, THE RISE AND
FALL OF FREEDOM OF CONTRACT 399-412 (1979); LAWRENCE FRIEDMAN, CONTRACT LAW
IN AMERICA 20 (1965); GRANT GILMORE, THE DEATH OF CONTRACT (1974).
175. 111 U.S. 53, 61(1884).
 Yale Journal of Law & the Humanities [Vol. 15:1

creativity in the composition and lighting to entitle him to a

copyright. 76' At the end of the opinion, the Court described with
approval an English case involving an employee sent to photograph
an Australian cricket team. 77' In that case, the Queen's Bench held
that the photographer, not his employer, was entitled to the
copyright because "the author is the man who really represents,
creates, or gives effect to the idea, fancy, or imagination."'78 The
Supreme Court said that the English justices' "views of the nature of
authorship and of originality, intellectual creation, and right to
protection" confirmed the Court's own views. "9 This statement was
dicta in that Burrow-Giles did not involve a dispute between
employer and employee, but nevertheless it reflects the Court's view
that the work of the creator is crucial to copyright ownership.
The one other frequently cited employer-employee copyright
dispute of the era, Peters v. Borst, also articulated a default rule of
employee ownership. The employer-the Litchfield Observatory at
Hamilton College--claimed ownership of a star catalog prepared by
the director of the observatory and his assistant. ° Noting that there
was no "agreement that any production of these parties as authors
should be the property of the observatory or the college," the trial
court insisted, "it can hardly be claimed the observatory or college
would become the owners of the work they might, as authors,
produce and publish to the world."'' The institution could look only
to the more intangible benefits that flow from having in their employ
"men who might become eminent and distinguished by reason of the
mental labor and results they achieved."'" The Court of Appeals
placed more emphasis on the fact that the Director's modest salary
was provided by an endowment and that the college was unable to
increase it:
Since the college was financially unable to pay him anything
approaching a reasonable salary, it may easily be inferred that
the director would be allowed and expected to do for himself
much work of his own, for which he would not be accountable
to the college, and which he could use or dispose of as he
pleased. '

176. 111 U.S.at 61.

177. Id.(citing Nottage v. Jackson, 11 Q.B. 627 (1883)).
178. Id.
179. Id
180. 9 N.Y.S. 789 (Sup. Ct. 1889).
181. Id.at 790.
182. Id.
183. Root v. Borst, 36 N.E. 814, 814 (N.Y. 1894).
2003]

Presuming copyright ownership by the employee because of

otherwise insufficient compensation, the court appears to have
adopted a default rule that reflected the likely expectations of the
parties. If both employee and employer are relatively sophisticated
negotiators and both contemplate that the employment will produce
copyrightable work, courts need not impose default rules to force
one side to disclose its intentions regarding copyright ownership. In
later cases, as we shall see, courts felt the need to force disclosure by
employees who intended later to claim their copyrights justified a
diametrically opposite default rule of employer ownership.
During the 1890s, while courts continued to uphold a default rule
of employee ownership, some cases enforced express contracts
allocating the copyrights variously to employer and employee. Press
Publishing Co. v. Monroe involved a poem commissioned by the
World's Fair organizing committee."8 The Second Circuit concluded
that the contract allocated the copyright to the author of the poem
and granted the committee only a license to publish and reproduce
it. 5 In Mallory v. Mackaye, by contrast, the Circuit Court for the
Southern District of New York concluded that a contract expressly
provided that a theater manager owned the copyright to a play
written by an actor in the theater."s Carte v. Evans, which involved
the efforts of Gilbert and Sullivan to obtain a U.S. copyright for The
Mikado, established the principle that an express agreement that the
employer should own his employee's literary product vested the
copyright entitlement in the employer, even in the absence of a
formal assignment. 7 This principle made it easier for employers to

184. 73 F. 196 (2d Cir. 1896).

185. 73 F. at 198-99.
186. 86 F. at 122. As the court paraphrased the contract, Mackaye "agreed to give and
devote to the service of Mallory the whole of his time and energy, as Mallory might direct, in
any of the capacities of an author, a manager, an actor, a director, or in any other capacity
having any connection with theatrical labor, and that the entire product and income of his
intellectual and physical labor and skill should belong absolutely to Mallory." Mackaye
attempted to gain control of the play he had written and copyrighted, as well as the double
stage he had developed and patented under contract with Mallory and for an accounting of the
profits Mallory received. Courts found that the contract transferred both the copyright and the
patent to the employer and, eventually, decided that Mackaye had breached the employment
agreement. These decisions absolved Mallory of the obligation to account for the profits or to
continue paying Mackaye under the contract. See Mackaye v. Mallory, 6 F. 743 (C.C.S.D.N.Y.
1881); Mackaye v. Mallory, 12 F. 328 (C.C.S.D.N.Y. 1882); Mackaye v. Mallory, 80 F. 256
(C.C.S.D.N.Y. 1897); Mallory v. Mackaye, 86 F. 122 (C.C.S.D.N.Y. 1898); Mallory v. Mackaye,
92 F. 749 (2d Cir. 1899).
187. 27 F. 861 (C.C.D. Mass. 1886). The case held that Gilbert & Sullivan's contract with
Tracy to compose a piano arrangement of the orchestral score of The Mikado and then to
copyright the piano arrangement in the United States obviated the need for a formal
assignment of the U.S. copyright:
The proprietor as well as the author is entitled to enter the work for copyright. The
consent of Tracy was sufficient to constitute [the registered copyright holder] the
 Yale Journal of Law & the Humanities [Vol. 15:1

obtain copyright protection without the cooperation of the creative

employee and without the need to seek an order compelling the
recalcitrant employee to assign the copyright.
The doctrinal uncertainty about the nature of the default rules
raises the question of their role in actually structuring relations
between creative employees and firms. As John Witt has pointed
out, there is every reason to believe a huge gap existed between
employment doctrines and actual practices. Consequently, we cannot
be certain about the extent to which parties contracted around
defaults.'" It may have been especially difficult for nineteenth-
century employees and firms to contract for the allocation of
copyrights, even if both had understood the need to do so. As
Rochelle Cooper Dreyfuss has explained, it can be exceptionally
difficult for collaborators in innovation to agree in advance upon an
allocation of credit and profit, since the likelihood of profit, the
nature of profit, and even the contributions that collaborators might
make are highly uncertain. 89 Negotiating over the terms of the
association may be especially painful at an interpersonal level, as it
involves an attempt to value and assign creativity. Given the
complexity and uncertainty of ex ante valuation and the very high
transaction costs, it may be that few even attempted to contract
around the default rules allocating copyrights in the workplace.
Apart from questions about the effect of defaults on employment
practices, there are separate questions about what courts thought
they were accomplishing by establishing default rules. In some cases,
courts might have adopted a default rule of employers' rights simply
because they deemed it more likely to reflect the actual
understanding of most employers and creative employees
("majoritarian defaults"). In other cases, they may have wanted to
force employees who planned to claim copyright to disclose that9
intent and negotiate for it in advance ("penalty defaults"). 1 0
Boucicault v. Fox seems to have imposed a penalty default, as the

proprietor for this purpose, without a formal assignment. The effect of the transaction
was the same as if Tracy had made the entry in his own name and then assigned to [the
plaintiff].
Id. (citing Lawrence v. Dana, 15 F. Cas. 26 (C.C.D. Mass. 1869). Another case enforcing an
express contract giving a publisher the entire "right, title, or interest" in books written by an
employee was Clark v. West, 86 N.E. 1 (N.Y. 1908), which involved law books published by
West.
188. Witt, supra note 161, at 635-50.
189. Dreyfuss, supra note 2, at 1173-74.
190. See Ian Ayres & Robert Gertner, Filling Gaps in Incomplete Contracts: An
Economic Theory of Default Rules, 99 YALE L.J. 87, 91 (1989) (introducing the concept of
penalty defaults); Ayres & Gertner, supra note 170; Frank H. Easterbrook & Daniel R.
Fischel, The Corporate Contract, 89 COLUM. L. REV. 1416, 1444 (1989) (describing default
rules as those "that almost everyone will want to adopt").
2003] Fisk

court explained that it expected employers to contract expressly for

copyright ownership when they deemed it possible and desirable.'
In still other cases, however, such as such as Peters v. Borst, the
tenor of the court's opinion suggests that it believed a rule of
employee ownership genuinely reflected what the parties likely
understood, or at least should have. 2
The change in the default rule around 1900 may reflect both a
changing judicial perception of the likely intent of the parties and a
changing judicial perception of which party-the employer or the
employee-likely has the best information about its future plans to
claim the copyright, and which party should, thus, be forced to
disclose its intent under penalty of losing the copyright. Whether is
the ensuing rule is best described as a penalty default or an "off-the
rack" default, the change seems to reflect a changing judicial
perception of whether employers or employees ought to be able to
control the copyright. This shift began gradually after 1860 and
rapidly accelerated after 1900.

IV. THE FOUNDATIONS OF THE MODERN ERA: 1900-1910

The principle of employer ownership of copyrights in employee

works was established in the first decade of the twentieth century.
The ambivalence of earlier courts about whether the employer's
right arose only from an express, or relatively clear, understanding
between the employer and employee disappeared. Instead, it became
quite clear in cases, and was finally enacted by statute, that the
employer was entitled to the copyright as a matter of law unless
there was an express agreement to the contrary. In other words,
where earlier courts had attempted to find an agreement to
employer ownership implied in the facts, twentieth-century courts
found such agreements implied by operation of law in every
employment relationship in which an employee created copyrighted
work during the scope of his or her employment.

A. Corporationsand Creativity
The ultimate legal fiction underlying modern copyright law is the
fiction of corporate authorship. If the fiction were merely a
shorthand way of saying that the corporation is the assignee of the
works of an author, it would be one thing. But not every case that
recognized employer ownership did so simply by assuming that the
employee had expressly or impliedly agreed to assign a copyright.

191. See supra text accompanying notes 100-15.

192. See supra text accompanying notes 180-83.
 Yale Journal of Law & the Humanities [Vol. 15:1

Some courts persisted in analyzing the matter as if the employer

itself had to be the creator. In this context, the rise of the corporation
played an ambiguous role. On the one hand, if judges saw authorship
as a collaborative process, the fact that a business entity was a
corporate body helped the employer's case. The claim of a corporate
"author," composed of many different people working toward a
common end, had greater rhetorical appeal than the claim of one
individual partner to the work of another. On the other hand, the
rise of corporate power threatened the very individualist premises
and values of much nineteenth-century law. Inasmuch as courts
regarded authorship as uniquely individual, corporate authorship
was simultaneously oxymoronic and repugnant. Courts did not
immediately capitulate to the empire-building tendencies of the
growing corporations in this area any more than they did in any
other area of law. One way to understand the schizophrenic state of
the law of ownership of ideas is as a working-out of the tension
between individualism and corporatism in American law and society.
The first case in which a court explicitly held that a corporation
owned a copyright in an employee's creation did not require
significant departure from past authorship rhetoric because the
president of the corporation, "himself an artist of respectable
attainments," had been personally involved in the design of the
advertising woodcut at issue. 93 The employee allegedly painted the
print under the supervision of the corporation's president. 9 The
corporation's "money paid for the painting; its artist colored it; its
president designed it, his was the 'originating, inventive, and master
mind."' 195 Analogizing the corporation to the artist's studio, the
court said:
The fact that the artist Stecher [an employee of the corporation]
executed Schumacher's design cannot defeat the copyright. The
sculptor seldom touches the marble from which his statues are
carved. The fact that the brush which embodied Schumacher's
idea was held by another artist rather than by himself cannot be
important in considering a question of this character.'96
The personal involvement of the corporation's president made the
step to the fiction of corporate creation seem to be nothing more
than an easy step of agency law attributing to the corporation the
acts of its chief officer.'

193. Schumacher v. Schwencke, 25 F. 466,466 (C.C.S.D.N.Y. 1885).

194. Id.
195. Id. at 468.
196. Id.
197. Several years later, however, the court regarded corporate authorship as
2003] Fisk

At the same time that courts were first contemplating corporate

authorship, they also were grappling with growth in the types of
materials subject to copyright. What makes the rhetoric in the cases
particularly interesting is the two levels on which the law was
changing. On the one hand, courts were invoking eighteenth- and
nineteenth-century notions of Romantic authorship to justify the
existence of copyrights in new media such as advertising. By
analogizing these commercial and corporate creations to the great
artistic works of the past, judges and lawyers legitimated new
property rights in new media. Schumacher was "an artist of
respectable attainments," and his involvement as corporate president
was akin to the master sculptor in his studio who "seldom touches
the marble."' 98 Yet, at the same time, courts alternately invoked and
ignored the great authors and great painters of the past to justify
corporate ownership of the creative works of employees.
In a sense, the core problems of copyright law-what it is that a
copyright protects and why we protect it-underlie the difficulty in
allocating authorship between employer and employee. Deciding
what kinds of materials could be treated as intellectual property was
a new and highly controversial exercise for much of the nineteenth
century. It became even more difficult as the subjects of copyright
protection expanded from books into new media. Questions
naturally arose: is it the idea or the expression that is protected?
What is it about a particular work that makes it appropriate to grant
a monopoly over its use? If rewarding and incentivizing creativity
were crucial in justifying the existence of copyrights at all, it became
all the more important to identify and lionize the originator.
Attribution of authorship is another facet of the same. To the extent
that property rights are justified by the moral superiority of the
individual artist, corporate authorship is troubling. But to the extent
that intellectual property rights exist to encourage investment in
intellectual endeavor, corporate authorship is essential.

unproblematic even without evidence of corporate officers' involvement in the acts of creation.
In 1908, an author employed to write a book on the law of corporations challenged
enforcement of a contract assigning the entire interest in the manuscript to the employer. He
argued, among other things, that the defendant had breached the contract by causing the book
to be copyrighted in the name of a corporation that was not a party to the contract. The New
York Court of Appeals rejected the claim. Clark v. West, 86 N.E. 1 (N.Y. 1908).
198. Schumacher, 25 F. at 468; cf Nat'l Cloak & Sit Co. v. Kaufman, 189 F. 215, 216
(C.C.M.D. Pa. 1911) (holding that a corporation owned the copyright to a fashion catalogue
because it "made large outlays and expenditures employing in the preparation of its various
component parts artists and authors of peculiar skill and ability ...[and] the illustrations
forming component parts of said book were the work and embodied the personal reaction of
artists of recognized skill in their calling, and were pictures of artistic merit, and, in addition to
their merit as artistic productions, were of peculiar value as portraying original conceptions
and creations relating to wearing apparel, of great interest to a large proportion of the
public").
 Yale Journal of Law & the Humanities [Vol. 15:1

This difficulty of reconciling corporate ownership and individual

artistic expression is evident in Justice Holmes's opinion for the
Supreme Court in Bleistein v. Donaldson Lithographic Co.'9 The
Court held that three chromolithographs prepared by employees of a
corporation for use as advertisements were proper subjects for
copyright protection. Holmes justified copyright protection on the
basis of the artistic genius and the uniqueness or singularity of the
"personality" or self-hood of the artist, but he found corporate
ownership of copyright to be entirely unremarkable.2"° In
determining that a lithograph for use as an advertisement was the
sort of creative work that should be accorded copyright protection
even though it was intended to be a faithful representation of actual
persons, Holmes wrote a paean to the individuality of artistic genius
quite at odds with the previously asserted facts of corporate creation
and control. If the law were otherwise, it "would mean that a portrait
by Velasquez or Whistler was common property because others
might try their hand on the same face."2 1" Holmes insisted that the
pictures were
the personal reaction of an individual upon nature. Personality
always contains something unique. It expresses its singularity
even in handwriting, and a very modest grade of art has in it
something irreducible, which is one man's alone. That
something he may copyright unless there is a restriction in the
words of the act.2 2
Ultimately, the fiction that the employer is the author allowed
Holmes to elide the question of how a corporation could be entitled

199. 188 U.S. 239 (1903).

200. Id. at 248.
201. Id.at 249.
202. Id.at 248. Justice Harlan's dissenting opinion asserted that an advertisement could
not be copyrighted because it was not "art" and lacked sufficient "connection with the fine arts
to give it intrinsic value." Id.at 252 (Harlan, J., dissenting). Holmes rejected Harlan's view of
the proper scope of copyright law:
It would be a dangerous undertaking for persons trained only to the law to constitute
themselves final judges of the worth of pictorial illustrations, outside of the narrowest
and most obvious limits. At the one extreme, some works of genius would be sure to miss
appreciation. Their very novelty would make them repulsive until the public had learned
the new language in which their author spoke. It may be more than doubted, for
instance, whether the etchings of Goya or the paintings of Manet would have been sure
of protection when seen for the first time. At the other end, copyright would be denied to
pictures which appealed to a public less educated than the judge. Yet if they command
the interest of any public, they have a commercial value,--it would be bold to say that
they have not an aesthetic and educational value,-and the taste of any public is not to
be treated with contempt. It is an ultimate fact for the moment, whatever may be our
hopes for a change.
Id.at 251.
2003] Fisk

to copyright an advertisement if the justification for the

203
copyright is
"the personal reaction of the individual upon nature.
Holmes' opinion in Bleistein is a lovely illustration of his insight
about the essential characteristic of the common law method.2 ' He
invoked the traditional notion of authorship as artistic
creation-"the personal reaction of an individual upon
nature"-while giving it an entirely new significance: corporate
authorship of advertisements. To expand copyright protection into
new media, he equated "the etchings of Goya" and "the paintings of
Manet" with commercial advertising. He did so precisely because
giving new content to the old form of authorship was an effective
rhetorical strategy to expand copyright protection. By eliding the
distinction between the legal fiction of corporate authorship and the
fact of collaborative creation in a corporate setting, and by effacing
the (ever-shrinking) difference between advertising and art, Holmes
made a significant change in doctrine that seemed to be nothing
more than a simple analogy.

B. The Work-for-Hire Doctrineinthe Courts

The first cases in which courts found that the fact of employment
or the act of commissioning a work, without more, entitled the
employer to the copyright were decided in 1899 and 1900,
respectively. In neither case did the court acknowledge the
difference between the precedents, all of which involved a
contractual allocation of copyright, and the instant situations where
no contract existed.
In Collier Engineer Co. v. United Correspondence Schools, a
salaried employee had the job of preparing and revising instructional
materials for a correspondence school.0 5 In subsequent employment,
the employee-author sought to write similar materials. Although the
court declined a preliminary injunction, finding it unclear from the
evidence whether the new materials infringed because both the new
and the original materials were compilations drawn from other
sources, the court nevertheless stated without elaboration that the

203. Another court did not bother to justify it, simply relying on Schumacheras controlling
authority. See Mutual Adver. Co. v. Refo, 76 F. 961, 963 (C.C.D.S.C. 1896).
204. See O.W.HOLMES, JR., THE COMMON LAW 5 (Dover photo. Reprint 1991) (1881)
("The customs, beliefs, or needs of a primitive time establish a new rule or a formula. In the
course of centuries the custom, belief, or necessity disappears, but the rule remains. The reason
that gave rise to the rule has been forgotten, and ingenious minds set themselves to inquire
how it is to be accounted for. Some ground of policy is thought of, which seems to explain it
and to reconcile it with the present state of things; and then the rule adapts itself to the new
reasons that have been found for it, and enters on a new career. The old form receives a new
content, and in time even the form modifies itself to fit the meaning which it has received.").
205. 94 F. 152 (C.C.S.D.N.Y. 1899).
 Yale Journal of Law & the Humanities [Vol. 15:1

employer was entitled to the copyright on the materials. 2° The court

acknowledged the possibility that the employee would be unduly
constrained from using the knowledge he had acquired about
teaching and so attempted to find a middle ground:
[A]lthough Ewald was not at liberty to reproduce so much of his
work as had been copyrighted by the employers for whom it was
prepared, even by availing of his recollection of the contents of
the copyrighted pamphlets, he was not debarred, after his
contract terminated, from making a new compilation, nor from
using the same original sources of information, nor from
availing of such information as to the needs of students and the
best methods of getting in mental touch with them as he may
have acquired while superintending complainant's school. °7

Gone was the focus on the author's entitlement to the fruits of his
creative genius. Instead the court focused on the task of
accommodating employer ownership of intellectual property with
some measure of employee freedom to use his knowledge in
subsequent employment.
The next year, in Dielman v. Wite, °8 a court recognized the right
of the person who commissions a work to own the copyright.
Dielman had received the commission to design a mosaic for the
Library of Congress. He painted a sketch of the mosaic, copyrighted
it, and then sent it to Venice, where the mosaic "was manufactured
in the usual manner by workers in mosaic. '' 2' Both the sketch and
the mosaic bore the notation: "Copyright, 1896, by Frederick
Dielman. '212 Nothing was said in the correspondence between
Dielman and the Library of Congress about copyright ownership.
When officials of the Library published photographs of the mosaic,
Dielman sought an injunction against their publication, which the
court denied:
In general, when an artist is commissioned to execute a work of
art not in existence at the time the commission is given, the
burden of proving that he retains a copyright in the work of art
executed, sold, and delivered under the commission rests
heavily upon the artist himself. If a patron gives a commission to

206. Id.at 153.

207. Id. at 153.
208. 102 F. 892 (C.C.D. Mass. 1900).
209. Id. at 893. This characterization of the contribution of the Italian mosaic workers
raises the question whether anything that they did was sufficiently creative as to vitiate the
claim of either of the American contenders to be the true "author" of the mosaic. The court
did not address that question, revealing that its ability or willingness to understand the truly
collaborative nature of the creation was not without limits.
210. Id.
2003] Fisk

an artist, there appears to me a very strong implication that the

work of art commissioned is to belong unreservedly and without
limitation to the patron."'
The court recognized the constraints such a rule might impose on
an artist's ability to make further use of the design, but concluded
that those issues were matters "of artistic ethics rather than of
law. 212 What had been a legal rule only a generation before had
2
been downgraded to the status of "artistic ethics. 1
After these cases, courts more readily saw employees who
produced copyrighted materials as employees hired to write rather
than as independent creators whose individuality was reflected in
their creation. In Edward Thompson Co. v. American Law Book
Co., which concerned the copyright to multi-volume encyclopedias
of English and American law, the court deemed it "unnecessary, as it
might be impracticable" to identify the actual authors of the work,
since "[i]t sufficiently appears that complainant's publication is the
result of the intellectual labor of the editors and compilers employed
by the complainant."" 2 '

Similarly, in National Cloak & Suit Co. v. Kaufman, the court

easily accepted the proposition that a corporation could copyright a
fashion catalogue prepared by its employees." ' The basis for the
corporation's copyright protection was that the corporation
"exercised the most careful supervision and discrimination and made
large outlays and expenditures."2 6' The court did not find artistic or
literary creativity irrelevant; it only decided that the corporation was
entitled to the credit. The corporation employed "artists and authors
of peculiar skill and ability," and the pictures and text,
embodied the personal reaction of artists of recognized skill in
their calling, and were pictures of artistic merit, and... were of
peculiar value as portraying original conceptions and creations

211. Id.at 894.

212. Id.
213. Eventually, courts distinguished the rights of employees from those of independent
contractors and backed away from the presumption of employer ownership articulated in Dielman v.
White. In Edward Thompson Co. v. Clark, 109 N.Y.S. 700 (Sup. Ct. 1904), and W.H Anderson Co. v.
Baldwin Law Publishing Co., 27 F.2d 82 (6th Cir. 1928), both of which involved publishers of law
books, the courts distinguished the rights of independent contractors from those of employees. The
Sixth Circuit wrote:
[W]here a contract of employment is silent, there may be an implication in favor of the
employer. But in the present case plaintiff was an independent contractor, rather than an
employee; moreover, it may properly be inferred that the parties did not intend plaintiff
to surrender a copyright in consideration of a sum less than the bare cost of the work.
Id. at 88.
214. 119 F. 217,219 (C.C.S.D.N.Y. 1902).
215. 189 F. 215 (C.C.M.D. Pa. 1911).
216. Id.at 216.
 Yale Journal of Law & the Humanities [Vol. 15:1

relating to wearing apparel, of great interest to a large

proportion of the public on account of the originality and
exercise of trained aesthetic faculties displayed in said
illustrations. 7
Finally, in a few cases decided just before the enactment of the
1909 Copyright Act, courts enforced express contracts allocating to
employers the copyright in law books prepared by employees. The
courts rejected the employees' contentions that they were entitled to
credit as authors or to control the corporation's subsequent
alterations of their work. The cases thus made clear that employers
of works made for hire pursuant to express contracts owned all the
rights to the works, just as if the corporations were the authors
themselves.2 1

C. The Enactment of a Work-for-Hire Doctrinein the 1909

Copyright Act
The 1909 revision of the copyright law made concrete, as well as
catapulted forward, a change that had just begun in the case law. The
1909 Copyright Act added a provision stating that employers were
the authors of works made by employees. 19 The decision to label the
employer an "author," rather than create a default rule of implied
automatic assignment, appears to have been based on three
considerations. First, it was a matter of ease in statutory drafting
("author" is a term of art used throughout the statute). Second, it
avoided constitutional doubts about a default rule of employer
ownership stemming from the constitutional provision that Congress
may give "authors" a copyright.2" Third, and most importantly, the
drafters of the revision wanted to be sure that the employer would
be the initial copyright owner rather than an assignee, because only
the initial owner is entitled to obtain a renewal.2 ' Of course,
Congress could have achieved the same result by changing the law
regarding who may apply for a renewal, but that option was not

217. Id.
218. See. e.g., Chamberlayne v. Am. Law Book Co., 163 F. 858 (C.C.E.D.N.Y.), rev'd, 165
F. 313 (2d Cir. 1908) (holding that by assigning a copyright by express contract an employee
lost his right to sue for "trespass to literary property" for alterations to an article he wrote for
inclusion in an encyclopedia); Jones v. Am. Law Book Co., 109 N.Y.S. 706 (N.Y. App. Div.
1908) (holding that an express contract assigned a copyright to an employer and that therefore
the employee cannot insist that he be listed as an author of the work).
219. Act of March 4, 1909, ch. 320, 35 Stat. 1075.
220. U.S. CONST. art. I, § 8 empowers the Congress "[t]o promote the Progress of Science
and useful Arts, by securing for limited Times to Authors and Inventors the exclusive Right to
their respective Writings and Discoveries."
221. See 1 LEGISLATIVE HISTORY OF THE 1909 COPYRIGHT ACT 56 (B. Fulton Brylawski
& Abe Goldman eds., 1976).
2003] Fisk

considered, perhaps because it was deemed too big an incursion into

the public domain.
The work-for-hire provision originated in a pair of conferences
that Herbert Putnam, the Librarian of Congress, convened in 1905 at
the request of the Chairman of the Senate Committee on Patents.
Putnam invited representatives from a number of industries and
from the American Authors' League to discuss the need for and the
desirable terms of a revised copyright law. The early version of the
draft bill presented at the first conference stated that only "authors"
could obtain copyrights.222 In a series of discussions on that provision,
various representatives of publishing and lithographic industries
advocated recognizing the right of the employer to obtain copyright
under the statute. One Samuel J. Elder of Boston urged that
publishers of encyclopedias and other works requiring the assistance
of a large number of people needed some method other than
individual assignments to obtain effective ownership of the copyright
to the complete project. The problem, he explained, was that only
the "author" of a work, and not the "proprietor" who might own the
copyright because of an assignment by the author, could obtain a
renewal of the copyright after the expiration of the original term.223
To renew the copyright to an encyclopedia, the publisher would
"have to go searching all over the world from widows and legitimate
children, and the search is so great that the renewal term can hardly
'224
be obtained. Elder noted, however, that any revision to enable
publishers to obtain renewals would be "confined by the language of
the Constitution to authors and by the broadening of the idea that
the courts have given it to the assignees of authors. ' 225 He thus
seemed to suggest that employers must be defined in the statute as
the "authors" of the works in order to obtain a renewal term.

222. 2 id. at xxiv.

223. 1 id. at 56.
224. Id. Renewal rights were inheritable but not readily assignable. Prior to the 1909 Act,
the renewal right was given to the author, if living, or to his spouse and children if the author
died prior to the commencement of the renewal term. Act of Feb. 3, 1831, ch. 16, § 2, 4 Stat.
436. See generallySeymour M. Bricker, Renewal and Extension of Copyright, 29 S. CAL. L.
REV. 23, 24 (1955) (summarizing provisions of the 1831 Act). The 1909 Act included a similar
provision. It did not prevent authors from assigning future rights to renew during the original
copyright term. Such an assignment, however, granted only an expectancy, the fulfillment of
which depended upon the author's surviving to the end of the initial term. If the author did not
survive, the renewal right returned to the author's statutory successors. See Miller Music Corp.
v. Charles N. Daniels, Inc., 362 U.S. 373 (1960). The rationale for the restriction on renewal
was to protect authors and their dependents from improvident bargains; the compulsory
bequest that favored the author's heirs over assignees was an additional protection to allow the
author's family to renegotiate the value of the assignment when the value of the copyright was
known. Daniel A. Saunders, Comment, Copyright-s Broken Rear Window: An Appraisal of
Damage andEstimate of Repair,80 CAL. L. REV. 179, 185 & nn.23-27 (1992).
225. 1 LEGISLATIVE HISTORY OF THE 1909 COPYRIGHT ACT, supra note 221, at 54.
 Yale Journal of Law & the Humanities [Vol. 15:1

Robert Underwood Johnson, Secretary of the American Authors'

Copyright League, objected to the notion that an employer should
be deemed the "author" of works created by employees. He
proposed that the statute create two categories: "authors" and
"proprietors." Each term should be given "the most inclusive
definition possible," but he insisted that "a man who gets up a
cyclopedia and contracts with other people to write for him" was
only "the proprietor and ought to be considered the proprietor, and
not the author."226
' Underwood's objection was based on the symbolic

value of the term "author," rather than on the notion that the
encyclopedia publisher was not entitled to copyright, or even to a
renewal term.
Based on the comments made during the first few days of the
conference, the Copyright Office drafted a revised bill. It did not
have a general work-for-hire provision. Rather, it listed in separate
sections the various persons who would be entitled to claim a
copyright. Among them was the "publisher of a composite or
collective work (a 'series,' a 'library,' or an encyclopaedia) which has
been produced at his instance and expense."227 In a separate section,
the draft bill stated:
If a person employs another for valuable consideration to make
a portrait, that is to say a work whose principal object is the
likeness of any person in any form of art, the copyright in such
portrait shall belong to the employer as if he were the author, in
the absence of any agreement in writing to the contrary.228
The discussion on the revised draft circulated by the Copyright
Office revealed that publishers and employers remained dissatisfied.
One complained:
We have people who work for us who make engravings or
etchings for us under salary. Under the new law--if it becomes a
law as drafted-they would have the right to copyright, and I
think it would be well to express in such a law that where no
agreement exists to the contrary the payment of a salary to an
employee shall entitle an employer to all rights to obtain a
copyright in any work performed during the hours for which
such salary is paid. It seems to me these things should not be left
to the courts to decide. 29
This comment reflected a number of concerns. It reflected a

226. Id. at 56.

227. 2 id. at xxiv.
228. Id.
229. Id.at 65.
2003] Fisk

perception that the extant law did not operate as a default rule that
could be modified by contract since courts would not enforce
contracts for employer ownership. It suggested that courts were too
unpredictable in what they would find sufficient. It also suggests that
employers found it too onerous to have to contract for employer
ownership. In other words, this was one of those circumstances in
which the initial entitlement to copyright mattered because it would
be too expensive or difficult to contract for any other allocation.
Employers may also have wanted a default rule of employer
ownership because they thought employees were unlikely to realize
the rule and would need to contract for their copyrights. As John
Witt has suggested, the uncertainty of the default rule may have
systematically advantaged employers "who were able to spread the
cost of legal expertise and contract negotiation across multiple
employment contracts."" 23 Whatever the
reason, employers plainly
realized that a default rule of employer ownership would be to their
advantage.
Richard Bowker of the American Authors' Copyright League
proposed alternative language for the definition of author to include
not only writers, composers, and painters, but also "the conductors
of a periodical, the joint authors of a collaborative work, a corporate
body with respect to the publications of such corporation, and a
person or persons at whose instance and expense a composite work
is produced."231
The representative of the Lithographers' Association and the
Reproductive Arts Copyright League argued that "the case of
manufacturers" should be distinguished from the case of other artists
and writers:
Now, it seems to me that we have forgotten that, and that the
right belonging to that artist who is employed for the purpose of
making a work of art so many hours a day, or that literary
producer who is employed for so many hours, should be very
different from the right that is held by the independent artist or
man who makes a painting for art's sake. I would suggest,
therefore, that in this clause there should be an exception that
this does not apply, where contractual relations exist between
the owner and the artist.2
In response, it was suggested that "an exception [be] made in
233
behalf of the person at whose expense such works were made.

230. Witt, supra note 161, at 654.

231. 2 LEGISLATIVE HISTORY OF THE 1909 COPYRIGHT ACT, supranote 221, at 143-44.
232. Id. at 188.
233. Id.
 Yale Journal of Law & the Humanities [Vol. 15:1

After that conference, the Librarian of Congress and the

Copyright Office circulated a new draft bill. This version contained a
new section 21 defining who may obtain a copyright. The provision
stated:
That the author of any work made the subject of copyright by
this Act, or his executors, administrators, or assigns, shall have
copyright for such work under the conditions and for the term
specified in this Act, including: -
The author of any original literary, dramatic, artistic, musical,
cartographical, geographical, or other similar production:

An employer, in the case of a work produced by an employee

during the hours for which his salary is paid, subject to any
agreement to the contrary.?"
Finally, in December of 1906 joint hearings were held before the
House and Senate Committees on Patents on S. 6330 and H.R.
19853, which included the results of the Librarian of Congress's
conferences. The version of the work-for-hire provision that was
enacted into law appeared in those bills. It was section 63, which
read, in full:
That in the interpretation and construction of this act "the date
of publication" shall in the case of a work of which copies are
reproduced for sale or distribution be held to be the earliest
date when copies of the first authorized edition were sold or
placed on sale; and the word "author"
35
shallinclude an employer
in the case of works made for hire.
This statutory definition of the author as including the employer in
the case of "works made for hire" has been in the statute ever
since.236
It would be too simple to say that the drafters of the 1909 Act
simply codified what was already clear in the cases, or that the cases
changed their tune as soon as the statute was enacted. Rather, both
the courts and the drafters seemed to arrive at the same conclusion

234. 3 id. at xxx.

235. 4 id. at pt. J.
236. See Borge Varmer, Works Made for Hire and on Commission, Study No. 13,
Copyright Law Revision, Studies Prepared for the Subcommittee on Patents, Trademarks, and
Copyrights of the Committee of the Judiciary, U.S. Senate, 86h Cong., 1" Sess. 128 (1960)
(noting that the provision of the Copyright Act stating that "the word 'author' shall include an
employer in the case of works made for hire" was enacted in 1909 "and represented the first
legislative attempt to deal with the relationship of employer and employee in the field of
copyright").
2003] Fisk

at roughly the same time, and evidenced similar sorts of practical

concerns as they changed the law. Moreover, neither the courts nor
the drafters described their efforts as creating a new rule. Rather,
both seemed to think they were simply stating what a sensible
analysis of what the employment contract already was.

V. CONCLUSION
Throughout the nineteenth century, the rights of employee authors
(like those of employee inventors) received considerable respect
from courts that were not otherwise inclined to grant significant
protections to employee rights. From the Supreme Court's widely
cited decision in Wheaton v. Peters until 1860 (and in some cases
beyond), virtually every court that confronted the issue determined
that, as a default rule, employees who produced copyrighted work
owned the copyright, even if they did so in the scope of their
employment. Around 1860, courts began to recognize employer
ownership, either through express contracts as a matter of equity, or,
eventually, by discerning in the employment relationship the
existence of implied contracts to the same effect. In the first decade
of the twentieth century, the default rule became the exact opposite
of what it had been in the last century, and employers became
entitled to their employees' creative works in the absence of an
agreement to the contrary.
It had been well understood for hundreds of years that employers
had a property interest in the fruits of their employees' labor.
Employers owned the goods they produced, the minerals they
extracted, the crops they harvested. Why were creative works
different? And why did the rule flip 180 degrees over the course of a
few generations?
One task of this article has been to illuminate the many reasons
why creative works were different. They were different because they
were creative and because the rhetoric of authorship carried special
force on the facts of the cases the courts confronted that it did not
carry with other forms of property. They were different because the
employees who generated them were not servants. They were
different because courts thought that employee ownership was a
principle that most likely reflected the actual understanding between
the parties and, even if they did not, courts were prepared to punish
the employer who failed to negotiate explicitly for copyright
ownership. And, finally, they were different because employee
ownership was most likely to produce the greatest production of
copyrighted material that courts cared about most: law books.
Toward the end of the nineteenth century and the early twentieth,
however, things had changed. The rise of the corporation made
 Yale Journal of Law & the Humanities [Vol. 15:1

employer control of all intellectual output of the persons in their

employ more logical and more rhetorically appealing. One could say
a corporation was the "author" of materials produced by a group of
persons in its employ in a way that one could not say that Dion
Boucicault's business partner was the "author" of a play that both of
them conceded Boucicault had in fact written. The kinds of materials
that were subject to copyright had expanded to include more
materials prepared in a collaborative way in a corporate setting. It
became apparent that employee ownership of copyrights thwarted
the publication of encyclopedias by making it difficult to obtain
renewals. Workers who created copyrighted material could more
easily be subsumed into the class of employees (not servants). The
rise of a unified law of employment, and newly uniform implied
rights and obligations, also favored employers.
The development of the work-for-hire doctrine was not inevitable.
If American law had recognized moral rights as French law does, it
might have been more difficult to imagine how the corporation could
acquire all the rights to the employee's works. One of the drafters of
the 1909 Act expressed precisely this concern: He worried that
employer ownership might allow a firm to alter and degrade a work
after its creation and injure the reputation of the individual
employee who was known to have been its creator. 37 Even as late as
1909, those with reservations about a wide-ranging work-for-hire
doctrine might have won the day. Conversely, if more map cases and
fewer theater and law publishing cases were litigated at mid-century,
perhaps the doctrine would have been settled in favor of the
employer much earlier. Even a rule that today seems intuitively
obvious to many might have taken a turn in a different direction at a
number of points in the past.
This episode of doctrinal change suggests a larger point about the
crucial importance that legal fictions and the common law method
play in mediating profound legal change. The cases are rife with
tension between crucial competing policies: the power of the
individual and the felt imperatives of economic development, the
ideology of individual authorship, and the premises of master and
servant law. The basic conflict between individual and corporate
rights was obscured in vague discourse about the nature of legal
rules. Courts recharacterized the nature of authorship, filling an
empty, or perhaps a very malleable, vessel like "authorship" with an
evolving meaning. The old form received new content, even as the
rhetorical force of the form carried on. Given the force of the
concept of "authorship," it was especially important that the courts

237. 3 LEGISLATIVE HISTORY OF THE 1909 COPYRIGHT AcT, supra note 221, at 340.
2003] Fisk

use the old form rather than reject it and invent a new one, because
the old form legitimated a profound change in property rights at a
time when the property rights of the individual were near and dear
to the hearts of judges and the accumulating power of corporations
was viewed with suspicion.
Though this is not the place to assay a theory of jurisprudence and
the common law method, a few preliminary thoughts must have their
say. I think it was the extraordinary case, not the ordinary one, which
drove the development of the law."3 In Wheaton v. Peters and
Boucicault v. Fox, the employee's claim to authorship was
compelling. The absence of an applicable overarching legal category
that directed an opposite result (neither Wheaton nor Boucicault
was a "servant" and there was not yet a unified law of "employees")
made it logically possible to create a rule of employee authorship
without regard for the old master-servant doctrine that would place
ownership of the servant's work product in the hands of the master.
True to what Llewellyn called the "Grand Style," the Justices or
judges felt free to create a rule of employee authorship because there
was not a contrary rule and because it suited their philosophical
inclination (in both cases) and their instrumental goals (especially in
Wheaton). Thereafter, judges felt themselves sufficiently restrained
by the dictum of Wheaton and the holding of the two Boucicault
cases to rule in most cases that employees owned the copyrights to
their works. But they were not constrained entirely, as illustrated by
Llewellyn's "selection of available impeccable precedent techniques"
for distinguishing precedent.239 In Keene v. Wheatley, the court could
create a new equitable principle of employer ownership because
copyright rules were not at issue--see Llewellyn's principle No. 19
(the prior rule is kept from application because its reason does not
fit); No. 35 (the case falls outside the rule); and No. 54 (the case
establishes a new concept)."
When Congress administered the coup de grace in the 1909
Copyright Act, one could plausibly claim that the change was simply
a small one of statutory drafting, not a large one divesting creative
employees of the fruits of their intellectual labor. But since the kind
of works subject to copyright had begun to include things less

238. On the question whether the ordinary case or the extraordinary one is the engine of
common law change, see BENJAMIN CARDOZO, THE NATURE OF THE JUDICIAL PROCESS 99
(1921), who identifies as factors in common law change "the force of logical consistency," the
"gradual breaking down [of doctrines] before the demands of practical convenience in isolated
or exceptional instances," and "the generative force of the exceptions as a new stock." See also
KARL LLEWELLYN, THE COMMON LAW TRADITION: DECIDING APPEALS (1960); ROSCOE
POUND, INTERPRETATIONS OF LEGAL HISTORY (photo. reprint 1986) (1923).
239. LLEWELLYN, supra note 238, at 77-91.
240. Id.at 81, 85, 89. •
70 Yale Journal of Law & the Humanities [Vol. 15:1

intellectual, and as the replacement of small partnerships by large

corporate enterprises was well underway, the moral claim of the firm
seemed stronger (and the notion of corporate authorship less
fictional) just as the moral claim of the employee waned.