Kreiss, ACCESSIBILITY AND COMMERCIALIZATION IN COPYRIGHT THEORY, 43 UCLALR 1

43 UCLALR 1

(Cite as: 43 UCLA L. Rev. 1)

UCLA Law Review

October, 1995

*1 ACCESSIBILITY AND COMMERCIALIZATION IN COPYRIGHT THEORY [FNa1]

Robert A. Kreiss [FNaa1]

Kreiss

INTRODUCTION ................................................................ 2

I. ACCESSIBILITY AND COMMER CIALIZATION IN COPYRIGHT THEORY .............. 6

A. The Goals of Copyright Law ........................................ 6

B. The Central Role of Accessibility in Achieving Copyright's

Primary Goals .................................................. 10

C. Copyright's Incentive Structure: Commercialization as the Means

to the Primary Goals ........................................... 14

D. Copyright's Quid Pro Quo ......................................... 20

II. REFLECTIONS ON ALTERNATIVE COPYRIGHT MODELS .......................... 22

A. Historical Perspective: No Need for an

Accessibility/Commercialization Model .......................... 22

B. Features of the 1976 Act Teaching Away from the

Accessibility/Commercialization Model .......................... 24

C. Alternative Model: Access at the End of the Copyright Term ....... 26

D. Alternative Model: The (Pro-Author) Incentives Argument:

E. Alternative Model: All Copyrighted Works Must Be Accessible ...... 32

1. The (Lack of) Textual Argument ................................ 34

2. The (Lack of) Constitutional Argument ......................... 37

III. ACCESSIBILITY, COMMERCIALIZATION, AND FAIR USE ....................... 41

A. Mass-Marketed Books .............................................. 45

B. Private Letters and Diaries ...................................... 47

C. One-of-a-Kind Works: Letters and Diaries in Library Collections;

Paintings and Sculptures in Museums ............................ 48

D. Limited-Edition Works ............................................ 51

E. Documents Containing Trade Secrets ............................... 54

F. Computer Programs ................................................ 56

G. Braille, Foreign Language Works, and Works on Records or CDs ..... 64

H. Television Broadcasts and News Clipping Services ................. 68

I. Accessible, Uncommercialized Works ............................... 70

IV. ACCESSIBILITY, COMMERCIALIZATION, AND WITHHOLDING INJUNCTIVE RELIEF .. 71

A. Out-of-Print Works ............................................... 72

B. Broadcast Television News Programs ............................... 74

CONCLUSION ................................................................. 74

*2 INTRODUCTION

Copyright's raison d'etre is to benefit the public by encouraging the production and dissemination of new copyrighted works. [FN1] In order for the means to serve the ends, the newly produced copyrighted works must be accessible to the public. [FN2] In light of this, it is somewhat surprising that copyright law [FN3] protects works that may not be disseminated at all, [FN4] and protects other works that may be disseminated in forms that are not humanly perceptible. [FN5] Even more surprising is the fact that while the accessibility *3 of a work should be of central importance to copyright theory, its role in that theory is more often assumed than carefully discussed by courts or commentators. [FN6] This lacuna is addressed by this Article.

Two recent cases [FN7] involving the disassembly of computer programs [FN8] have focused attention on the role of accessibility in the scheme of copyright. This attention has brought out articulate partisans of opposing views. On the one hand, proponents of accessibility argue that disassembly should be allowed in order to encourage authors to create new works that build on the unprotected ideas, processes, methods, and procedures embedded in a *4 first copyrighted work. [FN9] On the other hand, proponents of innovation argue that disassembly should be denied in order to maintain the proper amount of incentive for the creation of innovative works. [FN10]

To function properly, copyright law must strike a balance between the rights given to copyright authors and the access given to copyright users. The arguments of the advocates in the cases involving disassembly of computer programs simply highlight the fact that the issue of accessibility is one which lies at the heart of copyright law.

In spite of the energy with which the proponents have argued for their respective positions, the resolution of the accessibility issue is as simple as one-two-three. One. Accessibility is of central importance to copyright's primary objective of promoting public knowledge and learning [FN11] by encouraging the creation and distribution of new copyrighted works. [FN12] If works are not accessible, neither the public nor competing authors will get the benefit of these new works. Benefit to the public is important because the basic purpose of copyright law is to advance learning and knowledge. Benefit to competing authors is equally important because the authors produce competing works that allow the public a choice of views and expressions of each kind of work, thus leading to further advances in public learning and knowledge.

Two. The copyright system seeks to achieve the promotion of knowledge and learning by means of an incentive system. This system uses the economic rewards of the marketplace to stimulate the production and dissemination of new works. The exclusive rights granted to authors by the *5 Copyright Act permit authors to commercialize [FN13] their works without competition from copycats. [FN14] The potential to earn income from the commercialization of new works is the incentive that helps motivate authors to put in the long hours necessary to create those works. [FN15]

Three. There is a quid pro quo at the heart of the copyright system: if an author seeks benefits by commercializing a work, then the public should be able to benefit by having access to the work. If access is denied to the public for a work where the author is receiving the economic benefits made available by the copyright system, then the goals of copyright are not being served. For such works, the exchange is not fair: the public (both users and other authors) does not receive appropriate access to the work, yet the author's exclusive commercialization of the work is made possible by the copyright rights given by the public.

The quid pro quo can be restored, in part, by an appropriate use of copyright's fair use doctrine and by an appropriate limitation on the remedies given for copyright infringement. Under the fair use doctrine, competitors should be allowed to make copies and derivative works of a copyrighted work in order to gain access to the author's expression and to the author's ideas, systems, procedures, and methods. Having obtained access, the competitor must, of course, create his or her own expression of the ideas, systems, procedures, and methods.

In addition, copyright's remedies should not be used to prevent access to the expression, ideas, systems, procedures, and methods in a copyrighted work that has been commercialized. In appropriate cases, courts should withhold injunctive relief if an injunction would prevent competitors from gaining access to a copyrighted work that has been commercialized.

Authors need not take advantage of the economic aspects of the copyright system. They have the right to withhold a work from the market if they choose to do so; the grant of exclusive rights is not conditioned on an author's commercialization of the work. [FN16] Reciprocally, the public is not entitled to access to all copyrighted works. For example, private letters, *6 diaries, and in-house proprietary trade secret documents may never be published [FN17] and the public has no reason to expect access to these uncommercialized works. [FN18] The lack of accessibility does not mean that the copyright system has been subverted; it does not mean that there is an imbalance in the quid pro quo of giving rights to an author in order to encourage the creation and dissemination of a new work. For a work that is not commercialized, the exchange is fair: the public receives no access and hence nothing of value; similarly, the author receives no economic return from a copyrighted work that is not commercialized.

In Part I, this basic conceptual model of copyright will be elaborated. Part II will discuss and critique a number of alternative views of copyright's basic structure. Parts III and IV will examine the implications of accessibility and commercialization for the fair use doctrine and the granting of injunctive relief, respectively.

I. ACCESSIBILITY AND COMMERCIALIZATION IN COPYRIGHT THEORY

A. The Goals of Copyright Law

Copyright has primary and secondary goals. Understanding both sets of goals is important for an understanding of the roles of accessibility and commercialization in copyright theory.

Copyright law seeks "to encourage the widest possible production and dissemination of literary, musical and artistic works." [FN19] These are copyright's primary goals [FN20] and they are mandated by the constitutional *7 Enabling Clause on which the Copyright Act is based. Article I, Section 8 of the Constitution authorizes Congress "to promote the Progress of Science by securing for limited times to Authors the exclusive right to their respective Writings." [FN21]

The constitutional language of "promot[ing] the 'Progress of Science"' means promoting learning and knowledge. [FN22] The Copyright Act achieves this goal by encouraging authors to create and disseminate new works. [FN23] In copyright theory, the more works that are disseminated, the more this goal is advanced. [FN24] Different works may reach different members of the public through the use of different media or different methods of dissemination. [FN25] In addition, the fact that different authors may express themselves in different expressive manners is beneficial because different members of the public may respond to and learn from different expressions. [FN26] As the *8 Supreme Court has frequently pointed out, the rights given to copyright authors are a means to an end rather than an end in itself. "[T]he ultimate aim is, by this incentive, to stimulate artistic creativity for the general public good." [FN27]

The goals of encouraging the creation and dissemination of new works require a carefully balanced set of rights given to authors and privileges granted to users of copyrighted works. It must give authors an incentive to create, but it must also limit this incentive so that other authors can create new works that build on original works. Put simply, if there were no copyright rights, many authors would be unwilling to spend the time necessary to create valuable works because there would be no way for them to be compensated for that time. At the other extreme, if copyright granted overly extensive rights, society might find itself limited to one photograph of a puppy, one painting of a landscape, one country and western song, and one romance novel. [FN28]

The Copyright Act of 1976 attempts to avoid the extremes of underprotection and overprotection in a number of ways. The first is the limitation of copyright protection to an author's expression. The author's expression is protected to a significant extent, but the author's ideas, processes, methods, and procedures are not protected. [FN29]

The idea/expression [FN30] distinction promotes an appropriate balance *9 between the rights of copyright authors and the privileges of users by allowing authors to help themselves to ideas and methods used by previous authors. [FN31] For example, an individual rock and roll song is protected as expression, but the ideas and rhythms of the rock and roll genre are treated as ideas. Even if the genre could be traced to one individual, later musicians were allowed to use the ideas to create their own expressions of rock and roll. Society benefited from this balance by the widespread creation and dissemination of rock and roll songs.

A second way in which the Copyright Act avoids overprotection and underprotection is through the doctrine of fair use. [FN32] The protection given to an author's expression gives way to a variety of beneficial uses included under the general term of "fair use." For example, a scholar can quote portions of a novel as part of a critique of the author's work. [FN33]

In addition to copyright's primary goals, copyright also has secondary goals. [FN34] In enacting the Copyright Act of 1976, Congress also protected an author's privacy and gave authors control of the right of first publication. Protecting privacy for works such as diaries and private letters is accomplished by giving the author the exclusive right to make and distribute copies. [FN35] Protecting the right of first publication allows an author to withhold publication of a work until the author deems it ready for public distribution.

The 1976 Act may also have reflected a movement toward harmonization of the United States copyright system with some of the systems found elsewhere in the world. As more fully reflected in the passage of the Berne *10 Convention Implementation Act of 1988, [FN36] Congress may have had in mind some concern for an author's "moral rights." These rights protect an author's reputation and allow the author more control over the uses of the work. They are based on concerns about the personal and artistic relationship between the author and the work and not on the author's economic interest in the work.

While these non-utilitarian goals are secondary to copyright's utilitarian goals, they are essential in analyzing the importance of "commercialization" in copyright's system and in analyzing copyright's quid pro quo.

B. The Central Role of Accessibility in Achieving Copyright's Primary Goals

Accessibility plays a central role in copyright theory. In order for the copyright system to promote the Progress of Science, the public must have access to copyrighted works. If the public is not given access to copyrighted works, the public cannot learn from those works; learning and knowledge would not advance.

In this Article, the terms "access" and "accessibility" are used to refer to the public's ability to learn the ideas and expression from a copyrighted work. [FN37] In order for access to exist, two conditions are required: (1) a user of the work must be able to obtain a physical copy of the work and (2) the ideas and expression must be available in human-understandable terms.

A mass-marketed novel or musical composition passes both conditions. A private diary or letter fails the first condition: users cannot obtain copies. A sound recording on a compact disk fails the second condition: users cannot, without the help of a machine, read or hear the work. Other works are partly accessible and partly inaccessible. A foreign language work is accessible to some, inaccessible to others--accessibility depends upon who the user is. A musical work played on a CD player is accessible to the human ear but inaccessible to the eye--accessibility depends upon the type of understanding the user seeks. A computer program used on a personal computer may allow some access to the user interface while not allowing access to the structure of the source code--accessibility depends upon the aspect of the computer program the user wishes to learn.

*11 The statement "in order for the copyright system to promote the Progress of Science, the public must be able to have access to copyrighted works" must be expanded in two ways. First, the public's access must be to the author's expression in the work, and through that expression to the author's ideas. Copyright encourages authors to produce various expressions of the same ideas so that the public will benefit from those diverse expressions. The public gains access to the ideas explicitly or implicitly from its access to the expressions.

An example makes this point transparent. It is not enough for the public to be aware that two painters have painted portraits. The public benefit comes when the public is allowed to see the different expressions achieved by, for example, Rembrandt and Picasso.

The second expansion of the statement lies in the recognition that the "public" is composed of two groups of people. Copyright seeks to provide access to both of these groups. One part of the public is the using public, i.e., the public that reads, hears, or sees the author's work. If this public does not have access, it cannot learn from the works or benefit from them. If it does have access, it has the potential to benefit from the works. [FN38]

The second part of the public is other authors--authors who are or will be competitors of the original author. These competitors also must have access to the new copyrighted works. [FN39] If competitors do not gain access, *12 then they cannot create new works that build upon the original works. [FN40] Historians write better histories by having access to and building upon the works of earlier historians. Musicians create different expressions of new genres by having access to and building upon the works of other musicians in the genre. Computer programmers create improved programs by having access to and studying the work of earlier programmers. [FN41]

Access by competitors is an explicit aspect of patent law. At the time a patent issues, the specification and claims of the invention are published by the Patent and Trademark Office. [FN42] The public accessibility of the invention is deemed to be an essential part of the patent system because it enables competitors to learn about the invention so that they can build upon it.

Access by competitors is at least as important in copyright law. [FN43] Competitors are encouraged to build upon the work of others, extending the expression into new areas. In addition, and unlike patent law, copyright seeks to encourage competitors to create many expressions of any *13 given idea or theme. [FN44] Copyright encourages the creation of hundreds of photographs of puppies, hundreds of mystery novels, and hundreds of situation comedies. Access to the ideas and expression in a work are essential for competitors to create competing expressions of the same ideas. Once one work of a given kind is created and made accessible, other authors can study that work and create their own version of the genre. [FN45]

Noting that competitors must have access to copyrighted works simply states an assumption that underlies standard copyright doctrines such as the idea/expression dichotomy and the fair use doctrine. [FN46] The idea/expression dichotomy serves to explain those aspects of a prior work that a competitor can make free use of and those aspects that may not be freely copied. Equally important, the existence of the dichotomy implies that copyright law does encourage competitors to create competing expressions of the same idea--and implicitly acknowledges that competitors must have access to the ideas and expression in order to create their own new works.

Similarly, the fair use doctrine analyzes how much of a prior author's expression can be freely used by competitors and the public in their own works. [FN47] This doctrine also presupposes that the public and competitors have access to the prior work. [FN48]

Access by the using public and by other authors yields two types of public benefit. The first public benefit is the direct benefit that occurs because the public can learn from the ideas and expression of the first work. The second public benefit is the additional benefit that the public may *14 receive by being able to learn from the competing expressions and ideas created by competitors who also had access to the first work.

In sum, both the using public and competitors must have access to copyrighted works in order to further copyright's goals of promoting learning and knowledge. That access must be access to the expression in copyrighted works, and through access to expression, access to the ideas in those works.

Although not all copyrighted works are accessible, [FN49] one should not jump to the conclusion that the copyright system, taken as a whole, is failing to promote the Progress of Science. The next two sections further develop the analysis and indicate how to reconcile copyright's goals with the lack of access to some copyrighted works.

C. Copyright's Incentive Structure: Commercialization as the Means to the Primary Goals

The copyright system seeks to promote the public benefit of advancing knowledge and learning by means of an incentive system. The economic rewards of the marketplace are offered to authors in order to stimulate them to produce and disseminate new works. The mechanism is the granting of exclusive rights to authors with regard to their works. [FN50] The exclusive rights given to a copyright author are the rights that involve making money from the work. These involve the creation of copies, the sale of copies, the creation of derivative works, the public display of the work, and the public performance of the work. The exclusivity enables the author to prevent others from engaging in these money-making acts, thereby reserving to the author the ability to gain the economic rewards from the work. [FN51]

While these rights are the essence of copyright infringement, [FN52] they are not what copyright is all about. Copyright is really about the creating *15 and marketing activities of copyright owners, not copyright infringers. [FN53] Copyright is about helping copyright owners make money. It helps them make money by encouraging them to engage in the activities that they can prevent others from engaging in--making and selling copies of their work, creating derivative works, displaying their work publicly, and performing their work publicly--and doing so for profit.

In this Article, the term "commercialization" means doing, in a commercial or public setting, any of the acts that are reserved exclusively to the copyright owner by the Copyright Act under 17 U.S.C. § 106 [FN54] and that would be copyright infringement if done by another without authority of the copyright owner, whether these acts are done for profit or not; except that the sale of the original copy of a work is not, in and of itself, a commercialization. [FN55]

There are numerous examples of commercialization. Commercialization exists when mass-market books and records are published; when movies or plays are performed publicly; when paintings and photos are placed on public display; when copies of computer programs are sold, leased, or licensed; when digital works are placed in databases that are publicly accessible; and when documents containing trade secrets are disseminated, pursuant to trade secret licensing agreements, to persons outside *16 of the company having the trade secrets. Commercialization also exists when an author of a screenplay authorizes the creation of a movie based on the screenplay; when a novelist assigns the rights to publish the book; and when a painter authorizes the public display of her paintings.

These examples illustrate the kind of acts that copyright owners typically do to earn money from their works. The Copyright Act reserves these money-making activities to the copyright owner by granting the owner the right to prevent others from engaging in these acts.

It is this potential to earn income from the commercialization of new works that presumably motivates authors to put in the long hours necessary to create those works. [FN56] The creation of the work is the first step for an author. Copyright recognizes the importance of this step by granting copyright protection to the work from the moment it is first fixed in a tangible medium of expression.

However, the copyright system is more concerned with the economic return a copyright owner can obtain than with the work's creation. [FN57] The set of rights given to a copyright owner under the Act makes this clear. [FN58]

Several aspects of the definition of "commercialized" should be noted. First, the test for commercialization focuses on the activities that are granted exclusively to the copyright owner by Section 106 of the Copyright Act. It is important to focus on Section 106 because it contains the so- called *17 "economic rights" in the copyright system. [FN59] Copyright is based on an economic incentive system that encourages authors to create and disseminate copyrighted works. When an author does or authorizes any of the acts reserved exclusively to copyright owners under Section 106, the author is acting as Congress hoped the author would act. Focusing on these acts focuses our attention on the acts that are central to copyright's incentive structure.

It is not enough, however, to focus solely on Section 106. The definition of commercialization specifically focuses on those acts that would be copyright infringement if done by another without authority of the copyright owner. This definition excludes acts that are within the scope of Section 106 but which nevertheless are excused by other provisions of the Copyright Act. [FN60] The point of the definition is to focus attention on the benefits given to copyright owners by the copyright system. Copyright owners benefit from exclusivity and the definition of commercialization captures this point by its focus on activities that would be infringing if done by another without authorization.

Second, to be a "commercialization," the owner's activities must be "in a commercial or public setting." While a work is being created, the owner may make and distribute copies of drafts to a few associates for critical comments. Such activities involve activities reserved exclusively to the copyright owner by Section 106, but the activities are not done in a commercial setting. However, when a professor distributes copies of a memorandum to her students, for example, the work should be deemed complete and the distribution done in a public setting.

Normally, the test for a "commercial setting" is whether there is a commercial transaction in which money is given in exchange for the dissemination of the work. When copies of documents containing trade secrets are made and disseminated to employees within a company, there is no commercial transaction. However, the making and disseminating of copies of the same secrets to people outside of the company, as part of a trade secret licensing agreement, probably constitutes a transaction within a commercial setting.

*18 Commercialization can also exist in the absence of a commercial transaction. The publication of a law review article rarely involves royalties being paid to the author. However, it does involve the dissemination of copies of the work to the public. This distribution in a "public setting" makes the activity one of "commercialization."

Third, commercialization does not depend upon a profit motive. An artist who allows a charitable organization to make and distribute copies of a work of art has authorized the kind of act that the Copyright Act reserves exclusively to the copyright owner, whether or not the artist receives or requires a fee in return for the authorization.

Fourth, the sale of the original copy of a work is not, in and of itself, a commercialization. A painter who paints a portrait of an individual and sells the original to the individual has transferred his personal property in that original. Copyright law may prevent additional copies from being made, but the author is not yet making money from the use of the exclusive rights granted by the Copyright Act. [FN61] Only when the author begins or authorizes the production and distribution of more copies should copyright then consider the author to be engaged in commercializing activities.

Fifth, the concept of "commercialization" coincides with that of "publication" [FN62] for many kinds of works and commercial activities. The typical book, map, chart, musical composition, or play is disseminated to the public by the sale of copies of the work. The sale of copies is both a "publication" and "commercialization." However, "commercialization" does not always coincide with "publication." A movie that is shown nation wide in movie theaters may not be published, but it is commercialized. In this example, the copyright owner is earning an economic return through the public performances of the work, even though the work has not been published. Similarly, commercialization exists when an author of a screenplay authorizes the creation of a movie based on the screenplay, even though the movie has not been created, let alone published. The broader concept of "commercialization" recognizes that copyright's incentives focus on the copyright owner's economic uses of the work, not on whether the work has been "published."

Historically, "publication" marked the point at which a work entered the commercial marketplace and the public simultaneously received the benefits of the copyrighted work. Prior to publication, common law copyright protected the author's right to first publication, and allowed the author *19 time to finish the work and prepare it for public distribution. Publication of the work divested the author of common law copyright and marked the time at which the author had to obtain federal copyright protection and use copyright notice. [FN63]

Terms of art, such as "published" and "publication," that were developed in the context of determining when copyright notice had to be used and when federal copyright protection began [FN64] should not be inappropriately imported into a different context.

For works such as books, publication marked the time when the public obtained copies and could benefit from the work. However, nothing in logic suggests that the public benefit should begin at the same time that the owner's benefit begins. In fact, the economic benefits to a copyright owner might begin before or after this public benefit begins. [FN65]

The concept of commercialization focuses our attention on the time when the author seeks the benefits under copyright which creation of the work makes possible. Once an author engages in any of the acts reserved exclusively to the author and attempts to obtain an economic return in this way, she places herself within the copyright incentive structure, whether or not those acts fall within the concept of "publication." At this point, we should treat the author as having sought the economic benefits of copyright law. [FN66]

*20 Finally, "commercialization" can exist even where a copyrighted work is not fully "disclosed," where disclosure means providing at least one copy of the work for public inspection. Disclosure is required to obtain a patent. [FN67] But, for some copyrighted works, neither the Copyright Act nor the Copyright Office requires the deposit of a copy of the full copyrighted work for public inspection. [FN68] The test for commercialization focuses on acts of a copyright owner through which the owner can realize money, whether or not disclosure of the work is required or exists.

D. Copyright's Quid Pro Quo

The constitutional goals of copyright are the advancement of learning and knowledge. The means to achieve those ends is the incentive system which induces authors to create and disseminate their works.

In order for the means to further the ends, the copyrighted works must be accessible. When an author commercializes a work, the author gets the economic reward intended by the copyright system. If the work is accessible, the public and competitors get the benefit of being able to study the work's ideas and expression. [FN69] In this circumstance, there is a quid pro quo. [FN70] Both sides get the benefit that the copyright system was designed to produce.

Copyright has an equally balanced quid pro quo in the circumstance of a work that is created, but never commercialized. Private letters, diaries, and in-house trade secret material fit into this category. For these works, the public receives no benefit toward the advancement of learning and *21 knowledge. However, because the author obtains no economic benefit from the use of the exclusive rights granted by copyright, [FN71] there is a balance in the copyright quid pro quo system. [FN72]

One can picture the relationship between accessibility and commercialization as a two-by-two matrix. Copyright's quid pro quo is in balance for commercialized, accessible works and for uncommercialized, inaccessible works. It is out of balance in the other two quadrants of the matrix.

THE QUID PRO QUO MATRIX [FN73]

Accessible Works Inaccessible Works

--------------------------------------------------------

Commercialized Works Quid pro quo in balance Quid pro quo out of balance

--------------------------------------------------------

Uncommercialized Quid pro quo out of balance Quid pro quo in balance

Works

--------------------------------------------------------

The matrix indicates that there are two circumstances in which copyright's quid pro quo is thrown out of balance. One circumstance is when *22 works are commercialized but not publicly accessible. [FN74] For such works, the author obtains the economic benefit that the copyright system was designed to give to authors, but the public does not receive the public benefit that was intended. Copyright's means benefit the author, but copyright's utilitarian ends are not achieved. [FN75] The balance in the quid pro quo should be restored by a sensitive application of copyright's fair use doctrine and by an appropriate limitation on the remedies given for copyright infringement. [FN76]

The second possible imbalance in copyright's quid pro quo is where works are uncommercialized, yet accessible. For these works, the author is not taking advantage of copyright's economic incentive structure, yet the public is obtaining the benefits of access. This situation should not trouble us because the copyright owner has control over whether or not to commercialize the work. Whenever the owner desires, he or she can begin commercializing the work, thereby restoring the balance.

II. REFLECTIONS ON ALTERNATIVE COPYRIGHT MODELS

A. Historical Perspective: No Need for an Accessibility/Commercialization Model

The relationship between accessibility and commercialization has been overlooked until now. There are a number of possible reasons for this. First, copyright arose in the context of book publishing and applied to books that were published. [FN77] The Copyright Act of 1909 carried forward this historical view by linking federal copyright protection to the publication of a work, with only a few exceptions. [FN78]

*23 Publication normally meant accessibility. [FN79] By definition, publication meant that copies were publicly distributed. [FN80] Even if that distribution was rather limited, registration meant that copies of the work would be deposited with the Library of Congress. Distribution and deposit meant that scholars and competitors could gain access to the work. As important, the kinds of works involved--books, maps, charts, musical scores, paintings, sculptures, etc.--were the kinds of works whose ideas and expressions are apparent from humanly examining the work. These ideas and expressions are not hidden in any way. [FN81] For this reason, copyright scholars who focused on the federal system [FN82] may have assumed that there would be accessibility; they did not discuss it. [FN83]

While the assumption of accessibility was reasonable in the early years of copyright, it became less reasonable as new technologies were developed, especially if one looked beyond federal copyright to works protected by common law copyright. For example, phonograph records and tapes containing musical expression are not humanly perceptible in that form and one can only access them by using a machine. While this does not give access to the full score of the work, it does give partial access.

*24 Radio and television broadcasting produced other technological challenges to full accessibility. A broadcast of a copyrighted work might reach thousands of people, but accessibility to the work would be limited to the short time of the broadcast. Unless copies of the broadcast were made and distributed, on-going access would not exist.

Under the 1909 Act, these access problems did not bother the federal copyright system because that system did not always treat those works as protectable. Piano music rolls and records were held not to be entitled to federal copyright protection. [FN84] Broadcast radio and television programs were also unprotected because copies of programs were not distributed to the public. [FN85]

The commercial importance of these new technologies made it increasingly important for them to be brought inside the federal copyright system. [FN86] Those working on the copyright revision leading up to the Copyright Act of 1976 were aware of the need to adapt copyright to the realities of new technologies. [FN87]

B. Features of the 1976 Act Teaching Away from the Accessibility/Commercialization Model

The second major reason why the relationship between accessibility and commercialization has not been explicated until now stems from the structure of the Copyright Act of 1976. The Act led scholars astray in a number of ways: it eliminated any requirement that a work be published in order to qualify for copyright protection; it abolished common-law copyright for works fixed in a tangible medium of expression; it eliminated the requirement that works be deposited with the Library of Congress; and it used "publication" rather that "commercialization" as a relevant dividing line for some purposes.

*25 In spite of these features of the Copyright Act of 1976, the Act did not change copyright's constitutional goals of furthering learning and knowledge. [FN88] Nor did it change the fact that accessibility is the instrument by which those goals are achieved. Nor did it change the use of an economic incentive system to motivate authors. However, the changes wrought by the new act did obscure these basic features. One thing that misled people is the 1976 Copyright Act's continued use of the concept of "publication" rather than "commercialization." [FN89] While in many contexts these terms will coincide, they do not always do so.

The Copyright Act defines "publication" as "the distribution of copies or phonorecords of a work to the public by sale or other transfer of ownership, or by rental, lease, or lending." [FN90] For works such as mass-marketed books, this may coincide with commercialization. However, for works that are not distributed to the public--works such as the source code of computer programs or broadcast television news programs--the terms do not coincide. It is important to realize that copyright's incentive system does not concern itself with "publication" as that term is defined in the Act. Such a concern would be too narrow. The Act is concerned with creating an economic incentive for authors. When authors attempt to benefit from that incentive structure, the copyright system is working properly. The term "commercialization" more accurately captures the stage at which an author attempts to exploit the copyright. [FN91]

The other key changes that misled people were that copyright now attaches from the moment a work is fixed in a tangible medium of expression [FN92] and that common-law copyright is abolished for most works. [FN93] This means that many works that would not have received federal copyright protection under the 1909 Copyright Act now have protection. Such works include private letters and diaries, drafts of works, research and laboratory notes that contain written expressions of trade secrets, secure tests, unpublished television programs, and computer programs.

Access to many of these works is limited or even non-existent. Given this fact, a number of scholars may have believed that a new copyright *26 theory would have to be created that treated access differently from the historical view that access coincided with publication.

C. Alternative Model: Access at the End of the Copyright Term

Some have asserted that public access is satisfied by placing copyrighted works in the public domain after the copyright term has expired. [FN94] This view is unsatisfying for a number of reasons. First, it confuses accessibility with copying. When the copyright term ends, anyone can freely make copies of the copyrighted work. The public benefit that arises from this is the possibility of having competing versions of the same work. This may involve different prices, different paper quality, different type fonts, and other differences. Moreover, other authors can then freely copy the first author's expression in their own works.

The public benefit from this kind of copying is quite different from the public benefit derived from accessibility. Accessibility to a work's ideas and expression contributes to public learning and knowledge. Copyright's quid pro quo has historically been designed to advance learning and knowledge, not simply to enable competing versions of the same work to be produced in some distant future time.

Second, the view that the quid pro quo under the 1976 Act involves public access at the end of the copyright term represents a dramatic departure from copyright's origins without any indication that Congress intended such a departure. Historically, access occurred at the time of publication. [FN95] It is difficult to believe that Congress wanted to defer access to novels, music, and other paradigmatic works for more than fifty years. [FN96]

Third, this view is unsatisfying because of the very long time period that might exist before inaccessible works would become accessible. [FN97] *27 America has long been a country with rapid development and diffusion of creative works. This has come about, in part, because innovation has been coupled with rapid disclosure of new ideas. New inventions are disclosed at the time a patent is issued. New artistic ideas and copyrighted expression have historically been disclosed at publication. It is almost inconceivable that Congress would have wanted to replace this successful system of rapid change for one with much slower change; yet, that is what is suggested by a system wherein some commercialized copyrighted works could be inaccessible for decades.

This Article offers a copyright theory that does not suffer from these drawbacks. It accepts the fact that copyright now includes inaccessible works and uses the commercialization test to distinguish between those works for which there should be public accessibility and other works. This allows copyright theory to maintain its traditional goals of furthering learning and knowledge, and does not ask society to wait for more than fifty years in order to achieve those goals.

D. Alternative Model: The (Pro-Author) Incentives Argument: Copyright as a Pro- Creation or Misappropriation Statute

Other scholars have tried a different approach in attempting to reconcile the 1976 Copyright Act with copyright's goals; they have tried to redefine those goals. For example, some have argued that copyright's goals are now focused on the creation of new works. [FN98] Others have argued that copyright is primarily an anti-misappropriation statute, protecting the fruits of intellectual labor. [FN99] Under these views, dissemination of a work is no longer of central importance. Instead, any claims of access are adequately met by whatever access the copyright owner chooses to provide.

*28 These views are certainly supported by a superficial view of the Copyright Act of 1976. It is easy to argue that accessibility is not important when the Act expressly grants copyright protection to unpublished works. This argument is reinforced by the fact that copyright owners no longer always have to make deposits of their work with the Copyright Office. [FN100]

However, these views should also be rejected. The constitutional goal of copyright--to "promote the Progress of Science"--is unchanged, and the need for access in order to accomplish this goal is unchanged. There is no indication that Congress intended to discard either of these.

Copyright's history also belies the view that copyright's goals emphasize creation, but not dissemination. Prior to 1978, both common law copyright and federal copyright focused on publication. Common law copyright protected an author's right of first publication and federal copyright encouraged that publication by withholding rights until publication occurred. Neither type of protection provided any economic return to an author simply because of the creation of a work.

The 1976 Act did not drop this emphasis on commercialization of the work. Although federal copyright protection now exists from the moment a work is created, there is no economic return from a created, but uncommercialized work. The incentive is that of encouraging the creation and dissemination of the work.

Scholars who argue that access to a work is not essential to copyright sometimes couch their arguments in terms of the incentives given to copyright authors. They argue that without the ability to control and limit access to a work, fewer copyrighted works will be created. They particularly make this argument in the context of computer programs, forcefully arguing that it is easy for knock-off artists to disassemble or decompile computer programs and create unauthorized copies for sale. Unless disassembly and decompilation are made illegal, copyright authors will be unable to protect their investment in the computer programs.

Analytically, this argument must be rejected. These scholars are correct that copyright creates an incentive for authors to create works. However,*29 because creation without commercialization yields no economic return for the author, it cannot justify the initial effort to create a work.

Moreover, copyright's incentive is not unlimited. The scope of the incentive is defined by the statute: the author is given those exclusive rights, and only those rights, provided by the statute, and the statute provides for appropriate public access as part of the quid pro quo. This "incentive" will produce a certain number of copyrighted works. The fact that more copyrighted works might be produced if less public access were permitted does not mean that copyright's existing "incentive" system is not working. If those arguing for less access were to prevail, there might be more works, but there would be less public benefit to learning and knowledge. [FN101]

The alteration in copyright's quid pro quo sought by these scholars may be short-sighted. It appears to be pro-author in that it seeks to give copyright authors stronger rights. However, giving a copyright author stronger rights comes at a price. To the extent that competitors cannot get access to the ideas and expression in a work, they will be unable to produce new expressions of the same ideas. These competitors are also "authors." The short-term benefit for the first author can turn into a long-term barrier to many later authors. This barrier harms the public, who are then foreclosed from the works that those later authors might produce. These considerations mean that it is probably impossible to resolve empirically whether such a change in copyright's balance would yield more copyrighted works or not. [FN102]

*30 In addition to being short-sighted, the "incentives" argument makes no sense from the public's point of view. In effect, this argument asks the public to offer the copyright owner exclusive rights and leaves it up to the copyright owner to determine what access the public should receive in return. It is not a reasonable bargain for the public to give copyright owners this kind of a blank check. [FN103]

The "incentives" argument is also flawed to the extent that it confuses distinct issues. If there is a problem in proving infringement of a work, that issue should be dealt with directly. It should not be dealt with by focusing on the distinct issue of the incentive to create works. For purposes of discussion, one might agree that new technologies may make it difficult to prove infringement: computers can rapidly make many copies of a computer program, and computers may aid in scrambling the code in a program to make it more difficult to detect infringement. [FN104] However, these points do not justify altering copyright's quid pro quo. That quid pro quo focuses on the creation of and public accessibility to new works, not on infringement.*31 If there are problems with policing copyrights, then they should be dealt with as the problems that they are. [FN105]

The bottom line is that the copyright system offers an incentive as part of a quid pro quo bargain. Authors can walk away from that bargain if they don't like its terms. If an author refuses to create a work because the incentive is not large enough, then the work will not be created. Complaining that the incentive is not large enough is not a reason for society to change the system for that author or group of authors.

The challenge has been to build an understanding of copyright that allows the 1976 Act to protect unpublished works while at the same time furthering the constitutional goal of and need for access. The present Article meets this challenge. It focuses on copyright's traditional economic incentives to encourage authors to commercialize their works and argues that public access is important for these works. At the same time, this Article notes that federal copyright serves non-economic ends for works that are not commercialized. [FN106] For these works, copyright is largely irrelevant as an incentive for the creation of the work, and this Article argues that these works need not serve copyright's utilitarian goals.

*32 E. Alternative Model: All Copyrighted Works Must Be Accessible

A third group of scholars has responded to the Copyright Act of 1976 in a more confrontational manner. Rather than arguing that copyright's goals have changed or that access is unimportant, these scholars have argued that there is something wrong with a copyright statute that grants protection to inaccessible works. [FN107]

For some of these scholars, the accessibility issue did not crystallize until the widespread creation and dissemination of computer programs. [FN108] Computer programs are protected under copyright as literary works or audiovisual works. At the same time, the source code for many commercialized computer programs is not "published" and the programs are usually distributed in object code format, [FN109] a format that is not readable by humans. [FN110]

*33 This format was probably adopted because programs that have been compiled or assembled into object code format run more quickly than programs that have not been converted into that format. Speed of operation is one competitive advantage gained by distributing programs in object code format.

A second competitive advantage is that because it is not readable by humans, object code impedes competitors from learning the structure or methodologies used in a computer program. Program distributors reinforced this fact by attempting to impose licensing restrictions that prevent users from disassembling, decompiling, or reverse engineering the program. [FN111] The combination of licensing restrictions and distribution of programs in object code format presents a significant barrier to accessibility of computer programs by users and by competitors--the impact on competitors being the more significant one. [FN112]

The lack of full public access to the ideas and expression in commercialized computer programs began to trouble some scholars. [FN113] These works receive copyright protection, yet do not provide the kind of public benefit associated with copyright law; [FN114] they do not "promote the Progress *34 of Science." [FN115] The lack of access has led to litigation [FN116] and has raised the question of whether copyright protection is appropriate.

As mentioned above, some scholars have argued that copyright should not be granted to inaccessible works, [FN117] citing the fact that the public does not receive the requisite benefit from such works. [FN118]

1. The (Lack of) Textual Argument

In this and the next subsection, we see that neither the copyright statute nor the Constitution requires access to copyrighted works.

As an introduction to these issues, it is useful to note that limited accessibility arises in contexts other than that of computer programs. These include office memoranda that are written, read, and destroyed; private letters and diaries which are never copied and distributed, but which may be placed in a library; portrait paintings which are sold to the subject and may eventually be placed in a museum, but of which copies are never made; limited edition books whose few copies go to a variety of buyers; secure tests that are never released publicly; television programs that are broadcast but for which no copies are sold; and laboratory and research notes that are circulated as trade secrets under non-disclosure restrictions.

*35 Each of these examples can be contrasted with the kinds of works which are widely accessible--works such as mass-marketed novels, popular music, or movies. The point is that traditional copyrighted works vary greatly in the degree of access that may exist. For office memoranda and secure tests, there may be no public access. For private letters and diaries placed in a library, there will be access for those able to go to the appropriate library, but no access for others. For limited edition books or paintings, there will be access for those fortunate enough to find a copy someplace, but no access for others. For trade secret material, there will be virtually no public access. For unpublished television programs, there will be transitory access at the time the program is broadcast, but there may be no later access.

The point of these examples is that while accessibility may be the engine driving the production and proliferation of a multitude of works of any given genre, many of those works will not be destined for the market at all, and other works may be unsuccessful in the marketplace or will be destined for a very limited market. These examples should serve to curb the temptation to place undue emphasis on accessibility. [FN119] With this caution in mind, we now turn to the question of whether accessibility is required by the Copyright Act.

The starting point for analyzing whether copyright requires access is the Copyright Act itself. Is there any statutory language that requires access to the expression and/or ideas in a work as a condition of granting copyright protection to that work? The answer is that there is no such language. The Act states, in very broad terms:

Copyright protection subsists, in accordance with this title, in original works of authorship fixed in any tangible medium of expression from which they can be perceived, reproduced, or otherwise communicated, either directly or with the aid of a machine or device. [FN120]

Nothing in the statutory language requires that a work be accessible in order to be protected. [FN121] The language "original works of authorship" is *36 sweeping enough to include letters that are sent to only one person and never published, diaries that are written and never shown to anyone, research notes that are maintained by a company as a trade secret and not disseminated outside the company, secure tests that are administered nationally but never published, drafts of novels and other works that never see the light of day, and other unpublished or inaccessible works.

All of these examples involve the kind of "authorship" that is protected by the Copyright Act and all are original. The only other statutory requirement is that they be fixed in a tangible medium from which they can be perceived, reproduced, or otherwise communicated. All of these examples are indeed fixed in a tangible medium and they can be perceived from that medium.

Thus, the text of the Copyright Act supports the conclusion that computer programs, documents containing trade secrets, private letters and diaries, and other undisclosed works are copyrightable. Nothing in the text indicates that accessibility is a prerequisite for copyright protection.

This conclusion is supported by the legislative history of the Copyright Act of 1976. Prior to this act, federal copyright protection did not begin for most works until a work was published. Thus, for most copyrighted works, federal copyright protection coincided with accessibility. Under the 1976 Act, federal copyright protection begins at the moment of a work's creation. When Congress enacted this law, did Congress intend to limit protection to those works which were accessible? [FN122] Obviously not, for between the creation and publication of most works, the works will not be accessible, yet Congress extended federal copyright protection to these works during this period of inaccessibility. In addition, Congress was well aware that diaries, letters, and other works that might never become accessible were going to be protected by federal copyright. [FN123] Finally, one must note that Congress intended to give protection to phonograph records, which, like computer programs, were distributed in a form unreadable by *37 humans. [FN124] The inaccessibility of a copyrighted work was not intended by Congress to bar copyright protection.

2. The (Lack of) Constitutional Argument

In this subsection, we see that the Constitution does not require access in order for federal copyright to be granted. The Constitution states that Congress can enact copyright statutes "[t]o promote the Progress of Science by securing for limited Times to Authors the exclusive Right to their respective Writings." [FN125] Two portions of this clause are relevant to the discussion. First, the argument is made that inaccessible works, such as computer programs, do not "promote the Progress of Science" and hence cannot be given copyright protection. [FN126] Second, the argument is made that the only "exclusive Right" Congress can grant is the right to publish in a form that is publicly accessible. [FN127]

While the first argument is often couched in terms of computer programs, the nature of the argument makes it applicable to any kind of inaccessible work. [FN128] Private letters, diaries, secure tests, and documents containing trade secrets are examples of other kinds of works that are inaccessible. The argument requires an examination of whether the Constitution granted Congress the power to give copyright protection to these works.

The latter part of the constitutional Enabling Clause contains various words of limitation. The Supreme Court has held that the words "Authors" and "Writings" are words of limitation, requiring some threshold amount of creativity to be present before copyright protection can be granted. [FN129] Similarly, the phrase "limited Times" forbids Congress from enacting a copyright statute that would grant perpetual protection for copyrighted works.

The question raised by inaccessible works is whether the Introductory Clause, "promote the Progress of Science," contains words of limitation. Does each individual copyrighted work have to advance learning and knowledge? [FN130] If so, then an individual's private diary or other work which the person never intends to publish would be denied copyright protection. Unpublished works of this kind would give the public no benefit: there would be no "Progress of Science" from the production of these works. Similarly, private letters and documents containing trade secrets would be denied protection. [FN131] And computer programs that are distributed in object code form would be denied protection.

The alternative is to read the Introductory Clause, "promote the Progress of Science," as containing words of purpose. If this clause is read as stating the purpose, then Congress would be allowed to give copyright protection to a wide class of works in order to ensure that "Science" would be promoted by at least some of those works. Under this reading, Congress could protect all Writings [FN132] even though many of those writings might be unpublished or otherwise inaccessible to the public.

The proper answer to the question posed is that there is no constitutional requirement of accessibility; the Introductory Clause in the Constitution contains words of purpose, not words of limitation. [FN133]

First, as a matter of grammar, the constitutional Enabling Clause has two parts, separated by a comma. The first portion plainly speaks of the purpose to be achieved: "to promote the Progress of Science." The second portion grants Congress the means by which to achieve this purpose: "by securing for limited Times to Authors the exclusive Right to their respective Writings " The means language contains limitations: copyright protection must be for "limited times" and for "Authors." On the other hand, the language of purpose is simply that--the objectives that Congress must seek to achieve. The language is not language of limitation. *39 Thus, the grammar of the text refutes the idea that inaccessible works must be denied copyright protection.

Moreover, on a textual level, the words are a broad directive to Congress to "promote the Progress of Science"; the words do not indicate that individual works must do this. [FN134] Congress has the power to enact appropriate legislation to achieve this goal.

Second, on a policy level, one notes that courts are ill-equipped to take on the task of applying the alleged constitutional prohibition. The task would involve terribly difficult questions of judgment on which the Copyright Clause provides no guidance and that would trouble courts for many decades to come: Does this science fiction novel promote knowledge and information? Is this poem beneficial to society? Does this painting or photograph promote the public welfare? Is this sculpture or dance worth-while? Is an obscene work protectable? These are questions that courts have wisely shied away from, following Justice Holmes' admonition:

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. [FN135]

A number of circuit courts [FN136] and commentators [FN137] have reached the same conclusion. The Fifth Circuit's analysis of this issue is often cited. It noted that while Congress could "require that each copyrighted work be shown to promote the useful arts (as it has with patents), it need not do so." Instead, Congress could "conclude that the best way to promote*40 creativity is not to impose any governmental restrictions on the subject matter of copyrightable works." [FN138] While the case in which this sentiment was uttered involved an allegedly obscene work, the point is the same. The copyright system (particularly the courts) is not particularly well suited to evaluate whether individual works provide a public benefit and, hence, will not concern itself with such a determination at that microlevel. [FN139] Rather, the copyright system as a whole is designed to promote knowledge and learning.

Because questions about whether a particular work does or does not promote learning and knowledge are so subjective, the decisions as to what kinds of works to protect and what level of protection to give those works are questions which Congress is better equipped to address. Congress can take into consideration a variety of factual considerations and can establish appropriate protection based on its sense of social and economic priorities. Where, as in this instance, the constitutional language does not compel that decisions be based on constitutional grounds and where there are good reasons for desiring to retain the flexibility to adapt a statute to changing circumstances, it is better policy to interpret the Constitution to allow the flexibility.

The second constitutional argument is that the only "exclusive Right" Congress can grant is the right to publish in a form that is publicly accessible. [FN140] As noted above, [FN141] historically the English copyright statutes arose in the context of book publishing. Because published books are accessible, one can argue that the "exclusive Right" referred to in the Constitution means that only publicly accessible works can be protected.

The Constitution should be interpreted flexibly in light of changing technologies and changing societal needs. If the "exclusive Right" referred to in the Constitution should be interpreted only to deal with publishing of books, then paintings and sculptures could not be protected under the Copyright Clause of the Constitution. There is no need for a constitutional*41 amendment to protect creative works that are in different art forms from books.

Similarly, one notes that records, tapes, CDs, computer programs, and semiconductor chips did not exist in the eighteenth century when the Copyright Clause was drafted. This author believes that the Constitution's Copyright Clause should be interpreted flexibly to allow it to embrace new technologies, so long as the policies underlying the Clause are being furthered.

III. ACCESSIBILITY, COMMERCIALIZATION, AND FAIR USE

This Part of the Article explores how accessibility, commercialization, and copyright's quid pro quo affect fair use analysis. [FN142]

The Copyright Act states that "the fair use of a copyrighted work is not an infringement of copyright." [FN143] The problem for courts has been that of distinguishing fair uses from unfair ones. On this point the statute provides guidance and suggestions, but the ultimate decision should be made in light of copyright's purposes and objectives. [FN144]

There are a variety of ways to look at the fair use section of the Copyright Act. [FN145] One way is to ask whether the particular use being made of *42 the work has enough social benefit that it outweighs harms to the copyright owner. For example, one historian may create a work of history and quote brief passages from another historical work in order to comment on or criticize that work. The second historian may be a competitor, [FN146] may be commercially (as well as scholastically) motivated, and the second work may have a negative impact on the market for the first work. Notwithstanding these factors cutting against a finding of fair use, the social benefit of having a second historian build on an earlier work may excuse, as fair use, the fact that she has quoted portions of the earlier work. [FN147]

Indeed, in this instance the fair use doctrine reinforces the basic incentive structure of copyright. [FN148] Copyright seeks to encourage the creation *43 and dissemination of new works. If the second historian's use was held to be infringing, then copyright would be discouraging the creation of the new work. By recognizing the social value of having later historians build on the works of earlier historians, fair use furthers copyright's basic incentive structure. [FN149]

What role do accessibility and commercialization of a copyrighted work have in the fair use analysis? As with almost all fair use questions, the answer is that they play an important, but not dispositive role. [FN150] For instance, in the example of a second historian quoting from an earlier historical work, access to that first work is a prerequisite to being able to quote from that work. Access promotes the public benefit of the creation of new works that build on the work of earlier authors. Indeed, access is a prerequisite for almost any fair use claim. Activities such as "teaching, scholarship, and research" are given special favor under the fair use analysis, being blessed by statutory language. Obviously, they presume that the work is accessible.

Commercialization is also important. [FN151] In discussing the second fair use factor, [FN152] courts have distinguished between published and unpublished works. [FN153] In his treatise, Goldstein notes that "[c]ourts more closely confine the scope of fair use of unpublished works than they do the fair *44 use of published works on the ground that the right of first publication secures particularly important marketing interests." [FN154]

A moment's reflection on the rationale identified by Goldstein--to "secure [[] marketing interests"--indicates that the appropriate distinction should be couched in terms of "commercialization" rather than "publication." [FN155] For works such as published mass-market books, there may be little difference between the two concepts. [FN156] However, for broadcast television programs, the "marketing interests" start with public performance, a form of "commercialization," not with "publication." [FN157] Similarly, the "marketing interests" for mass-marketed computer programs start with *45 distribution of copies of the object code, a form of commercialization, rather than with the unpublished source code.

Considerations of copyright's quid pro quo highlight instances in which fair use should be used to correct imbalances. Standard fair use analysis can be applied to commercialized, accessible works such as mass-market books. For these works, fair use plays a role in correcting market failures or protecting socially desirable uses that do not have an overly-deleterious impact on copyright owners.

With commercialized, inaccessible works such as mass-marketed computer programs, fair use should be used to prevent copyright owners from inappropriately benefiting from an imbalance in copyright's quid pro quo. The imbalance in these cases is not that of market failures; it is that the copyright owner is attempting to have his cake (not allowing public access to the expression of the work) and eat it too (profiting from the commercial distribution of the product). [FN158]

This Part will examine a variety of works that have different combinations of accessibility and commercialization, to determine how the accessibility or inaccessibility, and commercialization or lack of commercialization, of each particular work affects the fair use analysis. First, the extremes will be considered: works that are the most accessible and most commercialized (e.g., mass-marketed books) and the most inaccessible and least commercialized (e.g., completely private letters and diaries). Next, a variety of other works with various degrees of accessibility and commercialization (e.g., letters and diaries in library collections, paintings in museums, limited edition works, documents containing trade secrets, television broadcasts, foreign language works, records and CDs, and computer programs) will be considered.

A. Mass-Marketed Books

Mass-marketed books represent the archetypal example of "accessible, commercialized" works. [FN159] The effective working of copyright incentive's quid pro quo is seen most easily for this kind of work. It is presumed that copyright helped provide the incentive to the author to create the work and to distribute the work. In return, the public benefit from that access is *46 seen: [FN160] first, by access by the using-public so that they can gain information and knowledge, and second, by access by competitors so that the public will receive additional works that build on the first work. [FN161]

The standard fair use paradigm evolved in the context of these works. Consequently, standard fair use analysis applies to most of these works. In particular, courts will be aware of the high degree of accessibility, and this will be represented in two ways. First, as to the second fair-use factor, [FN162] courts will note that the work has been "commercialized." [FN163] Second, as to the fourth fair-use factor, [FN164] courts will be aware that the author has marketed the work and will inquire into uses of the work by a competitor that have an impact on the relevant market. [FN165]

Mass-marketed books that have gone out of print require a more discriminating fair use analysis. Over time, these works become less and less accessible, making the case for fair use stronger. [FN166] Even if fair use is not *47 found, courts should consider withholding injunctive relief in cases where users make copies of out-of-print works. [FN167]

B. Private Letters and Diaries

Private letters and diaries represent the opposite kind of copyrighted works. They are "inaccessible, uncommercialized" works. [FN168] A private diary may never be seen by anyone other than the author. Private letters may never be seen by anyone other than the author and the recipient. [FN169]

How does the inaccessibility of these kinds of works affect copyright's fair use analysis? [FN170] The first response is that fair use does not arise if a work is completely inaccessible. [FN171] Private letters, diaries, journals, drafts, sketches, and the like, which are never made available to anyone are completely inaccessible. Because other people never gain access to these works, they cannot make any use of the works and hence fair use issues do not arise. [FN172]

In this regard, it should be emphasized that the fair use provision is a shield against a finding of infringement; it is not a sword which can be used to gain access to the work. A scholar, teacher, critic, or other user cannot twist the words of Section 107 from "the fair use is not an infringement" into "the fair use shall compel access." This is in keeping with the fact that the copyright system is only an incentive to authors to create and distribute their works; it is not a system that requires distribution as a condition for the grant of the rights. [FN173] It bears repeating that copyright's quid pro quo is not out of balance: Although the public receives no benefit *48 from these works, neither does the author benefit economically from the copyright rights. [FN174]

Fair use issues for private letters and diaries only arise in situations in which someone gains access to these materials. These issues will be discussed in the next subsection.

C. One-of-a-Kind Works: Letters and Diaries in Library Collections; Paintings and Sculptures in Museums

There is a group of works that are publicly accessible, but only to a slight degree. These are one-of-a-kind works such as private letters, diaries, or journals that have been given to a library, or original paintings or sculptures that have been given to a museum, which are accessible to the public. [FN175]

Two points need to be made about this class of works. First, these works are not "commercialized" as that term is defined in this Article. [FN176] Where only the original document is sold, the sale is of personal property rather than the transfer of copyright rights. [FN177] If the library or museum has been given the rights to make andsell copies of the work, then the *49 work has been "commercialized." [FN178] In this part of the Article, we are assuming that these rights have not been granted.

Second, these uncommercialized works are relatively inaccessible. Public access exists, but access is limited because there is only one copy (the original) and because anyone wanting access must go to the specific location where the original is located. Only those readers or scholars who have the time, money, and inclination to travel to the particular library will gain access. A competitor who wishes to learn and use the ideas contained in such works can do so if he is willing to pay that price. The vast majority of potential readers and scholars will not, as a practical matter, have access. The result is that members of the using public and competitors will be more restricted in the amount of expression that they can use than would be the case if the work were a mass-marketed work.

These two points combine to make a third point: There is no imbalance in copyright's quid pro quo. As with the case of purely private letters and diaries, the author receives no economic benefit from copyright because the work has not been commercialized. This is matched by the limited amount of public benefit. Copyright remains in balance. [FN179]

Fair use claims have arisen in situations where private letters, diaries, or journals have been given to a library and scholars, biographers, or others wished to make use of those works. Two recent decisions by the Second Circuit created great controversy by suggesting that the unpublished nature *50 of the works might almost preclude a finding of fair use. [FN180] Scholars and publishers reacted strongly to these decisions [FN181] and in 1992 Congress enacted Public Law 102-492 to reaffirm that the unpublished nature of a copyrighted work is not a bar to a finding of fair use. [FN182] Notwithstanding this enactment, the "uncommercialized" nature of a work is a factor that counts against a finding of fair use. [FN183] A proper understanding of accessibility and commercialization shows that there is no reason to change this analysis simply because of the lack of public access. In fact, any change in the analysis would be one that would alter the existing balance in copyright's quid pro quo.

The analysis in this Article contrasts with that of some of those who have written about fair use. For example, Judge Leval has argued that copyright's utilitarian goals are designed to promote the creation and dissemination of works rather than to encourage the production of shopping lists, memos, and private love letters. From this premise, he argues that works intended for publication are deserving of stronger copyright protection *51 than are private communications, and that the second fair use factor should focus on this distinction. [FN184]

Judge Leval's argument appears to be based on the view that copyright's quid pro quo is out of balance for unpublished works, [FN185] and that fair use should redress the imbalance unless the work is going to be published by the author.

The present Article refutes the view that there is an imbalance in copyright's quid pro quo in the case of uncommercialized works. Public access is not an issue for uncommercialized works; copyright protects these works for non-utilitarian reasons until the work is commercialized. Not until a copyright owner has taken advantage of copyright's incentive system should copyright concern itself with public access. [FN186]

On a practical level, Judge Leval's proposed solution is even more troublesome. It would require courts to determine which unpublished works are intended for publication and which are not--an examination into an author's subjective intent that is likely to be very difficult. The model presented in this Article avoids this problem. With no imbalance that needs to be corrected, there is no need to be concerned with the author's subjective intent.

D. Limited-Edition Works

The case of limited-edition works is very instructive because, surprisingly, they should be treated as "accessible, commercialized" works. "Limited-edition works" are humanly-accessible copyrighted works in which the copyright owner commits himself or herself to limiting the number of copies to a pre-set, fixed number of copies. [FN187] These differ from mass-market *52 works in that the publisher of the latter may make additional printings of a work if earlier printings have sold out.

Limited-edition works include art prints and photographs that are limited to a certain number of copies. [FN188] Such works share some of the inaccessibility attributes of one-of-a-kind works, yet the fair use analysis suddenly shifts.

The limited number of copies of limited-edition works serves to limit access by the public and access by competitors. In fact, if the limited number of copies have been acquired by private collectors, there may be less public access to these works than there is to an original painting in a museum or a private letter in a library.

Fair use analysis in this case, however, changes from that of one-of-a-kind works. This change occurs because a limited-edition work has been "commercialized" in the sense that multiple copies have been made and distributed to members of the public. [FN189] The author is making and distributing copies, thereby doing the acts that the Copyright Act encourages by giving the author these exclusive rights. Even though the author is only making a limited number of copies, the author is seeking the economic benefits made available under the Act.

Fair use analysis reflects the fact that the authors of limited-edition works are engaged in commercialization of the work. [FN190] It does this by liberalizing the uses that the public and competitors can make of the work. In particular, fair use seeks to allow competitors to use the work in order to create new works that build on the ideas and expression in the first work. In addition, fair use should allow scholars and others somewhat greater leeway in using expression from the first work than might be the case with works that have not been commercialized at all.

*53 The more interesting question is whether fair use analysis should treat limited-edition works differently from mass-marketed works. The striking difference in the degree of public access might suggest that fair use should treat these works differently. However, lack of access alone is not sufficient to justify a different fair use analysis. The question is whether the lack of access to a limited-edition work indicates an imbalance in copyright's quid pro quo that should be addressed in fair use analysis.

The answer is that limited-edition and mass-marketed works should be treated the same under fair use analysis. In both situations the authors are engaging in the kind of economic behavior that underlies copyright's incentive system. The mass-market author is opting for a larger quantity, perhaps at a lower per- unit price. The limited-edition author is opting for a lesser quantity, perhaps at a higher per-unit price. Both authors are producing multiple copies of the work, presumably seeking to optimize their income streams through decisions about quantity and price.

Is the public receiving an appropriate quid pro quo in this case? There are two reasons for thinking that the answer is yes. First, the author is limiting his or her use of copyright's economic rights. Unlike the publisher of a mass- marketed book who will print more editions if the demand grows, the publisher of a limited-edition work does not do so. Thus, the copyright owner of a limited-edition work forgoes the opportunity to gain additional income. The owner's limited use of the economic rights matches the public's limited access to the work.

Second, public demand for access is probably limited. Under normal economic theory, if there was a foreseeably larger demand by the public for the work, presumably the copyright owner would have printed a larger edition. [FN191] Even if the full extent of the demand was unforeseen, the public can still satisfy most of its demand through an increase in price in the resale market. [FN192] Thus public access is presumably available through normal market channels. [FN193]

*54 Implicit in this analysis is the recognition that the Copyright Act does not guarantee free or even cheap access to copies of copyrighted works. The copyright owner is free to set any price he or she desires for copies of the work. [FN194] Limits on public access that arise because of a high price do not justify changing fair use principles.

E. Documents Containing Trade Secrets

There are a great many documents that contain trade secret information. [FN195] Laboratory results, research data, customer information, marketing plans, and secure tests [FN196] are among the kinds of works that are protected by trade secrets. While some of the data in the documents may be factual, hence unprotected by copyright, a considerable portion of the documents may be protected by copyright.

Access to the ideas and expression in these works is restricted by limiting who is given access to the documents and through non-disclosure and confidentiality agreements.

Does the fair use provision of the Copyright Act give the recipient of these documents the right to disclose or make use of the information in ways not permitted by the non-disclosure and confidentiality agreement? The answer is no. Fair use exists as a defense to a claim of copyright infringement; it is not a defense to a claim of breach of confidence or breach of contract. [FN197]

*55 These works may also be protected by copyright. [FN198] In an action for copyright infringement, the law of trade secrets should be disregarded. That does not mean that any facts concerning secrecy should be ignored. However, they must be considered under copyright principles. For instance, if a company has trade secrets that it uses solely in-house, and an ex-employee has copied documents containing those secrets, the documents would be considered uncommercialized works. Fair use analysis would take into consideration the uncommercialized nature of the works and the fact that the works are inaccessible, [FN199] just as would be the case for private letters and diaries. [FN200]

On the other hand, if the company copied and disseminated documents containing trade secrets to outside companies as part of a trade secret licensing agreement, the documents would be treated as commercialized, inaccessible works for copyright purposes. The documents are commercialized because of their distribution, in a commercial setting, to outsiders. [FN201] They are inaccessible by virtue of the restrictions placed on them by the copyright owner.

The primary means by which access to these documents can be limited is contractual. [FN202] To the extent the company wants to prevent access, it should strive to implement appropriate non-disclosure agreements.

*56 On the other hand, access restrictions impede the normal workings of copyright's quid pro quo. When a copyright owner seeks the benefit of copyright law and commercializes the copyrighted work, the public can rightfully expect some appropriate amount of access. Contractual restrictions on access lead to an imbalance in copyright's quid pro quo.

This balance can be restored by allowing a greater scope of fair use than would be the case for an accessible, commercialized work. In performing the fair use analysis, courts should recognize that the "nature of the copyrighted work"--that it is commercialized, yet inaccessible--is a factor strongly supporting greater usage under fair use. [FN203]

F. Computer Programs

Computer programs raise inaccessibility issues because they are, in many cases, distributed in object code rather than in source code. As a practical matter, object code is not readable, even by skilled programmers. [FN204] In analyzing the fair use of these inaccessible works, it is necessary to distinguish three groups of computer programs.

First, there are computer programs that are never distributed to anyone. Lack of access to these programs is based on physical control over the programs, just as is the case for private letters and diaries. The copyright analysis appropriate to this group of programs is the same as for private letters and diaries or, in appropriate cases, one-of-a-kind works given to libraries or museums.

*57 Second, there are computer programs which are kept as trade secrets and not distributed outside a company. The discussion above concerning "Documents Containing Trade Secrets" applies to this group. [FN205] In short, for purposes of trade secret analysis, the user cannot claim greater use privileges based on provisions of the Copyright Act.

The third group of computer programs consists of those that are commercialized. [FN206] In some situations, these programs may be held as trade secrets and distributed subject to non-disclosure and confidentiality agreements or other valid and enforceable contractual restrictions. [FN207] A discussion of the enforcement of these contractual restrictions and actions for trade secret misappropriation is beyond the scope of this Article. In many cases, these programs are distributed in object code form and the proprietors rely on copyright for protection. [FN208] While there are few decisions on the *58 point, this group probably includes many of the mass-marketed computer programs for personal computers. [FN209]

For purposes of fair use analysis, these works should be treated as "inaccessible, commercialized works." [FN210] Access to these works is far different from the kind of access that is available to limited-edition and mass-marketed works. While copies are available so that the public can acquire them and obtain the benefits of the utilitarian aspects of the programs, public access is only to the object code version, a form that is not humanly understandable. [FN211] The ideas and expression in the programs are inaccessible, [FN212] and competitors are unable to produce competing versions unless and until they decipher the copyrighted work. [FN213] Without copies that can be perceived and studied by the public, the public is not getting the kind of access that is at the heart of the copyright system.

On the other hand, the works are commercialized in that the author has engaged in acts reserved exclusively to the copyright owner (i.e., making and distributing copies) by the copyright system. [FN214] Where the author is getting the economic benefits that the copyright system encourages, but the public and competitors are not getting the ability to study and learn the *59 ideas and expression of the work, there is an imbalance in copyright's quid pro quo. [FN215]

This balance can be reestablished, in part, by a judicious use of the fair use doctrine. [FN216] That doctrine should be used to privilege otherwise infringing acts that give the public and competitors the kind of access that normally exists for mass-market and other commercialized copyrighted works. The copyrighted work is the source code of the computer program. Disassembly [FN217] of the object code of a computer program does not give a competitor the source code, but it provides greater access to the ideas and expression in the program than does the object code. Using fair use to allow competitors to disassemble programs is one of the ways in which the copyright system can right the imbalance that exists with inaccessible, commercialized computer programs. [FN218]

This point gains support by thinking about the concept of an "intermediate copy." Where a competitor (or scholar) who disassembles a computer program and then produces an original work using only unprotected ideas and processes from the first work, one can think of the copy produced through disassembly as an "intermediate copy." [FN219] In this situation, the intermediate copy is the only possibly infringing copy. [FN220]

*60 The concept of an "intermediate copy" is both helpful and misleading. It is helpful in that it shows that this copy was used for research purposes so that a non-infringing final work could be produced. The intermediate copy was not sold to the public. The use that was made comes squarely within the kind of purposes for which the fair use doctrine was designed. [FN221]

One can emphasize this point by placing it in another light. Contrast this situation with one in which a computer program is commercialized in human- readable form. Suppose a competitor studies that form, without making any "intermediate" copies, in order to produce its final product. No one would even claim infringement in this second case. In this second situation, no intermediate copy is needed.

In contrast, in the first situation, the copyright owner has obtained, by commercializing a computer program in object code form, the benefit of copyright without disclosing a human-readable version of the work. The public does not receive the full amount of access. In comparing these two situations, the copyright system should not give a preferred position to the author who publishes a humanly-unreadable version over the author who publishes a readable version. Those who claim that a competitor should not be permitted to disassemble an object code version of a computer program want copyright law to give this preferred position to authors who seek the benefits of copyright without providing the public benefit of promoting the learning and knowledge. [FN222]

On the other hand, the concept of an "intermediate copy" is also misleading. If a competitor disassembles a computer program in order to study the ideas and expression in it, the fair use analysis should not depend upon whether that competitor has successfully used the ideas in another work or not. A competitor might not have had time to complete the work or might choose not to produce a work. In either of these situations, the disassembled work would not be "intermediate." However, the use--studying the first program--is the same in those two situations as it is if the competitor has already completed a new work.

*61 Nor should the fair use analysis depend on the timing of a complaint so that if the complaint is filed before the competitor completes the new work, there would not be fair use; whereas if more time has elapsed, the same use would be fair. Copyright analysis should not hinge on such a matter of timing. What is important is that accessibility exist under conditions where future growth and learning are possible.

In the situation in which a competitor disassembles a computer program and uses the ideas in order to produce a non-infringing competing program, [FN223] the fair use analysis used by the Ninth Circuit in Sega Enterprises Ltd. v. Accolade, Inc. [FN224] is correct in most respects. [FN225] A number of points are worth emphasizing. [FN226]

*62 First, the "purposes" of the use [FN227] in this situation include both research and producing a competing product. The parties may argue that the use is one or the other, but of course, both uses are present.

The issue is whether one of these purposes is condemned and the other is praised by the copyright system, or whether the copyright system takes some other view. The fair use provision explicitly declares that scholarship and research are among the kinds of uses which can be fair use. Therefore, the assertion that the use was for research should be treated as a plus for fair use analysis. More importantly, the purpose is to gain the kind of access that is available for traditional works under the standard copyright paradigm--works such as mass-marketed books and limited-edition works. The purpose is not to get more than is available for these other kinds of works.

On the other hand, the purpose of producing a competing product is a commercial one, hence, at first glance, suggesting that this weighs in favor of the copyright owner. However, as the Supreme Court has noted, one must go beyond the label of "commerciality" lest that label "swallow nearly all of the illustrative uses listed in the preamble paragraph of Section 107, including news reporting, comment, criticism, teaching, scholarship, and research, since these activities 'are generally conducted for profit in this country."' [FN228] One decides whether a particular "commercial" use is fair or not by keeping in mind that the primary goal of copyright is to promote the "progress of Science." [FN229] Science will not progress very far or very fast if competitors are denied access to the ideas of the works of their competitors. Thus, the fact that a competitor is trying to learn the ideas of a copyrighted work should in no way be treated as a strike against a fair use defense. [FN230] In fact, because copyright wants competitors to learn and use *63 the ideas of others, this purpose would seem to be either a strike in favor of fair use or irrelevant to the fair use analysis. [FN231]

Having said this, one must also reiterate that the competitor must create his or her own expression. Piracy of the first work is not allowed. It is in the context of a competitor who seeks to use the expression of the first author in a competing work that has led courts, properly, to emphasize that competing uses are disfavored in fair use analysis. [FN232]

The second point is that the nature of the computer program is that it is an "inaccessible, commercialized" work. The second fair use factor requires that courts inquire into the "nature of the copyrighted work." [FN233] The fact that the program is inaccessible and commercialized means that copyright's quid pro quo is out of balance. As noted elsewhere, the operative distinction in the second fair use factor should be whether the work has been "commercialized," not whether the work has been "published." [FN234]

The third point is that there is no cognizable market impact from the creation of the disassembled program. The Supreme Court has explained that the only cognizable market impacts are in those markets which the copyright owner will enter. [FN235] The fact that computer programs are distributed in object code form and, almost never, in source code form, indicates that the market for source code versions is not one which copyright owners are going to enter.

*64 Moreover, even if the source code market were a cognizable market, there is no market harm. Under the standard copyright paradigm, the copyright owner would make one sale to a competitor and the competitor would have access to the ideas in the work. Where the owner makes a sale of a copy of a machine- readable version, the author has still made that one sale, but the competitor has not yet obtained a version that gives access to the ideas. The disassembly of that version puts the competitor part of the way towards a similar position. If a competitor has to go through the extra effort of disassembly to try to get some of the ideas, [FN236] the competitor should not be penalized. Similarly, a copyright author who publishes a machine-readable version should not be preferred to one who publishes a human-readable version. [FN237]

Related to this is the fact that the only market impact that is relevant is the impact from the creation of one disassembled copy; it is not from the creation and sale of non-infringing competing works that incorporate unprotected ideas and processes. [FN238] Again the standard copyright paradigm makes this clear. A competitor who buys a copy of a copyrighted work and studies it in order to produce a non-infringing competing work is doing exactly what the copyright system hopes he or she will do. Copyright does not insulate a copyright author from competition.

G. Braille, Foreign Language Works, and Works on Records or CDs

Computer programs are not the only "inaccessible, commercialized" works. [FN239] Blind people cannot read printed words; [FN240] foreign language works may be in print, but they are inaccessible to people who do not understand that language; a mass-marketed record or CD may contain a performance of a new musical work, but the score of the musical composition *65 is inaccessible if the score has not been published. [FN241] For these kinds of works, the author has begun commercialization of the work, yet access to the information and expression in the work remains somewhat restricted.

As with computer programs, copyright's quid pro quo is out of balance, and fair use can and should be used to rectify that imbalance. The House Report on the 1976 Copyright Act reaches this conclusion with regard to making a Braille or recorded oral copy for a blind person, although it does not provide any analysis showing how it reached this conclusion. [FN242] The copyright model described in this Article helps show why the House Report reached the proper conclusion. An equally good case can be made that it would be fair use for a historian who does not read French to commission a translation of a French-language history of the battle and siege of Quebec in order to be able to comment on, criticize, and build on the ideas in that work, where there is no authorized translation of that work. Similarly, it should be fair use for someone to commission another person to transcribe the score of a musical work from a record or CD containing a performance *66 of the work, where there is no authorized translation or published score. [FN243]

In the case of the translation, one should note that if the historian could read the work in its original French, he could extract the ideas and freely use them to create his own work. Such a use--the creation of new works by competitors--is a paradigmatic example of the kinds of uses that the copyright system is designed to encourage. It is paradigmatic in several respects. First, the original work has been commercialized and is available to the public. Second, the use being made of the translation is for research and scholarship. The second historian can then make use of the ideas in creating the later work. And third, the public benefits from having available to it both the original work and the possibility of a new work. Whether the new work restates the ideas in a new form, criticizes the ideas, or comments on the ideas, the public benefits from having both works.

As far as the public is concerned, there is virtually no difference between the situation in which the historian can read the original work in French and that where the historian has commissioned a translation. In both cases the public gains the benefit of having the second historian have access to the original work.

Indeed, it is not important whether the second historian actually completes the second work. As noted above, [FN244] the "intermediate copy" analysis does not require the historian to complete a new work.

In terms of accessibility and commercialization, the fair use analysis in the case of foreign language works and musical compositions distributed on records and CDs is the same as for computer programs. In all of these instances, the author has begun commercial exploitation of the copyrighted work, yet has done so in a way that prevents humanly-readable access to the ideas and information in the original work. The user undertakes a private effort to convert the work into humanly-readable form and has the possibility of creating a new work using the ideas (and perhaps a "fair use" amount of the expression) of the first work. The copy made is not published, thereby leaving the market for the original work (the source code of *67 the computer program or the score of the musical composition) or an authorized translation open to the copyright owner. And, if the second person produces a new work, the new work builds upon the old work, providing the kind of public benefit that the copyright system seeks to promote. The conclusion one reaches is that the creation of a private intermediate work is fair use.

The analysis of computer programs, foreign-language works, and works on records and CDs is very fact-sensitive. For example, as between two historians who commission English translations of a French-language history, the historian who does not understand French is in a better position to assert fair use than the historian who understands French. This result is appropriate when one recognizes the importance of access in producing public benefits under the copyright system. It is the accessibility of the work to competitors that helps drive the production of new works, works that may reexpress the ideas of the original work and reach new audiences. The historian who understands French has sufficient access to produce new works without commissioning a translation. The historian who does not understand French must commission the translation in order to gain access.

Converting a work into Braille or a recorded oral reading so that a blind person can perceive it is just another form of foreign translation. While the House Report, in reaching its conclusion that making a copy for a blind person is fair use, did not make use of the framework of analysis put forward in this Article, it does mention many of the correct reasons. For example, the House Report notes that these works "are not usually made by the publishers for commercial distribution." [FN245] This sentence suggests two important points for the analysis. First, the work has been commercialized--a "publisher" has put out a printed version. Second, the work is inaccessible to blind people. One implication is that if the publisher had created a Braille or recorded oral version, then fair use might not be found so readily. The analysis presented in this Article agrees.

The House Report also properly distinguishes between commercial uses being made of the new version and private uses. It states that

the making of multiple copies or phonorecords of a work for general circulation requires the permission of the copyright owner, a problem addressed in section 70 of the bill, the making of a single copy or phonorecord as a free service for a blind person would properly be considered a fair use under section 107. [FN246]

*68 The analysis in this Article agrees with part of this sentiment. Commissioning one copy in order to obtain a copy of a commercialized work in a format that is humanly accessible (i.e., a Braille copy for a blind person) is far different from making a Braille copy that would then be sold commercially to many people.

In one respect, the present Article disagrees with the House Report. The House Report speaks about making a Braille or recorded oral copy "as a free service" for a blind person. [FN247] The analysis in this Article disagrees with the restriction that the service must be free. If a blind person purchases a copy of a book and commissions someone to make a Braille copy or to make a recorded oral version, the copyright owner is not harmed economically. The sale to the blind person has been made; the royalties to the copyright owner collected. That the blind person has to pay additional amounts to someone to translate the work into a form accessible to the blind does not affect the copyright owner's market. Put another way, if the copyright owner had published a Braille edition, the blind person would still have bought one copy, but that copy would have been the Braille edition, not the print edition.

H. Television Broadcasts and News Clipping Services

Many television programs are protected by copyright, yet access is limited. Two situations can exist: (1) copies of the program are publicly sold or distributed; and (2) copies are not publicly available.

The first situation presents no unusual fair use analysis problems. Where copies of a broadcast program are made and sold, there is public access. Fair use analysis should treat these kinds of copyrighted works in the same way it treats mass-market or limited-edition works. [FN248] The works are "accessible, commercialized" works.

The second situation presents a very different fair use situation. Many news programs on local television stations fall into this category. [FN249] Broadcast television programs for which no copies are made and sold by the *69 copyright owner should be treated as "commercialized, semi-accessible works" for purposes of fair use analysis. These programs are "commercialized" in that the copyright author has done one of the acts ("public performance") which the Copyright Act reserves exclusively to the copyright owner. [FN250]

The works are only "semi-accessible." Access to these programs is quite different from access to limited-edition and mass-market works: the only public access to a program may occur in the fleeting moments when the work is broadcast. Whereas the public may be able to continue to receive access to other commercialized works such as limited-edition and mass-market works by locating and viewing such works, there is little or no continuing access to these broadcast television programs. [FN251]

This "semi-accessibility" represents a departure from the kind of ongoing access that the copyright system seeks to promote. In order for a copyrighted work to advance learning and knowledge, the public and competitors need access over a period of time, not just at an instant of time. Semi-accessibility does not give the public the amount of benefit that the copyright system seeks. Thus, there is an imbalance in the quid pro quo of the copyright system. [FN252]

To reestablish the balance, fair use should allow competitors greater access and use of the work in order to create new works that build on the ideas and expression in the first work than would be the case if the program had never been broadcast. [FN253]

*70 Thinking of television programs as "semi-accessible, commercialized" works suggests that courts should be more sensitive than they have been to the primary goals of copyright and to copyright's quid pro quo balance when they do their fair use analysis in these cases. [FN254] In particular, it might be the case that fair use should be found where someone makes copies of television news broadcasts so that scholars and others can study them. It might also be fair use where television news clipping services make and distribute small portions of news broadcasts for which the copyright owner does not provide copies for sale. [FN255] Indeed, the unique circumstances involving television clipping services and the goal of providing more access to television news has led to the introduction of federal legislation to include this kind of activity among those listed in the fair use section of the Copyright Act. [FN256]

I. Accessible, Uncommercialized Works

The Quid Pro Quo Matrix [FN257] contains one quadrant that has not yet been discussed. This is the quadrant for "accessible, uncommercialized" works. One-of-a-kind public sculptures and public murals where the author has not authorized copies are examples of these kinds of works. [FN258]

*71 Copyright's quid pro quo is out of balance for "accessible, uncommercialized" works. The public is receiving the benefit of accessibility, albeit limited to the one copy in one location, while the copyright owner has not received the benefits of commercialization.

Because this imbalance is under the control of the copyright owner, it does not present any novel fair use issues. At any time, the copyright owner is free to authorize copies. Photographs, postcards, two-dimensional prints, and, for the sculpture, three-dimensional copies, can be authorized. Thus the uncommercialized status of the work can be changed to a commercialized status at any time by the copyright owner.

IV. ACCESSIBILITY, COMMERCIALIZATION, AND WITHHOLDING INJUNCTIVE RELIEF

There are some uses of copyrighted works which might not be fair use, but where the public access to the work is so limited that some other mechanism should be found to redress the imbalance in copyright's basic quid pro quo. This Part will discuss two such uses: first, some cases of commercial copying of out-of-print books for educational use and second, some commercial copying of unavailable broadcast television news programs. [FN259] In these cases, courts should award monetary relief, but consider withholding injunctive relief. [FN260]

The uses considered in this Part are meant to be illustrative rather than exclusive. They are included to indicate additional implications of *72 using the concepts of commercialization and accessibility in copyright analysis.

A. Out-of-Print Works

A teacher who desires to assign an out-of-print novel to her class faces a problem. When a copyright owner stops selling copies of the work, access begins to diminish. As noted above, the public and competitors may be able to obtain limited access to the work by finding second-hand copies. [FN261] However, the second-hand book market may not be adequate to meet the timing and volume requirements where a professor assigns that particular work in a class. [FN262]

If it would not be fair use for a commercial copying establishment to make copies of an out-of-print novel for use by students in a class, should the copyright owner be given an injunction? [FN263] What are the equities in such a case? Granting an injunction preserves the copyright owner's exclusive right to control the commercialization of the work through making and selling copies. The owner can obtain an economic return by requiring the user to obtain permission and pay a royalty. [FN264] If the royalty fee is higher than the user is willing to pay or if the owner refuses to grant permission, then the user will be unable to make the copies. If the flow of requests for making copies is great enough, the copyright owner might decide to put out a new edition of the work.

*73 On the other hand, granting an injunction against making and selling copies of an out-of-print novel can serve to reinforce the limited access to the work, and can inhibit socially productive uses of that work such as teaching.

The Supreme Court has remarked, in dictum, that close cases of fair use might warrant the judicious withholding of injunctive relief. [FN265] The Court contrasted "simple piracy" (where, presumably, injunctive relief is almost always appropriate) and cases where there is "a strong public interest in the publication of the secondary work." [FN266] While the Court made its remarks in the context of a parody--a situation in which a second author has created a new work the publication of which might further the public interest--the same rationale of considering the public interest should be considered in other situations as well.

The public benefit from allowing a commercial copying business to make copies of an out-of-print novel so that a professor could assign the work for a class would seem to outweigh the harm to the copyright owner and further the public interest. [FN267] If the circumstances indicate that this use would not be fair use, it would seem appropriate to grant monetary relief to the copyright owner without granting an injunction. [FN268]

*74 B. Broadcast Television News Programs

The case of broadcast television news programs for which copies are not sold or available from the copyright owner presents a similar situation where courts should hesitate to grant an injunction. As noted above, these works are "semi- accessible, commercialized" works. [FN269] The copyright owner has done those acts ("public performance") that the copyright system seeks to encourage, yet access is limited to the fleeting time during which the program is broadcast. Allowing commercial copying and distribution of these broadcasts would rectify this imbalance in copyright's quid pro quo. The copyright owner should be entitled to revenues from the distribution of copies, hence monetary relief is appropriate. On the other hand, withholding an injunction would further copyright's goals of promoting learning and knowledge by allowing public access to the ideas and expression in these works. [FN270]

The two situationsdiscussed in this Part are intended to illustrate the implications of the roles of accessibility and commercialization in copyright theory. As Judge Leval has noted, "[A]nalysis of [the propriety of granting an injunction] should reflect the underlying goals of the copyright law to stimulate the creation and publication of edifying matter." [FN271] While injunctions should issue in most copyright infringement cases, injunctions should not issue if they would serve to undermine copyright's fundamental goals.

CONCLUSION

Copyright's goals of promoting learning and knowledge can be accomplished only if copyrighted works are commercialized and accessible to the public and to competitors. This does not mean that every copyrighted work has to be accessible. Copyright protects private letters, diaries, and other works to which there is no public access. These works are not commercialized. Lack of access to such works is not contrary to the copyright statute or to the Constitution.

Accessibility becomes an issue when a copyright owner commercializes the work--doing any of the acts reserved exclusively to the copyright owner by the Act. These are the acts that copyright's incentive system encourages copyright owners to do. When an owner attempts to reap the rewards *75 offered by the copyright system, the public should expect to receive access. Without public access, there is an imbalance in copyright's basic quid pro quo.

Copyright offers two major avenues by which this imbalance can be rectified. One of these is through a judicious application of fair use. Different kinds of works present different kinds of accessibility issues under fair use. The paradigmatic kind of copyrighted work--mass marketed books--presents no accessibility issues. Somewhat surprisingly, this Article shows that limited- edition works likewise present no accessibility issues. Private letters and diaries, and one-of-a-kind works donated to museums and libraries also present few novel fair use questions because those works are not commercialized in the copyright sense.

There are a variety of works that are "inaccessible, commercialized" works. These include some computer programs, foreign language works, and works on records and CDs. For these works, the copyright owner is attempting to benefit from the incentives given by the copyright system, but the public is not getting full access. For these works, fair use should be used to allow the public and competitors to make humanly readable copies of these works so that the works can be studied. Allowing this gives the public somewhat the same kind of access that the public has to mass-marketed books and other accessible copyrighted works. This restores the copyright balance.

Similarly, some television news broadcasts are "commercialized, semiaccessible" works. The copyright owner is performing the work publicly, but public access may be limited to the short period of time during which the broadcast is made. The public does not get the kind of on-going access that will promote learning and knowledge, thus, there is an imbalance in copyright's quid pro quo. Allowing scholars and others the privilege of making a copy in order to study and learn from it would rectify this imbalance.

The second avenue by which the imbalance in copyright's quid pro quo can be remedied is the withholding of injunctive relief in situations where the amount of access offered for a commercialized copyrighted work is less than optimal. One can envision situations in which this remedy should be withheld for out-of- print works and some commercial uses of broadcast television news programs. The copyright mandate to benefit society by encouraging the creation and dissemination of new copyrighted works requires that those works be accessible in order for the public to have the benefit of the knowledge and information contained in the works. An injunction against an infringer reinforces the inaccessibility of the work, preventing the public access which is so important for copyright. Where a *76 copyright owner shows no interest in providing access and where damages will adequately compensate the copyright owner, a court should generally award damages and refuse the injunction.

The use of these two avenues, keeping in mind copyright's goals of promoting learning and knowledge, will enable the public to obtain an appropriate amount of benefit whenever copyright owners avail themselves of the exclusive rights given by the Copyright Act.

[FNaa1]. Professor of Law and Director of the Program in Law and Technology, University of Dayton School of Law. B.A. 1963, Reed College; Ph.D. (Mathematics) 1968, University of Oregon; J.D. 1977, Stanford University. This Article was prepared with the aid of a research grant from the University of Dayton School of Law.

The author thanks Tony Clapes, Paul Goldstein, Dennis Karjala, David Lange, Marshall Leaffer, David Rice, Ray Patterson, and Pam Samuelson for their perceptive comments on an earlier draft of this Article. Students Ginette Boersch, Jerry Goh, Mitchell Rubin, and Ajit Vaidya provided valuable research assistance.

[FN1]. 1 PAUL GOLDSTEIN, COPYRIGHT § 1.1 (1989) (Copyright law seeks "to encourage the widest possible production and dissemination of literary, musical and artistic works."); accord Fogerty v. Fantasy, Inc., 114 S. Ct. 1023, 1030 (1994) ("[C]opyright law ultimately serves the purpose of enriching the general public through access to creative works."); Feist Publications, Inc. v. Rural Tel. Serv. Co., 499 U.S. 340, 349 (1991) ("The primary objective of copyright is not to reward the labor of authors, but '[t]o promote the Progress of Science and useful Arts."') (alteration in original) (citation omitted); Twentieth Century Music Corp. v. Aiken, 422 U.S. 151, 156 (1975) ("Creative work is to be encouraged and rewarded, but private motivation must ultimately serve the cause of promoting broad public availability of literature, music, and the other arts.") (footnote omitted).

[FN2]. To oversimplify, "access" refers to access by the using public and by other authors to the ideas and expression in a copyrighted work. The concept is discussed more completely infra notes 37-49 and accompanying text.

[FN3]. The present federal copyright statute is the Copyright Act of 1976, as amended. 17 U.S.C. §§ 101-1010 (1994).

[FN4]. Copyright law protects "original works of authorship." 17 U.S.C. § 102(a) (1994). There is no requirement that the works be disseminated. Examples of undisclosed works are numerous. In addition to the vast quantity of shopping lists and other scribbled notes that people produce on a regular basis, there are personal diaries, drafts of stories, artistic sketches, unfinished paintings, and other materials that people generate and then destroy.

[FN5]. The Copyright Act only requires that works be "fixed in any tangible medium of expression, now known or later developed, from which they can be perceived, reproduced, or otherwise communicated, either directly or with the aid of a machine or device." 17 U.S.C. § 102(a) (1994) (emphasis added). Works that are fixed, but that cannot be perceived without the help of a machine, include computer programs that operate kitchen appliances or industrial equipment, and they are disseminated in the form of computer chips or on media that are not humanly readable. More common computer programs, such as word processing programs, are distributed in the form of object code that is not humanly perceptible in that form. There is limited direct accessibility in that the screen displays are perceptible when the program is used in a computer. Similarly, musical works distributed on records, tapes, or CDs are not humanly perceptible in that form, but there is limited direct accessibility in that the musical compositions and performances are perceptible when played by an appropriate machine.

[FN6]. There are some thoughtful articles and books that touch on this subject. See L. RAY PATTERSON & STANLEY W. LINDBERG, THE NATURE OF COPYRIGHT: A LAW OF USERS' RIGHTS (1991); Margaret Chon, Postmodern "Progress": Reconsidering the Copyright and Patent Power, 43 DEPAUL L. REV. 97 (1993); Anthony L. Clapes, Confessions of an Amicus Curiae: Technophobia, Law, and Creativity in the Digital Arts, 19 U. DAYTON L. REV. 903 (1994); Duncan M. Davidson, Common Law, Uncommon Software, 47 U. PITT. L. REV. 1037 (1986); Dennis S. Karjala, Copyright Protection of Computer Documents, Reverse Engineering, and Professor Miller, 19 U. DAYTON L. REV. 975 (1994); Arthur R. Miller, Copyright Protection for Computer Programs, Databases, and Computer- Generated Works: Is Anything New Since CONTU?, 106 HARV. L. REV. 977 (1993); Francis M. Nevins, Jr., Availability: The Hidden Value in Copyright Law, 15 COLUM.-VLA J.L. & ARTS 285 (1991); L. Ray Patterson, Copyright and "the Exclusive Right" of Authors, 1 J. INTELL. PROP. L. 1 (1993); Pamela Samuelson, CONTU Revisited: The Case Against Copyright Protection for Computer Programs in Machine-Readable Form, 1984 DUKE L.J. 663; Barry J. Swanson, The Role of Disclosure in Modern Copyright Law, 70 J. PAT. & TRADEMARK OFF. SOC'Y 217 (1988).

In addition, many authors note or assume the importance of access to achieving copyright's goals. Some of the more recent articles include: Pierre N. Leval, Toward a Fair Use Standard, 103 HARV. L. REV. 1105 (1990); Stanislaw J. Soltysinski, Legal Protection for Computer Programs, Public Access to Information and Freedom of Competitive Research and Development Activities, 16 RUTGERS COMPUTER & TECH. L.J. 447 (1990); Lloyd L. Weinreb, Fair's Fair: A Comment on the Fair Use Doctrine, 103 HARV. L. REV. 1137 (1990); Carl H. Settlemyer III, Note, Between Thought and Possession: Artists' "Moral Rights" and Public Access to Creative Works, 81 GEO. L.J. 2291 (1993); Stephen S. Zimmermann, Comment, A Regulatory Theory of Copyright: Avoiding a First Amendment Conflict, 35 EMORY L.J. 163 (1986).

[FN7]. Sega Enters. Ltd. v. Accolade, Inc., 977 F.2d 1510 (9th Cir. 1992); Atari Games Corp. v. Nintendo of Am. Inc., 975 F.2d 832 (Fed. Cir. 1992).

[FN8]. Disassembly "is a procedure for translating the machine language program into an assembly language program." OFFICE OF TECHNOLOGY ASSESSMENT, FINDING A BALANCE: COMPUTER SOFTWARE, INTELLECTUAL PROPERTY AND THE CHALLENGE OF TECHNOLOGICAL CHANGE 146 (1992). For an excellent and readable discussion of how disassembly is accomplished, see Andrew Johnson-Laird, Software Reverse Engineering in the Real World, 19 U. DAYTON L. REV. 843 (1994).

[FN9]. This view is found, for example, in an amicus curiae brief filed by eleven copyright law professors in Sega Enters. Ltd. v. Accolade, Inc., 977 F.2d 1510 (9th Cir. 1992), in which they wrote:

It is fundamental to the copyright balance that ideas and other unprotected elements be accessible to others. If access to unprotected elements in publicly distributed works can be technologically denied and enforced through copyright, the fundamental copyright equation is thrown out of balance.

Brief Amicus Curae [sic] of Eleven Copyright Law Professors in Sega Enterprises Ltd. v. Accolade, Inc., 33 JURIMETRICS J. 147, 155 (1992) [[hereinafter Sega Brief].

[FN10]. This view is found, for example, in a recent article by Arthur Miller, who writes:

[T]he claim that decompilation is necessary in order to gain "access" to a program's underlying ideas is misconceived. Nothing in the Copyright Act suggests that making every aspect of a work available to the general public is a precondition to copyright protection. The basic objective of copyright is for society to benefit from the availability of creative works--that the progress of science and the useful arts be "promoted"--whether or not the literal expressions or underlying ideas of those works are directly available to the public.

Miller, supra note 6, at 1028-29 (footnotes omitted).

[FN11]. See infra notes 19-22 and accompanying text.

[FN12]. See supra note 1.

[FN13]. To oversimplify, "commercialization" means doing any act that is likely to produce a direct economic return. This includes things such as the sale of books, public performances of music, and public displays of sculptures. The concept is discussed more completely infra notes 50-68 and accompanying text.

[FN14]. Patterson, supra note 6, at 43 (noting that copyright's "main function is to protect the entrepreneur against those who would pirate a work for competitive sale in the marketplace").

[FN15]. Mazer v. Stein, 347 U.S. 201, 219 (1954) ("Sacrificial days devoted to such creative activities deserve rewards commensurate with the services rendered.").

[FN16]. Stewart v. Abend, 495 U.S. 207, 228-29 (1990) ("[N]othing in the copyright statutes would prevent an author from hoarding all of his works during the term of the copyright.").

[FN17]. The Copyright Act defines "publication":

"Publication" is the distribution of copies or phonorecords of a work to the public by sale or other transfer of ownership, or by rental, lease, or lending. The offering to distribute copies or phonorecords to a group of persons for purposes of further distribution, public performance, or public display, constitutes publication. A public performance or display of a work does not of itself constitute publication.

17 U.S.C. § 101 (1994). Although they are related, publication is not synonymous with accessibility. For example, Dr. Martin Luther King's famous speech, "I Have a Dream," which was delivered orally to hundreds of thousands of people, was held not to have been "published," yet was widely accessible. King v. Mister Maestro, Inc., 224 F. Supp. 101 (S.D.N.Y. 1963). Computer programs that operate equipment (such as car radios) are presumably published, but are not "accessible."

[FN18]. Francis Nevins makes this point obliquely: "Copyright law should be so structured and interpreted that every creative work once offered to the public continues thereafter to be appropriately accessible to the public, except where this goal conflicts with another policy value of an equal or higher order." Nevins, supra note 6, at 286 (emphasis added).

[FN19]. See supra note 1. Alternative views of copyright's goals are considered infra Part II.

[FN20]. "Primary goals," "constitutional goals," and "utilitarian goals" will be used interchangeably throughout this Article.

[FN21]. The full clause reads: "The Congress shall have Power To 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." U.S. Const. art. I, § 8, cl. 8. The stylistic usage of the time indicates that part of the sentence deals with copyrights and part with patents. The copyright portion pairs "Science" with "Authors" and "Writings," while the patent portion pairs "useful Arts" with "Inventors" and "Discoveries." In this context, "Science" means knowledge or learning. Patterson & Lindberg, supra note 6, at 48; see also Oxford English Dictionary 648 (2d ed. 1989) ("Science" is defined as "1.a. The state or fact of knowing; knowledge or cognizance of something specified or implied; also, with wider reference, knowledge (more or less extensive) as a personal attribute.").

[FN22]. See supra note 21.

[FN23]. The Supreme Court has stated that "[t]he economic philosophy behind the clause is the conviction that encouragement of individual effort by personal gain is the best way to advance public welfare." Mazer v. Stein, 347 U.S. 201, 219 (1954). Other authors have also emphasized that copyright law has economic underpinnings. See, e.g., William M. Landes & Richard A. Posner, An Economic Analysis of Copyright Law, 18 J. LEGAL STUD. 325 (1989).

[FN24]. By encouraging new works to be produced, the public receives a constant flow of new expressions of ideas in literature, music, drama, dance, painting, sculpture, map-making, movies, records, tapes, and architecture. The availability of this profusion of new works is a primary goal of the copyright system. Nevins, supra note 6, at 285.

[FN25]. For example, televised works may reach peoplewho rarely read books, and historical novels may reach people who do not read histories.

[FN26].

Literature and the arts are centrifugal, aiming at a wide variety of audiences with different tastes. We cannot say that one novel treating the theme, say, of man's continuing struggle with nature is in any ultimate sense "better" than another novel--or musical composition or painting--on the same subject. The aim of copyright is to direct investment toward abundant rather than efficient expression.

Paul Goldstein, Infringement of Copyright in Computer Programs, 47 U. PITT. L. REV. 1119, 1123 (1986). As another commentator writes,

Wary of unreliable value judgments about art and incapable of predicting which of even the most successful authors' future works will capture or recapture the public's fancy, the mature copyright paradigm embraces all literary and artistic works simply by virtue of their being creations and leaves the assessment of merit entirely to the market.

J.H. Reichman, Design Protection and the New Technologies: The United States Experience in a Transnational Perspective, 19 U. BALT. L. REV. 6, 142-43 (1989) (footnotes omitted).

[FN27]. Twentieth Century Music Corp. v. Aiken, 422 U.S. 151, 156 (1975); accord Fogerty v. Fantasy, Inc., 114 S. Ct. 1023, 1030 (1994); Sony Corp. of Am. v. Universal City Studios, Inc., 464 U.S. 417, 429 (1984); United States v. Paramount Pictures, Inc., 334 U.S. 131, 158 (1948).

[FN28].

Thus, the copyright law seeks to establish a delicate equilibrium. On the one hand, it affords protection to authors as an incentive to create, and, on the other, it must appropriately limit the extent of that protection so as to avoid the effects of monopolistic stagnation. In applying the federal act to new types of cases, courts must always keep this symmetry in mind.

Computer Assocs. Int'l Inc. v. Altai Inc., 982 F.2d 693, 696 (2d Cir. 1992).

[FN29].

In no case does copyright protection for an original work of authorship extend to any idea, procedure, process, system, method of operation, concept, principle, or discovery, regardless of the form in which it is described, explained, illustrated, or embodied in such work.

17 U.S.C. § 102(b) (1994). This doctrine was forcefully stated more than one hundred years ago in Baker v. Selden, 101 U.S. 99 (1879).

[FN30]. This expression is a convenient shorthand for the distinction between expression on the one hand and ideas, procedures, processes, and similar unprotected elements on the other. Although the shorthand uses the word "idea," one must be careful to recognize that the unprotected elements include more than just ideas. See, e.g., Lotus Dev. Corp. v. Borland Int'l, Inc., 49 F.3d 807, 815 (1st Cir. 1995); Gates Rubber Co. v. Bando Chem. Indus., 9 F.3d 823 (10th Cir. 1993).

[FN31].

For as Justice Story explained, "In truth, in literature, in science and in art, there are, and can be, few, if any, things, which in an abstract sense, are strictly new and original throughout. Every book in literature, science and art, borrows, and must necessarily borrow, and use much which was well known and used before."

Campbell v. Acuff-Rose Music, Inc., 114 S. Ct. 1164, 1169 (1994) (quoting Emerson v. Davies, 8 F. Cas. 615, 619 (C.C.D. Mass. 1845) (No. 4436)). Law review articles, full of footnotes that attribute ideas to earlier authors, amply illustrate the truth of this assertion.

[FN32]. This doctrine is codified at 17 U.S.C. § 107 (1994).

[FN33]. Fair use is discussed more extensively infra Part III.

[FN34]. See Weinreb, supra note 6.

[FN35]. But see Jessica Litman, Copyright and Information Policy, 55 LAW & CONTEMP. PROBS., Spring 1992, at 185, 202 (noting that the protection of authors' privacy interests is "largely incidental to its protection of their economic copyright interests" and questioning whether the extension of copyright to unpublished works was intended "to incorporate common law copyright's solicitude for privacy" into copyright law). For purposes of this Article, it does not matter whether copyright's protection of privacy is a specific goal or a fortuitous byproduct of the copyright's protection of works from the moment that they are fixed in a tangible medium of expression.

[FN36]. Pub. L. No. 100-568, 102 Stat. 2853 (1988).

[FN37]. Access to a work sometimes coincides with the time when the copyright owner begins to commercialize the work. For some works, however, access and commercialization occur at different times. See infra notes 63-66 and accompanying text.

[FN38]. The Supreme Court has stated: "Creative work is to be encouraged and rewarded, but private motivation must ultimately serve the cause of promoting [[its] broad public availability." Twentieth Century Music Corp. v. Aiken, 422 U.S. 151, 156 (1975) (emphasis added).

[FN39]. In Feist Publications, Inc. v. Rural Tel. Serv. Co., 499 U.S. 340 (1991), the Court emphasized this point:

The primary objective of copyright is not to reward the labor of authors, but "[t]o promote the Progress of Science and useful Arts." To this end, copyright assures authors the right to their original expression, but encourages others to build freely upon the ideas and information conveyed by a work.

Id. at 349-50 (alteration in original) (emphasis added) (citations omitted); accord Fogerty v. Fantasy, Inc., 114 S. Ct. 1023, 1030 (1994) (stating that a successful defense to a copyright infringement action by an author of a second work "may further the policies of the Copyright Act every bit as much as a successful prosecution of an infringement claim by the holder of a copyright"). Commentators also acknowledge that

intellectual creative activity is in part derivative. There is no such thing as a wholly original thought or invention. Each advance stands on building blocks fashioned by prior thinkers. Second, important areas of intellectual activity are explicitly referential. Philosophy, criticism, history, and even the natural sciences require continuous reexamination of yesterday's theses.

Leval, supra note 6, at 1109 (footnote omitted); accord Diane Conley, Author, User, Scholar, Thief: Fair Use and Unpublished Works, 9 CARDOZO ARTS & ENT. L.J. 15, 20-25 (1990); William W. Fisher III, Reconstructing the Fair Use Doctrine, 101 HARV. L. REV. 1661, 1729-31 (1988). David Lange illustrates this point by describing how comedians have borrowed from others in building their own careers. David Lange, Recognizing the Public Domain, 44 LAW & CONTEMP. PROBS., Autumn 1981, at 147, 161-62 nn.63-64.

[FN40].

Unlike the patent paradigm, however, the copyright paradigm never prevents third parties from independently creating works of authorship similar to those already on the market. Nor does copyright law invest authors with any generally recognized right to control the end use of protected works as such. On the contrary, by encouraging third parties to make free and abundant use of nonprotectable matter underlying the protected expression, copyright laws foster a built-in process of "reverse engineering" that enables many copyrightable works to cluster around common themes or ideas.

Reichman, supra note 26, at 143-44 (footnotes omitted).

[FN41]. See, e.g., Sega Enters. Ltd. v. Accolade, Inc., 977 F.2d 1510 (9th Cir. 1992); Jerome H. Reichman, Computer Programs as Applied Scientific Know-How: Implications of Copyright Protection for Commercialized University Research, 42 VAND. L. REV. 639, 689-90 (1989) ("Computer scientists always borrow from one another; indeed, they must do so to remain scientifically valid.").

[FN42]. 35 U.S.C. § 154 (1994).

[FN43]. In its most recent fair use case, Campbell v. Acuff-Rose Music, Inc., 114 S. Ct. 1164 (1994), the Court recognized this point:

Accordingly, the mere fact that a use is educational and not for profit does not insulate it from a finding of infringement, any more than the commercial character of a use bars a finding of fairness. If, indeed, commerciality carried presumptive force against a finding of fairness, the presumption would swallow nearly all of the illustrative uses listed in the preamble paragraph of § 107, including news reporting, comment, criticism, teaching, scholarship, and research, since these activities "are generally conducted for profit in this country." Congress could not have intended such a rule, which certainly is not inferable from the common-law cases, arising as they did from the world of letters in which Samuel Johnson could pronounce that "[n]o man but a blockhead ever wrote, except for money."

Id. at 1174 (alteration in original) (citations omitted); see also GOLDSTEIN, supra note 1, § 5.2.1.4(d) (Supp. 1994) (noting that competitors are normally free to study a copyrighted work and extract its unprotectible elements).

[FN44]. This is a desirable kind of system for several reasons. First, authors vary greatly in talent. Encouraging the production of a variety of works of any given kind allows those authors to compete with each other. One hopes that good authors will succeed more often than bad authors. Second, different members of the public respond differently to different authors. What might appeal to one person might not appeal to another. The proliferation of works of a given genre helps provide diversity so that many different tastes are accomodated. See supra notes 24-26 and accompanying text.

[FN45]. See, e.g., Feist Publications, Inc. v. Rural Tel. Serv. Co., 499 U.S. 340 (1991) (competitor allowed to study white pages telephone directory in order to create its own competing directory).

[FN46]. See Sega Brief, supra note 9, at 156 n.29; see also Leval, supra note 6, at 1109-10.

[FN47]. See, e.g., Campbell v. Acuff-Rose Music, Inc., 114 S. Ct. 1164 (1994) (discussing fair use of one work by a parodist who creates a parody).

[FN48]. Judge Leval implicitly emphasizes this point when he contrasts transformative uses with uses that merely republish the original. Leval, supra note 6, at 1111-16. The former may involve competitors who comment on and criticize the first work. This certainly requires access to the expression of the first author. The Supreme Court agrees with some of Judge Leval's ideas concerning transformative use. See Campbell, 114 S. Ct. at 1177.

[FN49]. Private letters and diaries provide excellent examples.

[FN50]. 17 U.S.C. §§ 106-106A (1994). These rights are subject to limitations and exceptions found in §§ 107-120 (1994).

[FN51]. Copyright law does not grant a copyright owner the right to make and sell copies of a work. In the absence of a copyright law, everyone (including the author of the work) has the privilege to make and sell copies of a work. Thus, everyone can try to make money from these activities. The copyright law changes that by limiting the privilege to the copyright owner, allowing the owner the opportunity to have the market to himself, except as provided elsewhere in the Act. Presumably, the copyright owner will make more money in the absence of competition than he would make were everyone competing in selling the same work.

[FN52]. Anyone violating any of the exclusive rights of the copyright owner is an infringer. 17 U.S.C. § 501(a) (1994).

[FN53]. This, as noted before, is the means to the primary goals of promoting the Progress of Science. See supra notes 19-21 and accompanying text.

[FN54]. In its entirety, § 106 reads as follows:

§ 106. Exclusive rights in copyrighted works

Subject to sections 107 through 120, the owner of copyright under this title has the exclusive rights to do and to authorize any of the following:

(1) to reproduce the copyrighted work in copies or phonorecords;

(2) to prepare derivative works based upon the copyrighted work;

(3) to distribute copies or phonorecords of the copyrighted work to the public by sale or other transfer of ownership, or by rental, lease, or lending;

(4) in the case of literary, musical, dramatic, and choreographic works, pantomimes, and motion pictures and other audiovisual works, to perform the copyrighted work publicly; and

(5) in the case of literary, musical, dramatic, and choreographic works, pantomimes, and pictorial, graphic, or sculptural works, including the individual images of a motion picture or other audiovisual work, to display the copyrighted work publicly.

17 U.S.C. § 106 (1994).

[FN55]. Compare the definition of "commercially exploit" in the Semiconductor Chip Protection Act of 1984, 17 U.S.C. § 901(a)(5) (1994):

[T]o "commercially exploit" a mask work is to distribute to the public for commercial purposes a semiconductor chip product embodying the mask work; except that such term includes an offer to sell or transfer a semiconductor chip product only when the offer is in writing and occurs after the mask work is fixed in the semiconductor chip product.

[FN56]. The fact that some authors would produce new works even if there were no copyright system does not detract from this statement. Congress has apparently decided that there is a need for incentives.

[FN57]. See, e.g., BENJAMIN KAPLAN, AN UNHURRIED VIEW OF COPYRIGHT 75 (1967) ("[C]opyright has evidently more to do today with mobilizing the profit- propelled apparatus of dissemination--publication and distribution--than with calling the signals into first published existence; the latter process must be to a considerable extent self-generated."); Leval, supra note 6, at 1117-19 (Copyright's central concern is "for the protection of material conceived with a view to publication, not of private memos and confidential communications that its authors do not intend to share with the public.").

[FN58]. See supra text accompanying notes 50-53. Various parts of the Copyright Act reinforce the view that commercialization marks the boundary between works having formal copyright protection but little substantive rights and works which have "real rights." For example, the core rights under § 106 are the exclusive rights to make copies and to sell or lease copies. This exclusivity normally will not apply until after commercialization, because it is not until the author has distributed copies that another person can gain access and be in a position to infringe by making and selling copies.

In addition, the remedies granted by the Copyright Act are at their peak only after commercialization. The copyright owner will have a comparatively hard time proving lost sales if he is not making any sales. Statutory damages and attorneys fees only apply to a work that is registered and in most instances registration does not occur until after commercialization.

[FN59]. In contrast, § 106A grants "moral rights" for certain kinds of works. 17 U.S.C. § 106A (1994). These moral rights protect the author's honor and reputation.

[FN60]. Sections 107-120 identify activities that are non-infringing notwithstanding the provisions of § 106. 17 U.S.C. § 107-120 (1994). For example, the public display of a lawfully made copy of a work by the copy's owner is not a copyright infringement. 17 U.S.C. § 109(c) (1994).

[FN61]. See supra notes 50-53 and accompanying text.

[FN62]. See supra note 17.

[FN63]. The term "publication" is still important in the 1976 Copyright Act in a variety of contexts. These include 17 U.S.C. § 104 (1994) (national origin); § 302(c) (duration of work made for hire); §§ 401-406 (regarding copyright notice in works); § 407 (deposit for Library of Congress); § 408-412 (copyright registration).

[FN64]. Federal copyright protection under the 1909 Act was closely linked to publication and to the use of a copyright notice. Under that act, the failure to place copyright notice on copies of a published work would forfeit federal copyright protection. For works, such as movies, that were performed publicly without the sale of copies to the public, no notice was required because the work was not published.

[FN65]. Two examples illustrate this point. First, an author who obtains an advance when signing a book publishing contract obtains a benefit before the public benefits from publication. Second, an author who does not receive any royalties until after copies are sold obtains economic benefits after the public benefits from publication.

[FN66]. In one place, the Copyright Act of 1976 recognizes that acts of "commercialization" are as important as acts of "publication." The termination provisions of § 203(a) provide that any exclusive or nonexclusive grant of a transfer or license of copyright can be terminated

during a period of five years beginning at the end of thirty-five years from the date of execution of the grant; or, if the grant covers the right of publication of the work, the period begins at the end of thirty-five years from the date of publication of the work under the grant or at the end of forty years from the date of execution of the grant, whichever term ends earlier.

17 U.S.C. § 203(a)(3) (1994) (emphasis added). Assuming no prior commercialization, the date of execution of the grant is the date when the work is commercialized.

[FN67]. See supra note 42 and accompanying text.

[FN68]. For a variety of works, the Copyright Office only requires the deposit of "identifying material." 37 C.F.R. §§ 202.20(c), 202.21 (1994). See generally Miller, supra note 6, at 990; Leo J. Raskind, The Uncertain Case for Special Legislation Protecting Computer Software, 47 U. PITT. L. REV. 1131, 1140-41 (1986).

[FN69].

That copyright is a conditional right means that it is not so much a reward as it is a quid pro quo. The author receives the reward for making his or her original work of authorship accessible to all. Contrary to the common notion, the reward is not for the act of creation, but for the distribution to provide public access: public learning comes not from the creation of a work, but from reading and studying it, a truism that copyright owners have apparently managed to hide from courts for many years. Thus, to allow the author to retain the right--for life and beyond--to control access to a publicly disseminated work is to grant him or her the power to defeat the purpose of copyright even after having received its reward.

Patterson, supra note 6, at 37 (emphasis added); see also PATTERSON & LINDBERG, supra note 6, at 138 (1991).

[FN70]. Alternative views of copyright's quid pro quo are discussed infra Part II.

[FN71]. The author may receive non-economic benefits by using copyright rights to prevent others from disseminating a work. See supra notes 34-36 and accompanying text. For example, Howard Hughes and J.D. Salinger sought to protect their privacy by using copyright law to prevent others from writing biographical works. Salinger v. Random House, Inc., 811 F.2d 90 (2d Cir. 1987); Rosemont Enters., Inc. v. Random House, Inc., 366 F.2d 303 (2d Cir. 1966), cert. denied, 385 U.S. 1009 (1967). See generally Nevins, supra note 6; Weinreb, supra note 6, at 1145.

[FN72].

Our point here is not that every copyrighted item must demonstrably promote learning, enrich the public domain, and reward the author. But at the least, copyright law should not be used so as actually to inhibit learning, to avoid the public domain, or to deny the author a reward. We should never forget that copyright is not a right so much as it is a monopoly privilege that is granted in return for a quid pro quo--the creation of a work of authorship to benefit the public. Copyright can thus be viewed as being in the nature of a contract--a bargain and sale--between the author (copyright owner) and the public, represented by Congress.

PATTERSON & LINDBERG, supra note 6, at 138; see also Harold C. Moore, Comment, Atari v. Nintendo: Super Mario Uses "Expressive" Security Feature to "Lock" Out the Competition, 18 RUTGERS COMPUTER & TECH. L.J. 919, 939 (1992).

[FN73]. This matrix oversimplifies reality. For instance, accessibility is actually a continuum rather than a dichotomy.

[FN74]. Neither the Copyright Act nor the Constitution require copyright owners to provide access to copyrighted works. See infra notes 120-141 and accompanying text.

[FN75]. See Sega Brief, supra note 9, at 153-54; Karjala, supra note 6, at 994; Moore,supra note 72, at 939; Samuelson, supra note 6, at 705-06.

[FN76]. These points are elaborated in Parts III and IV, respectively.

[FN77]. See generally KAPLAN, supra note 57; PATTERSON & LINDBERG, supra note 6.

[FN78]. 17 U.S.C. § 10 (1909 Copyright Act, repealed by the 1976 Copyright Act). If federal copyright was not obtained at the time of publication, it was forever lost. One could obtain federal copyright protection for certain unpublished works upon registration of such works. 17 U.S.C. § 12 (1909 Copyright Act, repealed by the 1976 Copyright Act).

[FN79]. Baker v. Selden, 101 U.S. 99, 103 (1879) ("The very object of publishing a book on science or the useful arts is to communicate to the world the useful knowledge which it contains."); Samuelson, supra note 6, at 711 ("Authors affixed copyright notices to their works when the works were published, and publication meant that the work was disclosed to the public.") (footnotes omitted).

[FN80]. See supra note 17.

[FN81]. See, e.g., Reichman, supra note 26, at 142 ("[T]he mature copyright paradigm has largely evolved its present physiognomy because it deals with intellectual goods not protectable as trade secrets that require no reverse engineering to appropriate.") (footnote omitted).

[FN82]. State law protection under common law copyright protected an author's right of first publication and scholars recognized that works protected by common law copyright were not accessible. The lack of access did not trouble scholars because common law copyright was perceived as serving two important objectives: (1) protecting the right of privacy for works such as diaries and private letters that an author might want to keep unpublished; and (2) protecting the right of first publication while a work was being written and rewritten prior to publication.

[FN83]. For example, Alan Latman's 1958 study of fair use, as part of the copyright revision process leading up to the 1976 Copyright Act, does not discuss fair use of inaccessible works beyond a passing statement that there is a "general assumption that the fair use doctrine does not apply to [literary property protected by common law]." Alan Latman, Study 14: Fair Use of Copyrighted Works, reprinted in 1 GEORGE S. GROSSMAN, OMNIBUS COPYRIGHT REVISION LEGISLATIVE HISTORY (1976). One suspects that this assumption arose in the context of uncommercialized works such as book manuscripts, letters, and diaries. This Article argues that fair use analysis should distinguish such works from commercialized, inaccessible works such as computer programs.

[FN84]. The doctrine that works had to be humanly accessible without the aid of a machine stems from White-Smith Music Publishing Co. v. Apollo Co., 209 U.S. 1 (1908), a case holding that piano music rolls were not "copies" of musical compositions. Id. at 18.

[FN85]. King v. Mister Maestro, Inc., 224 F. Supp. 101 (S.D.N.Y. 1963); cf. Fortnightly Corp. v. United Artists Television, Inc., 392 U.S. 390 (1968) (holding that community antenna television system that retransmitted broadcast television signals did not "perform" the programs within the meaning of the Copyright Act).

[FN86]. These works would be protected by state common-law copyright, which meant that protection would vary from state to state and that separate infringement claims would have to be brought for each state in which infringement occurred.

[FN87]. William S. Strauss, Study 29: Protection of Unpublished Works 12-15, reprinted in 2 GROSSMAN, supra note 83.

[FN88]. Karjala, supra note 6, at 1009.

[FN89]. See supra note 63.

[FN90]. See supra note 17.

[FN91]. See supra notes 50-66 and accompanying text.

[FN92]. 17 U.S.C. § 102(a) (1994).

[FN93]. Common law copyright is preempted with regard to works that are fixed in a tangible medium of expression. 17 U.S.C. § 301(a) (1994). However, it may still exist for works that are not fixed.

[FN94]. David Lange, Copyright and the Constitution in the Age of Intellectual Property, 1 J. INTELL. PROP. L. 119, 126 (1993); Swanson, supra note 6, at 231; see also Sony Corp. of Am. v. Universal City Studios, Inc., 464 U.S. 417, 429 (1984) (Copyright "is intended to motivate the creative activity of authors and inventors by the provision of a special reward, and to allow the public access to the products of their genius after the limited period of exclusive control has expired.").

[FN95]. See supra notes 77-83 and accompanying text.

[FN96]. The duration of copyright for many works is the life of the author plus fifty years. 17 U.S.C. § 302(a) (1994). Works made for hire, anonymous works, and pseudonymous works have a copyright duration of seventy-five years from the work's first publication or one-hundred years from its creation, whichever expires first. 17 U.S.C. § 302(c) (1994).

[FN97]. Samuelson, supra note 6, at 712. Of course, many works are in single copies that are destroyed long before the end of the copyright term, preventing any public access at any time. This is true, of course, whether one's view of copyright's quid pro quo is that put forth in this Article or under some other theory of copyright.

[FN98]. See, e.g., WILLIAM PATRY, THE FAIR USE PRIVILEGE IN COPYRIGHT LAW 401 (1985) (arguing that the public's need for access is adequately satisfied by whatever the copyright owner makes available for sale); Clapes, supra note 6, at 942 (arguing that copyright is designed to encourage creativity, not dissemination); Davidson, supra note 6, at 1056-57 (arguing that copyright's goal is simply to promote creativity, not disclosure); Miller, supra note 6, at 1028-29 (arguing that copyright's goals are to promote the "availability" of works, not necessarily access to their underlying expressions or ideas); Swanson, supra note 6, at 225 (arguing that copyright's goal is protecting the creative output of commercial enterprises).

[FN99]. See, e.g., Note, Copyright Protection of Computer Program Object Code, 96 HARV. L. REV. 1723, 1740-42 (1983). The idea that copyright law embodies a general misappropriation doctrine is strongly undermined by Feist Publications, Inc. v. Rural Tel. Serv. Co., 499 U.S. 340 (1991), which held that copyright protection cannot be given to a work simply because of the amount of time and effort involved in creating the work.

[FN100]. See supra note 68.

[FN101]. It is even more likely that there would be fewer works along with less learning and knowledge. The lack of public access means that competitors will not have the benefit of learning from earlier works, and hence not be able to produce competing expressions of the same ideas. See supra notes 39-48 and accompanying text.

[FN102]. Robert Merges and Richard Nelson have attempted to gather empirical data on an analogous problem in patent law: the extent to which broad patent rights can impede further technological progress. Their discussion of progress in industries characterized by "cumulative technologies" (roughly speaking, this involves industries where today's advances build on and interact with existing technologies) suggests that broad patents may slow technological growth. They use this data to critique the "prospect theory" for granting broad patent rights. Robert P. Merges & Richard R. Nelson, On the Complex Economics of Patent Scope, 90 COLUM. L. REV. 839, 880-97 (1990).

[FN103]. Law professors who filed an amicus curiae brief in Sega Enters. Ltd. v. Accolade, Inc., 977 F.2d 1510 (9th Cir. 1992), responded in the following terms to William Patry's assertion that the public's need for access to computer programs is fully satisfied by the marketing of copies of the program:

This argument incorrectly assumes that the only public purpose served by access is consumer use of the work in question. In fact, distribution to the public for consumption purposes has nothing whatever to do with the idea/expression distinction or the goals of section 102(b). Access to both ideas and expression is provided by public distribution of traditional works, and that access both supplies and justifies the author's economic reward. The idea/expression distinction and section 102(b) are aimed at defining what later authors can freely use in creating new works. Congress has affirmatively stated in section 102(b) that later authors should be able to use the ideas and other unprotected elements in creating new works. This is also the historic message of Baker v. Selden.

Sega Brief, supra note 9, at 156 n.29 (citations omitted).

[FN104]. See, e.g., Casey P. August & Derek K.W. Smith, Understanding Some Intricacies of Software: Expression, Interfaces, and Reverse Assembly, COMPUTER LAW., Apr. 1994, at 16.

It should be noted that other authors, noting how difficult it is to reverse engineer a product, dispute the likelihood that a pirate would go to the trouble to do so. Johnson-Laird, supra note 8, at 900-01.

Moreover, most copying of computer programs is done without any attempt to disguise the copying. People simply make exact copies of the programs. Copyright law can deal with whether or not this is infringement in a very traditional and straightforward manner. See generally Karjala, supra note 6, at 993.

[FN105]. New technologies have made it easier for people to make copies of many other kinds of copyrighted works. Photocopy machines, tape recorders, and VCRs all aid copying. While copying of computer programs by a computer may be easier than other modes of duplication, the problems are similar.

Copyright owners have to be vigilant to prevent piracy. They have often formed associations to assist them. These include ASCAP, BMI, and CCC. Computer companies have emulated these other copyright owners by creating the Software Publishers Association and the Business Software Alliance to help prevent unauthorized copying of software. While no one doubts that many illegal copies of software are in use, there are indications that many major companies are policing themselves to prevent employees from making illegal copies.

[FN106]. Federal copyright protection for these works is important even though there is no public access. It replaces common law copyright and provides many of the same benefits formerly provided by that body of law. See supra note 82. The replacement is not exact, however. For example, common law copyright could prevent, forever, others from publishing private letters, diaries, or other works. In contrast, copyright protection under the 1976 Copyright Act is for a fixed term, after which those works will fall into the public domain. 17 U.S.C. §§ 302-305 (1994). After the copyright term has expired, the only barrier to publication will be physical control of the actual copies of the works. See infra notes 168-174 and accompanying text.

[FN107]. The discussion, infra, indicates that these scholars need to address their concern to the legislature rather than to the courts. The present Copyright Act does not correspond to their view of a properly functioning copyright system.

[FN108]. Earlier, other scholars were concerned about accessibility in the context of private letters. See, e.g., sources cited infra note 181. Access to private letters often involves access to the physical documents, a different access problem than that presented by mass-marketed computer programs.

[FN109]. Object code can be thought of as a string of zeros and ones. The pattern of zeros and ones provides instructions to a computer so that the computer will perform the tasks required of it. For an excellent discussion of the process of programming and the creation of source code and object code, see Anthony L. Clapes et al., Silicon Epics and Binary Bards: Determining the Proper Scope of Copyright Protection for Computer Programs, 34 UCLA L. REV. 1493, 1510-36 (1987).

[FN110]. The code is inaccessible for two reasons. First, it is unreadable because it is in magnetic or chip form in a game cartridge or diskette, and people cannot read magnetic or chip material. Second, the human-readable equivalent of the magnetic or chip material is a pattern of zeros and ones. Sega Enters. Ltd. v. Accolade, Inc., 977 F.2d 1510, 1525 (9th Cir. 1992) (citing Johnson-Laird, Technical Demonstration of "Decompilation," reprinted in REVERSE ENGINEERING: LEGAL AND BUSINESS STRATEGIES FOR COMPETITIVE DESIGN IN THE 1990's 102 (Prentice Hall Law & Business ed. 1992)). In Sega, the court recognized that a trained programmer could work directly from the zeros and ones of object code, but noted that "[b]ecause even a trained programmer cannot possibly remember the millions of zeros and ones that make up a program, however, he must make a written or computerized copy of the disassembled code in order to keep track of his work." Id. See generally Sega Brief, supra note 9, at 151-53; Johnson-Laird, supra note 8, at 864-69; Samuelson, supra note 6, at 683-84.

Commentators who have argued pro and con about whether it is possible (actually or theoretically) for a trained programmer to read object code are quibbling over an irrelevant issue. Copyright, like most of the legal system, is an intensely practical system. See, e.g., Robert A. Kreiss, Section 117 of the Copyright Act, 1991 B.Y.U. L. REV. 1497. The essential point is that object code, in the vast quantities found in normal computer programs, is not, as a practical matter, humanly accessible to programmers. The analysis of copyright's quid pro quo does not turn on whether there are or are not trained programmers who can read and understand some amount of object code.

[FN111]. David L. Hayes, Shrinkwrap License Agreements: New Light on a Vexing Problem, 15 HASTINGS COMM. & ENT. L.J. 653 (1993); Marshall Leaffer, Engineering Competitive Policy and Copyright Misuse, 19 U. DAYTON L. REV. 1087 (1994); David A. Rice, Public Goods, Private Contract and Public Policy: Federal Preemption of Software License Prohibitions Against Reverse Engineering, 53 U. PITT. L. REV. 543 (1992) [hereinafter Rice, Public Goods]; David A. Rice, Trade Secret Clauses in Shrink-Wrap Licenses, COMPUTER LAW., Feb. 1985, at 17 [hereinafter Rice, Trade Secret Clauses].

[FN112]. Scholars have considered whether these contractual restrictions conflict with copyright law principles. See, e.g., Leaffer, supra note 111 (discussing copyright misuse); Rice, Public Goods, supra note 111 (discussing preemption); John T. Soma et al., Software Interoperability and Reverse, 20 RUTGERS COMPUTER & TECH. L.J. 189, 223-35 (1994).

[FN113]. See, e.g., Karjala, supra note 6; Leaffer, supra note 111; Rice, Public Goods, supra note 111; David A. Rice, Sega and Beyond: A Beacon for Fair Use Analysis At Least as Far as It Goes, 19 U. DAYTON L. REV. 1131 (1994); Samuelson, supra note 6.

[FN114]. This is not meant to denigrate the obvious benefits from computer programs or to suggest that there is no access at all. Users are able to "use" the programs, whether to play a computer game, to use a word processing program, or to have the program operate a kitchen appliance. The user gains access to as much of the user interface of the program as is appropriate for the "use" of the program. That access might involve access to an appreciable amount of expression in the case of a computer game, or to hardly any expression in the case of a kitchen appliance. However, the user has no access to the protected expression contained in the source code of the program, and virtually no access to the program's logic, ideas, processes, procedures, or other unprotected elements.

The fact that the computer software industry has evolved into a multi-billion dollar industry indicates that users are deriving significant benefits from the functionality of the programs and from the degree of access that they are afforded. The extent to which these benefits are the kind that copyright seeks to promote (i.e., the "Progress of Science") as opposed to whether these benefits are from the functionality of many programs is a nice empirical question. Certainly for computer games, a major portion of the benefits to the user seem to be derived from the expression of the user interface. On the other hand, for highly functional programs such as those in kitchen appliances or operating system software, there seems to be very little access, and consumers seem to acquire these programs primarily for their functional aspects. For works, such as word processing programs, where both the user interface and the performance of the program is important to user satisfaction, the question is likely to have a mixed answer.

Accessibility to other authors, another important aspect of accessibility, see supra notes 39-48 and accompanying text, is present to the same limited extent as it is to users. To the extent that programmers do not have access to all of the aspects of a commercialized program, they cannot learn from that work in order to create their own expressions of the ideas, processes, procedures, and methods embodied in that work.

For competing programmers, the barrier to access to computer programs imposed by distribution in object code serves to thwart copyright's traditional objectives. In more traditional contexts, a competing author who wishes to learn the writing techniques in a published novel or a poem can purchase the work and study those techniques. Distribution of the work provides (or is assumed to provide) sufficient accessibility so as to enable competing authors this ability.

[FN115]. See supra note 21 and accompanying text.

[FN116]. Particularly in the Atari and Sega cases cited supra note 7.

[FN117]. As suggested by one court, patents and copyrights "should be limited to those things which serve in the promotion of science and the useful arts." In re Shao Wen Yuan, 188 F.2d 377, 380 (C.C.P.A. 1951). Scholars have argued that copyright should be denied to an author who suppresses a work. See, e.g., P. McCoy Smith, Copyright, Suppression and the Problem of the Unpublished Work: Lessons from the Patent Law, 19 AIPLA Q.J. 309 (1991).

[FN118]. Samuelson, supra note 6.

[FN119]. Some authors argue in favor of a disclosure requirement, based upon the common constitutional foundation of patents and copyrights and upon patent's disclosure requirement. Smith, supra note 117; Soltysinski, supra note 6. The authors do not consider such copyrighted works as diaries, private letters, secure tests, and other material that should not be published. If copyright did not give the author the right to prevent publication, some other body of law would have to be created to do so. See supra notes 82, 106.

[FN120]. 17 U.S.C. § 102(a) (1994).

[FN121]. See supra note 16.

[FN122]. Ascribing "intent" to a body of people has theoretical difficulties under the best of circumstances. In the case of recent copyright legislation, the difficulties are compounded because Congress has tended to enact legislation that is written by interest groups and represents compromises that the various interest groups have decided that they can live with. See, e.g., Jessica D. Litman, Copyright, Compromise, and Legislative History, 72 CORNELL L. REV. 857 (1987); Thomas P. Olson, The Iron Law of Consensus: Congressional Responses to Proposed Copyright Reforms Since the 1909 Act, 36 J. COPYRIGHT SOC'Y U.S.A. 109 (1989).

[FN123]. Strauss, supra note 87, at 32-33. Also, such works could have received federal copyright protection under the 1909 Copyright Act if they were registered. See supra note 78.

[FN124]. H.R. REP. NO. 1476, 94th Cong., 2d Sess. 52 (1976).

[FN125]. See supra note 21 and accompanying text.

[FN126]. Patterson, supra note 6, at 4, 18; cf. Samuelson, supra note 6, at 705-12.

[FN127]. Patterson, supra note 6.

[FN128]. Arthur Miller has argued that courts should administer copyright law in the same fashion for computer programs as they do for other kinds of copyrighted work. Miller, supra note 6, at 985-91.

[FN129]. Feist Publications, Inc. v. Rural Tel. Serv. Co., Inc., 499 U.S. 340 (1991).

[FN130]. As noted supra note 21, at the time the Constitution was written, "Science" referred to learning and knowledge, and was paired with "Authors" and "Writings" in the Constitution. This portion of the Enabling Clause dealt with copyrights rather than with patents.

[FN131]. While the recipient of a letter might benefit, the knowledge and information contained in the letter is not available to the public generally. Similarly, the recipient of business and technical documents containing trade secret information might benefit, but there would be no public benefit.

[FN132]. The word "Writings" is used here in its constitutional sense, a sense which includes not only books, but also charts, maps, paintings, sculpture, music, choreographic works, architectural works, etc. Were it not for Feist Publications, Congress arguably could be even more overinclusive and protect some works which might not be "Writings," provided that such overinclusiveness promoted the objectives of the constitutional clause.

[FN133]. Hutchinson Tel. Co. v. Fronteer Directory Co., 770 F.2d 128, 130 (8th Cir. 1985); Schnapper v. Foley, 667 F.2d 102, 112 (D.C. Cir. 1981); Mitchell Bros. Film Group v. Cinema Adult Theater, 604 F.2d 852, 859- 60 (5th Cir. 1979), cert. denied, 445 U.S. 917 (1980); see also Ralph S. Brown, Eligibility for Copyright Protection: A Search for Principled Standards, 70 MINN. L. REV. 579, 592 (1985).

[FN134]. Schnapper, 667 F.2d at 112; Mitchell Bros., 604 F.2d at 859- 60. But see Martinetti v. Maguire, 16 F. Cas. 920, 922-23 (C.C.D. Cal. 1867).

[FN135]. Bleistein v. Donaldson Lithographing Co., 188 U.S. 239, 251-52 (1903).

[FN136]. Hutchinson, 770 F.2d at 130; Pacific & Southern Co., Inc. v. Duncan, 744 F.2d 1490, 1498-99 (11th Cir. 1984), cert. denied, 471 U.S. 1004 (1985); Jartech, Inc. v. Clancy, 666 F.2d 403, 405-06 (9th Cir. 1982), cert. denied, 459 U.S. 826 (1982); Schnapper, 667 F.2d 102; Mitchell Bros., 604 F.2d 852.

[FN137]. See, e.g., WILLIAM F. PATRY, LATMAN'S THE COPYRIGHT LAW 16-17 (6th ed. 1986); Thomas W. McEnerney, Recent Development, Fraudulent Material Is Entitled to Copyright Protection in Action for Injunctive Relief and Damages, 74 COLUM. L. REV. 1351 (1974).

[FN138]. Mitchell Bros., 604 F.2d at 860.

[FN139]. In Belcher v. Tarbox, 486 F.2d 1087, 1088 (9th Cir. 1973), the court rejected a claim that allegedly fraudulent material should be denied copyright protection:

There is nothing in the Copyright Act to suggest that the courts are to pass upon the truth or falsity, the soundness or unsoundness, of the views embodied in a copyrighted work. The gravity and immensity of the problems, theological, philosophical, economic and scientific, that would confront a court if this view were adopted are staggering to contemplate. It is surely not a task lightly to be assumed, and we decline the invitation to assume it.

[FN140]. Patterson, supra note 6, at 31.

[FN141]. See supra notes 77-79 and accompanying text.

[FN142]. Showing how accessibility, commercialization, and copyright's quid pro quo affect fair use analysis should not be confused with a comprehensive theory of fair use. The former should be part of the latter, but it is not the entire comprehensive theory.

[FN143]. The full text of the fair use provision reads as follows:

Notwithstanding the provisions of section 106 and 106A, the fair use of a copyrighted work, including such use by reproduction in copies or phonorecords or by any other means specified by that section, for purposes such as criticism, comment, news reporting, teaching (including multiple copies for classroom use), scholarship, or research, is not an infringement of copyright. In determining whether the use made of a work in any particular case is a fair use the factors to be considered shall include--

(1) the purpose and character of the use, including whether such use is of a commercial nature or is for nonprofit educational purposes;

(2) the nature of the copyrighted work;

(3) the amount and substantiality of the portion used in relation to the copyrighted work as a whole; and

(4) the effect of the use upon the potential market for or value of the copyrighted work.

The fact that a work is unpublished shall not itself bar a finding of fair use if such finding is made upon consideration of all the above factors.

17 U.S.C. § 107 (1994).

[FN144]. Campbell v. Acuff-Rose Music, Inc., 114 S. Ct. 1164 (1994).

[FN145]. See, e.g., 2 GOLDSTEIN, supra note 1, § 10.1 (identifying (1) "a private benefit approach, excus[ing] uses that the copyright owner would have licensed but for insurmountable transaction costs" and (2) "a public benefit approach, excus[ing] a use, even in the absence of transaction costs, if the social benefit of the use outweighs the loss to the copyright owner"); see also Jay Dratler, Jr., Distilling the Witches' Brew of Fair Use in Copyright Law, 43 U. MIAMI L. REV. 233 (1988); Fisher, supra note 39; Wendy J. Gordon, Fair Use as Market Failure: A Structural and Economic Analysis of the Betamax Case and Its Predecessors, 82 COLUM. L. REV. 1600 (1982); William M. Landes, Copyright Protection of Letters, Diaries, and Other Unpublished Works: An Economic Approach, 21 J. LEGAL STUDIES 79 (1992); Leval, supra note 6; Weinreb, supra note 6.

[FN146]. Historically, fair use was developed in the context of allowing competitors to make use of a copyrighted work. In the nineteenth century, the doctrine of fair use excused a competitor who made an abridgement of a work or who made a translation of the work. See, e.g., 2 GOLDSTEIN, supra note 1, § 10.1 n.5; PATTERSON & LINDBERG, supra note 6, at 67-68.

It is unfortunate that, for a few years, the Supreme Court used language in its opinions suggesting that use by a junior party for commercial gain will almost always preclude a finding of fair use. See, e.g., Sony Corp. of Am. v. Universal City Studios, Inc., 464 U.S. 417, 451 (1984) ("every commercial use of copyrighted material is presumptively unfair"); Harper & Row Publishers, Inc. v. Nation Enters., 471 U.S. 539, 555 (1985) ("Under ordinary circumstances, the author's right to control the first public appearance of his undisseminated expression will outweigh a claim of fair use.").

Fortunately, in its most recent pronouncement in Campbell, the Court emphatically stated that commercial uses should not be condemned automatically:

If, indeed, commerciality carried presumptive force against a finding of fairness, the presumption would swallow nearly all of the illustrative uses listed in the preamble paragraph of § 107, including news reporting, comment, criticism, teaching, scholarship, and research, since these activities "are generally conducted for profit in this country."

Campbell, 114 S. Ct. at 1174 (quoting Harper & Row, 471 U.S. at 512 (Brennan, J., dissenting)).

The point is that competitors should be among the first to be given the benefit of fair use because it is competitors who will be producing the competing works which copyright law seeks to encourage. See supra notes 39-48 and accompanying text.

This is not to say that the competitor can then duplicate on a wholesale basis the expression of the first work in the competing work. On the contrary, while a competitor can learn from the ideas and expression of the first work, fair use does not protect the competitor from all claims of infringement as to the competing work which is produced. We expect the competitor to produce his or her own expression of the ideas in that work.

[FN147]. The same point can be made about other competitive, productive uses of a copyrighted work. For example, the fair use provision of the Copyright Act encourages scholarship, teaching, and learning by indicating that such uses are among the kinds of uses which may be fair uses.

[FN148].

Monopoly protection of intellectual property that impeded referential analysis and the development of new ideas out of old would strangle the creative process. Three judicially created copyright doctrines have addressed this problem: first, the rule that copyright does not protect ideas, but only the manner of expression; second, the rule that facts are not within the copyright protection, notwithstanding the labor expended by the original author in uncovering them; and finally, the fair use doctrine, which protects secondary creativity as a legitimate concern of the copyright.

Leval, supra note 6, at 1109-10.

[FN149]. See generally id. at 1110-11 (The four statutory fair use factors direct courts "to ask in each case whether, and how powerfully, a finding of fair use would serve or disserve the objectives of the copyright."). But see Weinreb, supra note 6, at 1140 (arguing that fair use is not solely designed to promote copyright's utilitarian goals).

[FN150]. Cf. Campbell, 114 S. Ct. at 1174 (no single factor should be treated as dispositive in the fair use analysis).

[FN151]. See supra notes 50-66 and accompanying text.

[FN152]. The "nature of the copyrighted work."

[FN153].

[A]n author's right to choose when he will publish is no less deserving of protection. The period encompassing the work's initiation, its preparation, and its grooming for public dissemination is a crucial one for any literary endeavor. The Copyright Act, which accords the copyright owner the "right to control the first public distribution" of his work, echoes the common law's concern that the author or copyright owner retain control throughout this critical stage. Under ordinary circumstances, the author's right to control the first public appearance of his undisseminated expression will outweigh a claim of fair use.

Harper & Row Publishers, Inc. v. Nation Enters., 471 U.S. 539, 554-55 (1985) (citation omitted).

[FN154]. 2 GOLDSTEIN, supra note 1, § 10.2.2.2.

[FN155]. See generally supra notes 62-66 and accompanying text. Virtually all courts writing about fair use seem to believe that whether a work is "published" or "unpublished" is important to the second fair use factor. See, e.g., Harper & Row, 471 U.S. at 554. The statutory language in § 107 does not compel this analytical distinction and, as this Article argues, the policies underlying the Copyright Act indicate that what is important is whether a work has been "commercialized" or not.

Section 107 contains only a single reference to "publication." This is contained in one sentence added to § 107 in 1992: "The fact that a work is unpublished shall not itself bar a finding of fair use if such finding is made upon consideration of all the above factors." 17 U.S.C. § 107 (1994). This language was added in response to two cases involving use of private letters in libraries. In that context, the term "unpublished" has the same meaning as "uncommercialized," and nothing suggests that Congress was even thinking about whether the second fair use factor should focus on the broader concept of commercialization as opposed to the concept of publication. See generally Jonathan Band & Laura L.F.H. McDonald, The Fair Use Bill: A Funny Thing Happened on the Way to Congress, 10 COMPUTER LAW., Mar. 1993, at 9.

For the most part, fair use discussions about whether or not a work has been "published" could as easily have been written using the term "commercialized." For example, in Sega Enters. Ltd. v. Accolade, Inc., 977 F.2d 1510, 1526 n.8 (9th Cir. 1992), the court labeled as "published" the computer programs that were copied. The court's rationale was that the "computer game cartridges [[were] held out to the public for sale." The rationale correctly recognized that the operative facts involve commercialization.

It is in the context of computer programs and a few other kinds of works or uses where the distinction between "published" and "commercialized" actually matters. For example, the source code of a computer program is rarely published, even though the object code version of the work may be widely sold in the marketplace. As the Sega court correctly concluded, what was important was the commercialization. The court's characterization of the works as "published" was actually inaccurate to the extent that it implied that the source code was available. See also Litman, supra note 35, at 199-201.

[FN156]. In many instances, commercialization will occur prior to publication because commercialization occurs when the author signs a publishing contract authorizing the copying and distribution of the book. During the interval between commercialization and publication, the author still has an interest in being allowed to complete the work. Harper & Row, 471 U.S. at 554-55. The different dates of commercialization and publication also affect when a grant can be terminated under § 203(a). See supra note 66.

[FN157]. This Article's use of "commercialization" rather than "publication" finds some support in the legislative history of fair use. Some commentators suggested that fair use analysis should apply to these "technically 'unpublished' but publicly performed works." See Patry, supra note 98, at 441- 42.

[FN158]. Cf. Gordon, supra note 145 (discussing fair use where transactions costs or other impediments interfere with free market transactions between copyright owners and users).

[FN159]. These works are in the upper left quadrant of the Quid Pro Quo Matrix, supra text accompanying note 73.

[FN160]. See supra notes 38-49 and accompanying text.

[FN161]. Some scholars have begun thinking about the time when much information will be available electronically. Such distribution will have much the same accessibility to the using public as is now the case for mass-marketed books. However, on-line accessibility is likely to be through licenses which may bar various uses. Competitors will probably not be able, for example, to down-load and resell material. To the extent that this material is the expression of authors, copyright law would also prevent the competitor from reselling the material. To the extent, however, that the material is factual, the licensing restrictions would go beyond that offered by copyright law. Whether the use of such license restrictions would or should be preempted by § 301 of the Copyright Act or by the Supremacy Clause of the Constitution presents a thought-provoking issue. See generally Open Discussion Following the Presentation of Jessica Litman, 17 U. DAYTON L. REV. 623 (1991); Open Discussion Following the Presentation of Paul T. Sheils, 17 U. DAYTON L. REV. 597 (1991); Presentation by Professor Jessica Litman, 17 U. DAYTON L. REV. 617 (1991); cf. Leaffer, supra note 111 (discussing whether the doctrine of copyright misuse should be applied to contractual restrictions against disassembly of computer programs); Rice, Public Goods, supra note 111 (discussing preemption of contractual restrictions against disassembly of computer programs).

[FN162]. This factor is the "nature of the copyrighted work." See supra note 143.

[FN163]. As noted supra notes 151-157 and accompanying text, the proper issue is that of commercialization, not publication.

[FN164]. This factor is the "effect of the use upon the potential market for or value of the copyrighted work." See supra note 143.

[FN165]. See, e.g., Campbell v. Acuff-Rose Music, Inc., 114 S. Ct. 1164, 1178 (1994) (distinguishing between second works that displace the first work in the market and those that harm the market for the first work through disparagement); Sega Enters. Ltd. v. Accolade, Inc., 977 F.2d 1510, 1523 (9th Cir. 1992) (distinguishing between original second works that compete with the first work and copies that replace the first work).

[FN166].

[A] key, though not necessarily determinative, factor in fair use is whether or not the work is available to the potential user. If the work is ' out of print' and unavailable for purchase through normal channels, the user may have more justification for reproducing it than in the ordinary case, but the existence of organizations licensed to provide photocopies of out-of- print works at reasonable cost is a factor to be considered.

2 GOLDSTEIN, supra note 1, § 10.2.2.2(b) n.100 (1989) (quoting S. REP. NO. 473, 94th Cong., 1st Sess. 64 (1965)); PATRY, supra note 98.

[FN167]. See infra notes 261-268 and accompanying text.

[FN168]. These works are in the lower right quadrant of the Quid Pro Quo Matrix, supra text accompanying note 73.

[FN169]. Private letters that have been donated to a library or similar institution that allows some access are considered infra notes 175-186 and accompanying text.

[FN170]. One can rightly question whether copyright has anything to do with encouraging the production of private diaries and letters. Landes, supra note 145, at 95. Some scholars have argued that if the copyright incentive is not required, then copyright rights should not be given to such works. John S. Wiley, Jr., Copyright at the School of Patent, 58 U. CHI. L. REV. 119, 152- 54 (1991).

[FN171]. Nevins, supra note 6, at 306-07.

[FN172]. This point throws into relief accessibility's central role in achieving copyright's goals of promoting learning and knowledge. Private letters and diaries do not promote these goals because they are completely inaccessible.

[FN173]. See supra notes 120-141 and accompanying text.

[FN174]. See supra notes 71-72 and accompanying text.

[FN175]. There may be situations in which donors give documents or objects to a library or museum under restrictions that prevent all access for some period of time. So long as those restrictions apply, these situations more properly fall within the previous section dealing with "Private Letters and Diaries." The present subsection deals with materials to which public access does exist. Of course, donors may impose some restrictions even in these cases. For examples of such restrictions on access, particularly in the case of private letters when some of the people involved are still alive, see Salinger v. Random House, Inc., 811 F.2d 90, 93 (2d Cir.), cert. denied, 484 U.S. 890 (1987). Breach of these restrictions presents a distinct legal issue that is not directly germane to this Article. Cf. Wright v. Warner Books, Inc., 953 F.2d 731 (2d Cir. 1991) (raising but not deciding issue of whether copyright owner is third party beneficiary of agreement with library).

[FN176]. The sale or transfer of the work to the institution is not a "commercialization" because the definition of that term provided that "the sale of the original copy of a work is not, in and of itself, a commercialization." See supra note 55. The public display of the work by the institution would not be a "commercialization" because commercialization is limited to acts that "are reserved exclusively to the copyright owner by the Copyright Act under 17 U.S.C. § 106 and which would be copyright infringement if done by another without authority of the copyright owner." Id. While "public display" is within the exclusive rights listed in § 106, § 109(c) states that the public display of a lawfully made copy by the owner of that copy is not an infringement. See supra note 60 and accompanying text.

[FN177]. See supra note 61 and accompanying text.

[FN178]. The emphasis should remain on "commercialization," not on "publication," although these works are not published either. See supra notes 151-156 and accompanying text. See generally supra notes 50-66 and accompanying text.

[FN179]. Actually, in terms of copyright's quid pro quo, the balance favors the public, because the public does have some limited access while the copyright owner is receiving no commercial benefit. Copyright theory is not concerned about this kind of imbalance because the "harm" to the copyright owner (the lack of commercialization) is under the owner's control. This is analogous to the situation in which an owner abandons a copyright. See sources cited infra note 258.

A number of authors, apparently seeing copyright as a grant of rights in exchange for public access, feel that there is an imbalance when the public does not gain access to uncommercialized letters. See, e.g., Conley, supra note 39; Nevins, supra note 6, at 308-09. These authors fail to appreciate that in copyright theory, public accessibility is linked to a copyright owner's commercialization of the work. See supra notes 69-76 and accompanying text. The Second Circuit accurately stated this when it said that

Salinger has a right to protect the expressive content of his unpublished writings for the term of his copyright, and that right prevails over a claim of fair use under "ordinary circumstances." Public awareness of the expressive content of the letters will have to await either Salinger's decision to publish or the expiration of this copyright.

Salinger, 811 F.2d at 100 (citation omitted).

[FN180]. New Era Publications Int'l v. Henry Holt & Co., 873 F.2d 576 (2d Cir. 1989), cert. denied, 493 U.S. 1094 (1990); Salinger, 811 F.2d 90. A later decision, Wright v. Warner Books, Inc., 953 F.2d 731, 740 (2d Cir. 1991), appeared to retreat from the language in New Era and Salinger, and stated that the unpublished nature of a work was not a per se barrier to a finding of fair use.

[FN181]. See, e.g., Mary S. Bilder, The Shrinking Back: The Law of Biography, 43 STAN. L. REV. 299 (1991); Conley, supra note 39; Karen B. LeFevre, The Tell-Tale "Heart": Determining "Fair" Use of Unpublished Texts, 55 LAW & COMTEMP. PROBS., Spring 1992, at 153; Lynn I. Miller, Fair Use, Biographers and Unpublished Works: Life After H.R. 4412, 40 J. COPYRIGHT SOC'Y U.S.A. 349 (1993); Nevins, supra note 6, at 300-09; Mary-Alice Pomputius, "Fair and Foul are Near of Kin": A Suggested Approach to the Fair Use of Unpublished Works, 15 COLUM.-VLA J.L. & ARTS 161 (1991); Smith, supra note 117; Andrea D. Williams, The Fair Use Doctrine and Unpublished Works, 34 HOW. L.J. 115 (1991).

[FN182]. In its entirety, the act reads as follows:

Be it enacted by the Senate and House of Representatives of the United States of America in Congress assembled, That section 107 of title 17, United States Code, is amended by adding at the end the following: "The fact that a work is unpublished shall not itself bar a finding of fair use if such finding is made upon consideration of all the above factors."

Pub. L. No. 102-492, 106 Stat. 3145 (1992).

[FN183]. Wright, 953 F.2d 731; see, e.g., Landes, supra note 6, at 107- 12; Leval, supra note 6, at 1116-22; Miller, supra note 181; Weinreb, supra note 6, at 1144-46. These sources assert that the "unpublished" nature of a work is a factor weighing against a finding of fair use. They should have referred to the works as "uncommercialized." See supra note 155.

[FN184]. Leval, supra note 6, at 1116-22.

[FN185]. While Judge Leval uses the terminology of "published/unpublished," the proper focus should be on whether or not the work has been "commercialized." This difference does not matter for the kinds of works that Judge Leval is discussing.

[FN186]. See supra notes 69-76 and accompanying text.

[FN187]. We assume that the number of copies is at least two (including the original). The analysis for one-of-a-kind works should apply if the copyright owner makes no copies other than the original. See supra notes 61, 175-186 and accompanying text.

[FN188]. While this category appears to be quite different from that of mass-marketed books, the differences may be somewhat illusory. In practice, the wildlife artist who limits the number of copies of a painting to 750, the book publisher who prints 5000 copies of a new book, and the book publisher who prints and reprints a best-seller all produce a finite number of copies and all limit the size of the printing based on their estimates of the number of copies that are likely to sell in a certain amount of time. For example, popular wildlife artists probably print and sell more copies of each work than do lesser-known artists, and books from new authors have smaller press runs than do those with more established names. The major difference is that the publisher of a mass-marketed book is willing to produce additional copies based upon market demand while the publisher of the limited-edition work stops at a pre-defined number of copies no matter what the demand proves to be.

[FN189]. See supra text accompanying note 55.

[FN190]. Under the second fair use factor, the "commercialized" nature of a work is more favorable to a copyright infringement defendant than if the work was "uncommercialized."

[FN191]. This is true for limited-edition works as well as for mass-market works. Editions of Robert Bateman's wildlife prints are typically much larger than editions of works by lesser-known painters.

[FN192]. Section 109(a) of the Copyright Act prevents the copyright owner from barring resales of a lawfully made copy of the copyrighted work. 17 U.S.C. § 109(a) (1994).

[FN193]. The second-hand market may not be adequate for all works. As anyone who has attempted to locate a rare book can attest, some works may be so rare that obtaining a copy is extremely difficult.

For other works, the second-hand market might provide a copy now and then, but a particular use might make demands that cannot readily be supplied by that market. For example, the second-hand book or print market might not be adequate to meet the timing and volume requirements where a professor assigns a particular work in a class. This situation is considered infra notes 261-268 and accompanying text.

[FN194]. See Paul Goldstein, Copyright and the First Amendment, 70 COLUM. L. REV. 983, 989 (1970).

[FN195]. The Uniform Trade Secrets Act defines a "trade secret" as

information, including a formula, pattern, compilation, program, device, method, technique, or process, that:

(i) derives independent economic value, actual or potential, from not being generally known to, and not being readily ascertainable by proper means by, other persons who can obtain economic value from its disclosure or use, and

(ii) is the subject of efforts that are reasonable under the circumstances to maintain its secrecy.

Uniform Trade Secrets Act § 1(4).

[FN196]. Secure tests are also protected by copyright law. For issues involving New York's attempt to gain disclosure of test questions, see Association of Am. Medical Colleges v. Cuomo, 928 F.2d 519 (2d Cir.), cert. denied, 502 U.S. 862 (1991), College Entrance Examination Bd. v. Cuomo, 788 F. Supp. 134 (N.D.N.Y. 1992), and College Entrance Examination Bd. v. Pataki, 889 F. Supp. 554 (N.D.N.Y. 1995), modified, 1995 U.S. Dist. LEXIS 10760 (N.D.N.Y. July 26, 1995).

[FN197]. In addition, breach of confidence and breach of contract claims are not preempted by § 301 of the Copyright Act because the causes of action are not equivalent. See, e.g., Gates Rubber Co. v. Bando Chem. Indus., 9 F.3d 823 (10th Cir. 1993); Trandes Corp. v. Guy F. Atkinson Co., 996 F.2d 655 (4th Cir.), cert. denied, 114 S. Ct. 443 (1993); National Car Rental Sys., Inc. v. Computer Assocs. Int'l, Inc., 991 F.2d 426 (8th Cir.), cert. denied, 114 S. Ct. 176 (1993); Computer Assocs. Int'l, Inc. v. Altai, Inc., 982 F.2d 693 (2d Cir. 1992).

[FN198]. See David Bender, Protection of Computer Programs: The Copyright/Trade Secret Interface, 47 U. PITT. L. REV. 907, 956-58 (1986). Some authors have suggested that copyright and trade secret law should not protect the same work. They note that under the 1909 Copyright Act, joint protection was nearly impossible because federal copyright focused on published works, and that the primary focus of the 1976 Copyright Act is also on published works even though it protects works prior to publication. See, e.g., Pamela Samuelson, Creating a New Kind of Intellectual Property: Applying the Lessons of the Chip Law to Computer Programs, 70 MINN. L. REV. 471, 517-19 (1985).

The structure of the 1976 Copyright Act makes this suggestion untenable as a matter of existing law. There are many ideas protected by trade secret law that are also fixed in a tangible medium of expression and protected by copyright law. Both bodies of law apply and copyright law does not preempt trade secret law. See supra note 197. However, the suggestion does point toward an element of truth, which is that once a work has been "commercialized," copyright law should not stand in the way of appropriate public access to the work.

[FN199]. See generally Emily Campbell, Comment, "Testing" the Copyright Clause: Copyright Protection for Educational and Psychological Tests, 69 NEB. L. REV. 791 (1990); Jocelyn Samuels, Note, Testing Truth-in-Testing Laws: Copyright and Constitutional Claims, 81 COLUM. L. REV. 179 (1981).

[FN200]. See supra text accompanying notes 168-174.

[FN201]. See supra note 55 and accompanying text.

[FN202]. The causes of action for breach of the non-disclosure agreement and for misappropriation of trade secrets are distinct from an action for copyright infringement.

[FN203]. Courts have failed to appreciate the relationship between commercialization and accessibility in a number of cases involving trade secrets. In its analysis of the second fair use factor in a case involving secure tests, one court noted that "[t]he very purpose of copyrighting the MCAT questions is to prevent their use as teaching aids." Association of Am. Medical Colleges v. Mikaelian, 571 F. Supp. 144, 153 (E.D. Pa. 1983), aff'd, 734 F.2d 3 (3d Cir. 1984). It seems that the court was suggesting that copyright approves of commercialized, inaccessible works. Such a suggestion is improper because copyright's quid pro quo is out of balance.

In a similar case, the Second Circuit commented that the nature of secure tests was such as to favor the copyright owner, apparently agreeing with the district court that the tests were "unpublished" works. Association of Am. Medical Colleges v. Cuomo, 928 F.2d 519, 524 (2d Cir.), cert. denied, 502 U.S. 862 (1991); accord College Entrance Examination Bd. v. Pataki, 889 F. Supp. 554 (N.D.N.Y. 1995), modified 1995 U.S. Dist. LEXIS 10760 (N.D.N.Y. July 26, 1995). The court should have treated the work as a "commercialized" work for purposes of fair use analysis.

The fact that these courts erred in their analyses of the second fair use factor does not necessarily mean that they reached an improper result in these cases. Fair use requires an analysis of all of the fair use factors.

[FN204]. See supra note 110.

[FN205]. See supra text accompanying notes 195-200.

[FN206]. As previously noted, the operative concept for purposes of analyzing the second fair use factor is whether the work has been "commercialized," not whether it has been "published." See supra notes 151-156 and accompanying text. See generally supra notes 50-68 and accompanying text.

[FN207]. See, e.g., Computer Assocs. Int'l, Inc. v. Bryan, 784 F. Supp. 982 (E.D.N.Y. 1992); ISC-Bunker Ramo Corp. v. Altech, Inc., 765 F. Supp. 1310, 1323 (N.D. Ill. 1990). Many computer programs are licensed under valid trade secret or contractual restrictions. Indeed, in the early years of the computer industry, this was the most common method of protecting computer programs. A great deal of software is still licensed using valid trade secret restrictions. As literary works (and possibly as audiovisual works), these programs are also protected by copyright law.

At the present time, a great deal of mass-market software is marketed with what purports to be license agreements inside the packages. These licenses often contain restrictions on reverse engineering, disassembly, or other use. Some courts have held these licenses invalid or unenforceable under certain conditions. See, e.g., Step-Saver Data Sys., Inc. v. Wyse Technology, 939 F.2d 91 (3d Cir. 1991); Vault Corp. v. Quaid Software Ltd., 847 F.2d 255 (5th Cir. 1988). But see Microsoft Corp. v. Harmony Computers & Elecs., Inc., 846 F. Supp. 208 (E.D.N.Y. 1994). See generally sources cited supra note 111.

Some have argued that computer programs are protected by trade secret law simply because they are distributed in object code--a form that is not humanly readable. Allen R. Grogan et al., Decompilation and Disassembly: Undoing Software Protection, 1 COMPUTER LAW., Feb. 1984, at 1. The argument made is that object code distribution (even without a confidential relationship or licensing restriction) is a reasonable method of maintaining the secrecy of the program, because the disassembly or decompilation of the program might take months or even years. Such an argument is contrary to trade secret's normal rules that reverse engineering from a lawfully obtained copy is lawful.

Others have argued that § 301 of the Copyright Act and the Supremacy Clause of the Constitution should preempt many contractual restrictions against disassembly or decompilation. See, e.g., Rice, Public Goods, supra note 111. The author, who tries to keep abreast of major new computer cases, knows of no cases that have adopted this view.

[FN208]. Some authors seem to wish that copyright law could be used to protect aspects of programs that are not publicly known and yet not protected by trade secret law. See, e.g., August & Smith, supra note 104, at 20 ("'Reverse engineering' is a term which has been used recently, and incorrectly, to refer to a process of revealing software concepts which are hidden within commercially successful computer programs.").

The proper response to such a wish is straightforward. If there are "secrets" in the programs, one can try to protect them by trade secret law. That requires reasonable efforts to maintain the secrecy by confidentiality and non- disclosure agreements, etc. If the program cannot be protected under trade secret law (for example, if it is mass-marketed without valid contractual restrictions), then one only obtains as much protection as is available under copyright law (or some other body of law such as patent law). If copyright law (or the other body of law) protects the secrets, fine. But if it doesn't, then it doesn't. The important thing is not to ask copyright law to do something it is not designed to do. As this Article argues, copyright law is only designed to protect secrecy for uncommercialized works.

[FN209]. See supra note 207 and sources cited therein.

[FN210]. These works are in the upper right quadrant of the Quid Pro Quo Matrix, supra text accompanying note 73.

[FN211]. See supra notes 109-110 and accompanying text. There is also access to the user screens for programs that have such screens, and to the functionality of the program, some of which may be expressive. However, there is no access to the original source code. Full access requires access to something approximating the source code.

[FN212]. This kind of copyrighted work presents access problems of a different kind than is the case with works discussed in earlier Parts of this Article. It is not the lack of copies that prevents access to this kind of work, it is the form in which the copies are distributed that inhibits access.

[FN213]. Johnson-Laird, supra note 8, at 843; Reichman, supra note 41, at 700-03.

[FN214]. For the definition of "commercialization," see supra text accompanying note 55. Sega Enters. Ltd. v. Accolade, Inc., 977 F.2d 1510, 1526 n.9 (9th Cir. 1992), rejected the argument that computer games held out to the public for sale were unpublished works for purposes of copyright's fair use analysis. Accord Lewis Galoob Toys, Inc. v. Nintendo of Am., Inc., 780 F. Supp. 1283, 1293 (N.D. Cal. 1991), aff'd, 964 F.2d 965, 970-71 (9th Cir. 1992). Had the Ninth Circuit considered the issue in terms of "commercialization" instead of "publication," it would have similarly concluded that the games were commercialized.

[FN215]. See sources cited supra note 75.

[FN216]. Atari Games Corp. v. Nintendo of Am., Inc., 975 F.2d 832 (Fed. Cir. 1992); Sega Enters. Ltd. v. Accolade, Inc., 785 F. Supp. 1392 (N.D. Cal. 1992). See generally Sega Brief, supra note 9, at 147; Litman, supra note 35, at 196-201; sources cited infra note 226.

[FN217]. Throughout this Article, the term "disassembly" should be taken to include both disassembly and decompilation.

[FN218]. Because disassembly does not recreate the complete structure of the original source code and because it does not contain the programmer's comments that are in the source code, disassembly does not give the public or competitors complete access to the original work. See generally Johnson-Laird, supra note 8, at 843. To the extent that complete access is not provided to a commercialized, inaccessible work, copyright's imbalance has not been fully rectified.

[FN219]. For purposes of this Article, it does not matter whether one thinks of disassembly as producing a copy or a derivative work.

[FN220]. The "intermediate copy" label gives rise to two versions of mistaken analysis. Both versions try to merge the intermediate copy and the final work for purposes of determining infringement. One version of this is that, at times, courts have seemed to adopt a philosophy of "tracing" the fruits of infringement. This means that if they find that some work is a copyright infringement, then they will enjoin works that are based on that infringing copy. Sega, 785 F. Supp. 1392 (N.D. Cal. 1992); accord Clapes, supra note 6, at 931-34; Davidson, supra note 6, at 1095; Ian N. Feinberg & Karen K. Williams, Injunctive Relief Based on Infringing Intermediate Copying: Reflections on Sega and Atari, COMPUTER LAW., Dec. 1993, at 12; Miller, supra note 6, at 1016 n.183.

This "tracing" is improper. Each work, or version of a work, must be analyzed separately for copyright infringement purposes. This is just as would be the case if an historian took notes about the contents of another historical work before writing her own history. Whether or not the notes are an infringing copy is distinct from whether or not the new history is an infringement. In Sega Enters. Ltd. v. Accolade, Inc., 977 F.2d 1510, 1518-19 (9th Cir. 1992), the Ninth Circuit properly rejected the contention that intermediate copying could not constitute copyright infringement unless the final end product infringed.

The second version of the problem is the reverse of the first one. In this version, courts might be tempted to excuse an infringing intermediate copy if the final work is non-infringing. This is "tracing" in reverse. It is equally improper. PATRY, supra note 98, at 400.

[FN221]. Others have asserted that it is fair use for an author/scholar to make copies of journal articles as part of scholarly research. See, e.g., Soma et al., supra note 112, at 210.

[FN222]. This is akin to letting the copyright owner have his cake and eat it too. Traditional fair use analysis applies to accessible, commercialized works. For inaccessible, commercialized works, the fair use analysis needs to change to reflect the lack of access. See supra note 158 and accompanying text.

[FN223]. These are essentially the facts of Sega, 977 F.2d 1510.

[FN224]. Id.

[FN225]. While the court recognized that the copyrighted computer games were commercialized, inaccessible works (to use the language of this Article), the court's access/commercialization analysis was not always proper. For example, while the court recognized the goal of promoting the creation of new works, it observed that dissemination of object code defeats this because it

precludes public access to the ideas and functional concepts contained in those programs, and thus confers on the copyright owner a de facto monopoly over those ideas and functional concepts. That defeats the fundamental purpose of the Copyright Act--to encourage the production of original works by protecting the expressive elements of those works while leaving the ideas, facts and functional concepts in the public domain for others to build on.

Id. at 1527. This language suggests that the purpose of access is to provide access only to the unprotectable ideas in a program. The court's view of accessibility is too narrow. Competitors need access to both the unprotected ideas and the protectable expression in a work.

[FN226]. A number of authors have reported on Sega and/or Atari Games Corp. v. Nintendo of Am., Inc., 975 F.2d 832 (Fed. Cir. 1992), a similar case in which the Federal Circuit used a similar, but less elaborate, fair use analysis. The University of Dayton School of Law held a Scholarly Symposium on "Copyright Protection and Reverse Engineering of Software" in November 1993. Eight papers from this Symposium were published in 19 U. DAYTON L. REV. (1994). They are: Clapes, supra note 6; David Friedman, Standards as Intellectual Property: An Economic Approach, 19 U. DAYTON L. REV. 1109 (1994); Johnson-Laird, supra note 8; Karjala, supra note 6; Leaffer, supra note 111; Vanessa Marsland, Copyright Protection and Reverse Engineering of Software-- An EC/UK Perspective, 19 U. DAYTON L. REV. 1021 (1994); Rice, supra note 113; Jaap H. Spoor, Copyright Protection and Reverse Engineering of Software: Implementation and Effects of theEC Directive, 19 U. DAYTON L. REV. 1063 (1994).

Other sources discussing Sega and/or Atari include: JONATHAN BAND & MASANOBU KATOH, INTERFACES ON TRIAL (1995); William S. Coats & Heather D. Rafter, The Games People Play: Sega v. Accolade and the Right to Reverse Engineer Software, 15 HASTINGS COMM. & ENT. L.J. 557 (1993); David L. Hayes, The Legality of Disassembly of Computer Programs, 12 COMPUTER/L.J. 1 (1993); Pamela Samuelson, Fair Use for Computer Programs and Other Copyrightable Works in Digital Form: The Implications of Sony, Galoob and Sega, 1 J. INTELL. PROP. L. 49 (1993); Soma et al., supra note 112; Julie Aguilar, Summary, Sega Enterprises Ltd. v. Accolade, Inc.: Setting the Standard on Software Copying in the Computer Industry, 23 GOLDEN GATE U. L. REV. 269 (1993); S. Carran Daughtrey, Note, Reverse Engineering of Software for Interoperability and Analysis, 47 VAND. L. REV. 145 (1994); Victor de Gyarfas, Note, Sega v. Accolade%i: A Step Forward for Reverse Engineering?, 23 SW. U. L. REV. 571 (1994); Lloyd G. Farr, Sega v. Accolade: Another Generation of Computer Program Copyright Cases Has Growing Pains, 27 GA. L. REV. 903 (1993); Joe L. Gage, Jr., Comment, Copyright Law and Reverse Engineering: Have Recent Decisions Taken the Fair Out of Use?, 46 BAYLOR L. REV. 183 (1994); Christopher W. Hager, Apples & Oranges: Reverse Engineering as a Fair Use After Atari v. Nintendo and Sega v. Accolade, 20 RUTGERS COMPUTER & TECH. L.J. 259 (1994); Virginia A. Johnson, Comment, Byte After Byte, the Courts Nibble Away at Copyright Protection of Software, 10 GA. ST. U. L. REV. 543 (1994); David C. MacCulloch, Note, Sega Enterprises Ltd. v. Accolade, Inc.% i: What's So Fair About Reverse Engineering, 14 LOY. L.A. ENT. L.J. 465 (1994); Stephen B. Maebius, The New Use of Fair Use: Accessing Copyrighted Programs Through Reverse Engineering, 75 J. PAT. & TRADEMARK OFF. SOC'Y 431 (1993); Timothy S. Teter, Note, Merger and the Machines: An Analysis of the Pro-Compatibility Trend in Computer Software Copyright Cases, 45 STAN. L. REV. 1061 (1993).

[FN227]. The first fair use factor is "the purpose and character of the use, including whether such use is of a commercial nature or is for nonprofit educational purposes." See supra note 143.

[FN228]. Campbell v. Acuff-Rose Music, Inc., 114 S. Ct. 1164, 1174 (1994) (quoting Harper & Row Publishers, Inc. v. Nation Enters., 471 U.S. 539, 592 (1985)).

[FN229]. See supra notes 19-33 and accompanying text.

[FN230]. See supra notes 145-149 and accompanying text.

[FN231]. See generally William F. Patry & Shira Perlmutter, Fair Use Misconstrued: Profit, Presumptions, and Parody, 11 CARDOZO ARTS & ENT. L.J. 667, 679-87 (1993).

[FN232]. William Patry, in discussing disassembly of computer programs, improperly states:

While a for-profit corporation (the typical reverse engineer) may indeed "study" a competitor's program, the corporation does not perform this "study" or "research" for the greater public welfare but only for the commercially selfish purpose of increasing its profits by marketing a product in direct competition with the original.

PATRY, supra note 98, at 400.

The flaws in his statement have already been exposed. To recapitulate: First, the public welfare is served, not disserved, by competitors producing directly competing works. Second, if the original work was humanly accessible (such as a published history or published source code), Patry presumably would not object if a competitor (a second historian or competing computer company) studied that work to produce the competing work. Copyright's promotion of the public welfare would be disserved by giving more protection to the inaccessible work than to the accessible one. Third, the disassembly in this case is done to learn from the first work and be able to produce non-infringing new works. Such study is the kind of study that furthers copyright's goals of promoting the creation and dissemination of new works.

[FN233]. 17 U.S.C. § 107 (1994).

[FN234]. See supra notes 151-156 and accompanying text. See generally supra notes 50-68 and accompanying text.

[FN235]. Campbell v. Acuff-Rose Music, Inc., 114 S. Ct. 1164, 1177-78 (1994).

[FN236]. See supra note 218.

[FN237]. Dennis Karjala adds another point, noting that the creation of a human-readable copy "does not create a single usable copy that can cut into the copyright owner's sales." Karjala, supra note 6, at 1010. His point is that sales of computer programs typically are made to users who desire to use them as functional tools. Such users don't want to buy a manuscript copy of assembly code. They want to buy an object code version that their computer can run.

[FN238]. Sega Brief, supra note 9, at 160-62; Donald S. Chisum et al., LaST Frontier Conference Report on Copyright Protection of Computer Software, 30 JURIMETRICS J. 15, 25 (1989). See generally Patry & Perlmutter, supra note 231, at 692.

[FN239]. Cf. Sega Brief, supra note 9, at 154 (asserting that "[c]omputer programs in object-code form are alone, among all types of publicly distributed copyright-protected works, in being unreadable by human beings").

[FN240]. Fair use for making Braille or recorded oral copies of works is considered in the House Report on the 1976 Copyright Act. H.R. REP. NO. 1476, 94th Cong., 2d Sess. 47, 73 (1976). See infra note 242 and accompanying text.

[FN241]. In addition, of course, a record's grooves are not humanly readable and the magnetic or electrical impulses on a tape or CD are not humanly readable.

The inaccessibility of works other than computer programs has been recognized by others. See, e.g., Clapes, supra note 6, at 935-44; Litman, supra note 35, at 198; Miller, supra note 6, at 989.

[FN242].

Another special instance illustrating the application of the fair use doctrine pertains to the making of copies or phonorecords of works in the special forms needed for the use of blind persons. These special forms, such as copies in Braille and phonorecords of oral readings (talking books), are not usually made by the publishers for commercial distribution. For the most part, such copies and phonorecords are made by the Library of Congress' Division for the Blind and Physically Handicapped with permission obtained from the copyright owners, and are circulated to blind persons through regional libraries covering the nation. In addition, such copies and phonorecords are made locally by individual volunteers for the use of blind persons in their communities, and the Library of Congress conducts a program for training such volunteers. While the making of multiple copies or phonorecords of a work for general circulation requires the permission of the copyright owner, a problem addressed in section 70 of the bill, the making of a single copy or phonorecord as a free service for a blind person would properly be considered a fair use under section 107.

H.R. REP. NO. 1476, supra note 240, at 73.

Without citing the House Report, one author suggested that a copyright owner would not object if a copy of a work is made line by line in a machine which then converts it into an audible text so that it can be heard by a blind person. Wayne A. Ely, Copyright and Trademark Protection of Computer Software-- Reverse Engineering of Competitor's Computer Game Software, Required to Comprehend Work, and Resulting in Display of "False Trademark," Not Violative of Copyright Act or Lanham Act, 12 TEMPLE ENVTL. L. & TECH. J. 137, 156 (1993). Whether the copyright owner would not object for altruistic reasons, for reasons based on comparing the costs and benefits of litigation, or because they would lose the case on fair use grounds is not stated.

[FN243]. Litman, supra note 35, at 198. These examples can be made to look more like the facts of Sega Enters. Ltd. v. Accolade, Inc., 977 F.2d 1510 (9th Cir. 1992) (discussed supra notes 223-238 and accompanying text), by assuming that the commissioning person, without publishing the translation or transcription, uses the ideas gleaned from it to produce his own work. In such case, the translation or transcription is an "intermediate" work. As noted above, the fact that the work became an "intermediate" work by virtue of the completion of some final work does not affect the analysis. See supra note 220 and accompanying text. But see Clapes, supra note 6, at 931-47 (assuming that creating a score in this situation would not be fair use).

[FN244]. See supra notes 219-220 and accompanying text.

[FN245]. H.R. REP. NO. 1476, supra note 240, at 73; see supra note 242.

[FN246]. H.R. REP. NO. 1476, supra note 240, at 73.

[FN247]. Id.

[FN248]. According to one case, Cable News Network, Inc. licenses a company to make and license certain of CNN's programs. CNN v. Video Monitoring Servs. of Am., Inc., 940 F.2d 1471 (11th Cir.), vacated, 949 F.2d 378 (11th Cir.), appeal dismissed, 959 F.2d 188 (11th Cir. 1992).

[FN249]. News clipping services have sprung up to fill a demand for access to broadcast television news programs after the time of the broadcast. See, e.g., Los Angeles News Serv. v. Tullo, 973 F.2d 791 (9th Cir. 1992); Pacific & S. Co., Inc. v. Duncan, 572 F. Supp. 1186 (N.D. Ga. 1983), aff'd in part, rev'd in part, 744 F.2d 1490 (11th Cir. 1984), cert. denied, 471 U.S. 1004 (1985).

[FN250]. See supra notes 55-62 and accompanying text; See also PATTERSON & LINDBERG, supra note 6, at 100. Broadcast television programs represent an instance in which the concepts of commercialization and publication may differ. If the copyright owner is the broadcasting station, the work in this second situation will be unpublished because no copies of the work are distributed and the statute explicitly states that "[a] public performance of a work does not of itself constitute publication." See supra note 17.

This Article's use of "commercialization" rather than "publication" finds support in William Patry's review of the legislative history of fair use. Some commentators suggested that fair use analysis should apply to these "technically 'unpublished' but publicly performed works." PATRY, supra note 98, at 441-42.

[FN251]. Video recorders have allowed people to make recordings of programs for later viewing. Such "time-shifting," when done by private citizens for personal use, has been held to be fair use. Sony Corp. of Am. v. Universal City Studios, Inc., 464 U.S. 417 (1984).

[FN252]. See generally PATTERSON & LINDBERG, supra note 6, at 99-102; Zimmermann, supra note 6.

[FN253]. Sony, 464 U.S. 417. The fact that a work is semi-accessible does not mean that all uses will be fair uses. Duplicating and selling copies of news broadcasts would be unlikely to be a fair use, while making a copy so that scholars and others could study it might be a fair use. As with all fair use analysis, all of the fair use factors must be evaluated and weighed.

If fair use is not found, courts should, in appropriate situations, limit the remedies for infringement to monetary awards. See infra notes 269-270 and accompanying text.

[FN254]. Neither the Supreme Court in Sony, 464 U.S. 417, nor the lower courts in the cases cited supra note 249 have been sensitive to the roles of accessibility and commercialization. In Sony, the Supreme Court's fair use analysis dealt primarily with the nature of the use being made (private "time- shifting") and the potential market impact. While time-shifting relates to accessibility, the Court did not express itself in terms of the role of accessibility in copyright's scheme, nor in terms of the semi-accessibility of the programs. The Court also failed to discuss the fact that the programs had been commercialized in that public performances were made by broadcasting them. The Court's fair use analysis would have been improved if it had been grounded in the fundamental copyright principles discussed in this Article.

[FN255]. Making and distributing small portions of news broadcasts is more likely to be fair use than making and selling copies of the entire news programs. The cases involving news clipping services have not used the copyright model described in this Article. See, e.g., cases cited supra note 249.

[FN256]. At least one bill was proposed in Congress that would add news clipping services to the list of fair use examples in § 107. Bill on T.V. News Clipping Fair Use Gets No Copyright Office Support, 44 BNA'S PATENT, TRADEMARK & COPYRIGHT J. 156 (1992). The analysis in this paper suggests that such legislation might be unnecessary if courts engaged in a more sensitive fair use analysis.

[FN257]. See supra text accompanying note 73.

[FN258]. Another situation is that in which the author abandons the copyright. Abandonment is a defense to copyright infringement and is based on a finding that the copyright owner intended to surrender the copyright rights. National Comics Publications, Inc. v. Fawcett Publications, 191 F.2d 594 (2d Cir. 1951), modified, 198 F.2d 927 (2d Cir. 1952). See generally Robert A. Kreiss, Abandoning Copyrights to Try to Cut Off Termination Rights, 58 MO. L. REV. 85 (1993). Abandonment can be found from non-publication, Pacific & S. Co., Inc. v. Duncan, 572 F. Supp. 1186 (N.D. Ga. 1983), aff'd in part, rev'd in part, 744 F.2d 1490 (11th Cir. 1984), cert. denied, 471 U.S. 1004 (1985), but non-publication does not necessarily indicate that a copyright author is abandoning her copyright.

[FN259]. Other situations might involve other out-of-print works. See, e.g., Rosemont Enters., Inc. v. Random House, Inc., 366 F.2d 303 (2d Cir. 1966) (preliminary injunction against biography based on magazine articles was vacated where magazine articles about Howard Hughes were out of print), cert. denied, 385 U.S. 1009 (1967).

[FN260]. Injunctions are not mandatory in cases of copyright infringement. The statute says that a court "may grant temporary and final injunctions on such terms as it may deem reasonable to prevent or restrain infringement of a copyright." 17 U.S.C. § 502(a) (1994) (emphasis added). In the vast majority of cases of infringement, an injunction should probably issue, but copyright "was never intended to serve the goals of secrecy and concealment. Thus, the copyright law on its own terms, and not merely in deference to the first amendment, demands caution in awarding oppressive injunctions." Leval, supra note 6, at 1132, 1135. Others have also advocated restraint in issuing injunctions in appropriate copyright cases. See, e.g., KAPLAN, supra note 57, at 73; MELVILLE B. NIMMER & DAVID NIMMER, NIMMER ON COPYRIGHT § 1.10[C][2] (1994); Ralph S. Brown, Civil Remedies for Intellectual Property Invasions: Themes and Variations, 55 LAW & CONTEMP. PROBS., Spring 1992, at 45, 46-49.

[FN261]. See supra note 193 and accompanying text.

[FN262]. Under some circumstances, it may be fair use for a professor to make and distribute copies of a work to students in a course. The House Report on the Copyright Act of 1976 contains guidelines for minimum standards for certain educational fair use that were agreed to by a number of interested groups. H.R. REP. NO. 1476, supra note 240, at 68-71. These guidelines were adopted by the House and Senate conferees as being part of their understanding of fair use. H.R. REP. NO. 1733, 94th Cong., 2d Sess. 69, 70 (1976).

[FN263]. The facts are similar to those with regard to some of the works copied in Basic Books, Inc. v. Kinko's Graphics Corp., 758 F. Supp. 1522 (S.D.N.Y. 1991). The limited public access for an out-of-print work, coupled with the arguably educational use being made of the work makes the fair use case a close one. What seemed to tip the scales in Basic Books against fair use was the fact that the copying and selling was done by a commercial establishment, which was not itself making the educational use. It remains an open question whether it would be fair use if the professor makes the copies for the students and provides them to the students for free or at cost.

[FN264]. See Meeropol v. Nizer, 560 F.2d 1061, 1070 (2d Cir. 1977) (commenting that even though letters have been out of print for 20 years, there may be a future market for them), cert. denied, 434 U.S. 1013 (1978); Basic Books, 758 F. Supp. at 1533 (noting that permission fees may be the only income for authors and copyright owners).

[FN265].

Because the fair use enquiry often requires close questions of judgment as to the extent of permissible borrowing in cases involving parodies (or other critical works), courts may also wish to bear in mind that the goals of the copyright law, "to stimulate the creation and publication of edifying matter," are not always best served by automatically granting injunctive relief when parodists are found to have gone beyond the bounds of fair use.

Campbell v. Acuff-Rose Music, Inc., 114 S. Ct. 1164, 1171 n.10 (1994) (quoting Leval, supra note 6, at 1134); see 17 U.S.C. § 502(a) (court "may grant injunctions on such terms as it may deem reasonable to prevent or restrain infringement") (emphasis added); Abend v. MCA, Inc., 863 F.2d 1465, 1479 (9th Cir. 1988) (finding "special circumstances" that would cause "great injustice" to defendants and "public injury" were injunction to issue), aff'd sub nom., Stewart v. Abend, 495 U.S. 207 (1990); Leval, supra note 6, at 1132 (while in the "vast majority of cases, [an injunctive] remedy is justified because most infringements are simple piracy," such cases are "worlds apart from many of those raising reasonable contentions of fair use" where "there may be a strong public interest in the publication of the secondary work [and] the copyright owner's interest may be adequately protected by an award of damages for whatever infringement is found").

[FN266]. Campbell, 114 S. Ct. at 1171 n.10.

[FN267]. In cautioning against automatically granting an injunction in every case of copyright infringement, Judge Leval properly warns that one should consider whether the denial of injunctive relief might "undermine the incentives of authorship for which copyright law was created." Leval, supra note 6, at 1134. It is hard to imagine that an author's decision to create or not create a work would be affected by the knowledge that injunctions might be withheld if the author allowed the published book to go out of print.

[FN268]. Cf. Basic Books, 758 F. Supp. 1522.

[FN269]. See supra notes 249-256 and accompanying text.

[FN270]. Cf. cases cited supra note 249.

[FN271]. Leval, supra note 6, at 1134.

END OF DOCUMENT