Copyright (c) 1997 Albany Law School; Laura G. Lape

I. Introduction
Let's start with a classic example transposed to a new setting. A contributor posts to a listserv [FN1] what looks like a poem, or maybe lyrics. A second subscriber adds a melody and posts the result. The first or second contributor then markets the song and makes a mint. Who owns the copyright to the song? Resolution of this question depends upon the copyright doctrine of joint works, which governs collaborative creation. [FN2]

Increased opportunities for collaborative efforts have developed during the tenure of the Copyright Act of 1976 in fields such as scientific research, the computer industry, and the entertainment industry. [FN3] In particular, interest in and use of computer-mediated collaboration has grown within the past decade. [FN4] Such computer-mediated collaboration includes forms such as video conferencing, desktop conferencing, and electronic mail applications. The growth of collaborative behaviors, combined with judicial restriction in the scope of the work-made-for-hire doctrine, [FN5] increase the significance of joint work doctrine under the Copyright Act.

This Article addresses the creation of joint works under the current Copyright Act, focusing on the courts' restrictive view of joint works. We turn then to a specific context, the Internet, to examine the impact of the use of express language reflecting intent regarding future use of transmitted material. Part II of this Article traces the history of joint work doctrine prior to the 1976 Copyright Act. Part III examines the courts' treatment of joint works under the 1976 Act. Finally, Part IV analyzes the effect of the presence and absence of express language regarding future use of matter communicated via the Internet.

II. Joint Work Doctrine Prior to the 1976 Act
The current copyright act, the Copyright Act of 1976, makes a break with prior law as to the definition of joint works in several respects. First, the 1976 Act contains the first statutory definition of a joint work, which was not defined in any prior act. [FN6] Secondly, the 1976 Act was intended to abandon one development of prior case law, the much criticized 1955 decision of the Second Circuit commonly known as the 12th Street Rag case. [FN7] Given the relative discontinuity between joint works under the 1976 Act and the prior case law, why should we devote any time to the definition of joint works before 1976? First, familiarity with prior judicial treatment of joint works aids our understanding of congressional intent in adopting the 1976 Act since prior case law was the backdrop against which Congress acted. [FN8] Further, there was not a complete break between the 1976 Act and prior case law. The legislative history of the 1976 Act indicates that the Act's definition of a joint work was intended to be generally consistent with case law prior to the 12th Street Rag case. [FN9] Finally, we may examine case law prior to the 1976 Act to search for any precursors to the limitations recently imposed by courts on the definition of a joint work in the form of requirements of 1) an independently copyrightable contribution and 2) an intent to be joint authors. [FN10]
It has been asserted that the doctrine of joint authorship under United States law owes its origins to Maurel v. Smith, [FN11] a 1915 case decided by Judge Learned Hand. [FN12] Learned Hand himself created that impression in Maurel by noting that he had "been able to find strangely little law regarding the rights of joint authors" and by relying on only one joint works case, an 1871 British case. [FN13] However, nineteenth-century judicial opinions show that courts assumed that joint authorship could be created, although their opinions did not specify how. [FN14] For example, nineteenth-century opinions refer to parties as joint authors or report allegations of joint authorship. [FN15] Thus, nineteenth-century parties were treated as joint authors by both the parties themselves and the courts.

What was missing until the early twentieth century was an attempt by the judiciary to expressly define what constituted a joint work. The 1909 Act cases used a "common design," the intent to contribute to one work, as a basis for finding joint authorship. [FN16] Along with common design, however, early cases discussed inseparability, that is, how well blended the contributions were in the final product, as a basis for finding joint authorship. [FN17] For example, Learned Hand reasoned in Maurel that the scenario and dialogue of an opera were parts of a joint work, in part because a "scenario followed as much as this goes into the bone and flesh of the production." [FN18] Learned Hand meant that the scenario and dialogue could not be teased apart. Distinguish this use of inseparability from use of the terms "inseparable" and "indivisible" to describe the legal, rather than the artistic, result, the legal result being the melding of rights which is joint authorship. [FN19]

A series of 1909 Act cases addressed the question whether co-authors need have worked together and at the same time. [FN20] In the leading case, Edward B. Marks Music Corp. v. Jerry Vogel Music Co., [FN21] Learned Hand clarified that all that is needed is intent for the contributions to be used in a single work: "[I]t makes no difference whether the authors work in concert, or even whether they know each other; it is enough that they mean their contributions to be complementary in the sense that they are to be embodied in a single work to be performed as such." [FN22] The Marks holding was stretched beyond what came to be considered the breaking point in 1955 in Shapiro, Bernstein & Co. v. Jerry Vogel Music Co., known as the 12th Street Rag case. [FN23] In that case, the Second Circuit held that the intent to contribute to a larger work can be conceived long after the first contribution has been produced, and can be conceived by someone other than the author of that contribution. [FN24] Note that the Marks rule is consistent with the definition of a joint work adopted in the 1976 Act, while the 12th Street Rag rule is not. [FN25]

The requirement imposed by courts beginning in the late 1980s, that each joint author make an independently copyrightable contribution, is not found in the 1909 Act cases. [FN26] However, two opinions appear late in the tenure of the 1909 Act, in which the Southern District of New York mentioned in passing, and the Second Circuit placed reliance on, the parties' failure to consider themselves joint authors in order to resist the 12th Street Rag doctrine. [FN27]

Thus, these cases may be precursors to the requirement of an intent to be joint authors imposed by courts beginning in the late 1980s. Note that these cases are not direct precedent to the 1976 Act joint work cases, because that Act's legislative history indicated that its definition of a joint work was to be consistent with the line of cases prior to the 12th Street Rag case. [FN28]

In the first of these two cases, Picture Music, Inc. v. Bourne, Inc., [FN29] it was only under the Second Circuit's 12th Street Rag doctrine that the song in question could be a joint work, since the assignee, Berlin, rather than the original author, conceived the intention for the second contribution to be made. [FN30] The Southern District of New York, unhappy with a finding of a joint work, which seemed compelled by the 12th Street Rag precedent, noted that that case had been criticized by commentators and would be overruled by what would become the 1976 Act. [FN31] The court then relied primarily on the insufficiency of the second contributor's contribution to find that the song was not a joint work. [FN32] After stating its conclusion and rationale, the court added that "[i]t was not the intention of the parties that [the second contributor] should become a joint owner of the source material or of the popular song so closely derived therefrom." [FN33] The court continued, in what appears to have been intended as explanation, that the second contributor did not become a co-owner of the source material because the "original work was not transferred to her," nor did she become a co-owner of the final song because she had not made a "significant contribution." [FN34] The court's preceding reference to the parties' lack of intention to become joint owners is inexplicit at best. [FN35]

In Gilliam v. American Broadcasting Cos., [FN36] the Second Circuit's partial reliance on the parties' lack of intent to be joint authors in order to conclude that a finding of no joint authorship was likely, is only slightly more illuminating. [FN37] After a very brief discussion, the court concluded that the "matter is subject to further exploration at the trial," but would not bar a preliminary injunction. [FN38] The Second Circuit apparently believed that the 12th Street Rag doctrine, criticism of which the court noted, was necessary for a finding that the television program at issue was a joint work. [FN39] In fact, joint authorship could have been found in Gilliam under Marks, since the writers had intended, at the time they wrote the script, that it would be embodied with other contributions in a single work, the television program. [FN40] Further, the contract provision that the script writers would "retain all rights in the script not granted in the agreement," used by the court as evidence that "the parties did not consider themselves joint authors," in fact functioned much more directly as an agreement by the television producer to assign any joint authorship rights it might have, other than those specified in the agreement, to the script writers. [FN41] Precursors under the 1909 Act to the requirement of intent to be joint authors appear to be limited to the brief and ambiguous discussions in Picture Music and Gilliam.

The picture of the definition of a joint work as of the adoption of the 1976 Act, then, is one of a doctrine relatively recently developed, with the definition expanding beyond the tolerance of the courts and Congress, but with little in the way of precursors to the limits on joint works recently imposed by courts. Congress proceeded in 1976, in defining a joint work as "a work prepared by two or more authors with the intention that their contributions be merged into inseparable or interdependent parts of a unitary whole," [FN42] to 1) explicitly reject the possibility of requiring inseparable contributions; [FN43] 2) reject case law permitting intent for a larger work to be formulated later and by one other than the first contributor; and 3) agree with case law permitting joint authors to work apart and at different times so long as there was intent to contribute to one work. [FN44] Thus, Congress struck a middle ground. We turn next to the courts' treatment of joint works under the 1976 Act.

III. Treatment of Joint Work Doctrine Under the 1976 Act
In examining cases decided under the 1976 Act, we find that courts generally show hostility to the joint work doctrine. [FN45] This hostility manifests itself in a variety of ways: the language in which joint work doctrine is discussed in judicial opinions; the subordination of joint work doctrine to the work-made-for-hire doctrine; conclusory findings that contributions have not properly merged; and narrow definitions of a joint work adopted by courts.

Given the courts' wariness about joint work doctrine, it is not perhaps surprising that the very tone in which courts discuss the existence of a joint work is at times inhospitable to the doctrine. [FN46] A striking example was the Second Circuit's decision in Childress v. Taylor, [FN47] in which the court affirmed a grant of summary judgment to the plaintiff, a professional writer, on the grounds that an actress was not a co-author of a play. [FN48] The Second Circuit justified its decision to require that each co-author make a copyrightable contribution [FN49] by arguing that this requirement "might serve to prevent some spurious claims by those who might otherwise try to share the fruits of the efforts of a sole author." [FN50] Aside from the circularity of the reasoning (it is the court's requirement which makes certain otherwise valid claims spurious), the picture drawn by the court is of a dishonest interloper trying to horn in on the hard-won product of another's labors. The court struck the same righteous tone in concluding that "[a] playwright does not so easily acquire a co-author." [FN51]

In addition to the language used by courts, the courts show hostility to joint work doctrine in their integration of that doctrine with work-made-for- hire doctrine. The choice made by courts to subordinate joint work doctrine to work-made-for-hire doctrine represents avoidance of joint works. Where both doctrines are implicated, courts regularly permit work-made-for-hire doctrine to take priority over joint work doctrine by considering the possibility of a joint work only after having considered and rejected the possibility of a work made for hire. [FN52] Consider the discussion of the District of Columbia Circuit in Community for Creative Non-Violence v. Reid, [FN53] a case in which the court appeared otherwise well-disposed toward joint work doctrine. [FN54] The court stated that had it adopted a broader interpretation of the scope of work made for hire, it would have affirmed the district court's judgment that CCNV, as employer, was the owner of the sculpture at issue as a work made for hire. [FN55] The court made this statement in spite of the fact that it thereupon characterized the sculpture as a "textbook example of a jointly-authored work." [FN56] Thus, although the sculpture looked to the court like a joint work, if the definition of a work made for hire had been satisfied, the court would have ignored the possibility of a joint work. The Second Circuit, in Childress v. Taylor, similarly made it clear that the work- made-for-hire doctrine trumps joint work doctrine. [FN57] Where an editor is employed by an author, the court explained, the author would own the editor's contributions under work-made-for-hire doctrine. [FN58] Only where the editor was not the employee of the author would a claim of joint authorship "not be defeated by the 'work made for hire' doctrine." [FN59] This result, the priority of work-made-for-hire doctrine over joint works, is not required by the Act, which defines both types of work, but nowhere indicates the relationship between the two. [FN60] The choice of courts to make work-made- for-hire doctrine trump joint work doctrine, although it may be justifiable, results in restricting the scope of joint work doctrine. [FN61]

Hostility to joint work doctrine is also apparent in conclusory and insupportable statements made by some courts to the effect that the parties' contributions have not, in fact, properly merged, [FN62] or that the parties did not intend merger. [FN63] For example, where a client made a "thumbnail sketch" of the desired floor plan for an architect, the Eleventh Circuit stated in MGB Homes, Inc. v. Ameron Homes, Inc. [FN64] that there was no evidence that it was the intent of either party for the sketch to become part of the finished plans. [FN65] One wonders why a client would produce such a sketch, if not to have it incorporated in the final plans, or why an architect would agree to take the job, if he had no intention of producing a plan to the client's specifications. Could the court have been failing to distinguish between any copyrightable material contained in the sketch, which would be incorporated in the final plan, and the paper the sketch was drawn on, which would not?

Most significantly, the joint work doctrine has been treated grudgingly in that some courts have defined a joint work more narrowly than required by the statutory language and its legislative history. Section 101 defines a joint work simply as "a work prepared by two or more authors with the intention that their contributions be merged into inseparable or interdependent parts of a unitary whole." [FN66] Thus, the statutory language requires only that each contributor intend that her contribution merge into a whole work, [FN67] and does not specify what the nature of that contribution must be. [FN68] The legislative history expands on the statutory definition by clarifying that joint authors need not work at the same time or together, but does not require, or even support, the narrower definitions adopted by some courts. [FN69] This judicial restriction of the joint work doctrine has been accomplished by making two adjustments to the statutory definition of a joint work. These adjustments deal respectively with the state of mind of the joint authors and with the nature of their contributions. Courts have narrowed the scope of joint work doctrine by: 1) requiring intent on the part of the creators of a work that they be joint authors and joint owners, rather than simply intent that their contributions be merged; and 2) requiring that each joint author make a copyrightable, or even more than a copyrightable, contribution.

In spite of the fact that the Copyright Act requires, in its definition of a joint work, only the creators' intent for their contributions to merge, many courts impose a stricter requirement. [FN70] Although such judicial opinions both state and apply a higher intent standard than that provided by s 101, they frequently quote the statutory language. [FN71] Indeed, some courts do not appear to realize that they are departing from the express language of s 101. For instance, the opinions of such courts employ in the same discussion both the standard of intent to merge, and the standard of intent to be joint authors, as if there were no difference between the two. [FN72] A few opinions make this deviation from the express language of the statute explicit and offer a justification for imposing a higher intent standard. [FN73] The justification offered is that the scope of the joint work doctrine needs to be narrowed beyond the statutory definition in order to avoid including as joint authors those to whom Congress never intended to extend that status. [FN74]

The stricter intent standard imposed by these courts requires that the parties have the intent to be joint authors, rather than simply the intent that their contributions merge, as required by the statutory definition. What the courts mean by the intent to be joint authors is an intent to share ownership in a work. [FN75] Thus, good evidence of intent for joint authorship is, according to the courts, evidence of intent for all authors to be given credit or billing as joint authors, [FN76] or evidence of intent that all authors' names appear as authors on a copyright registration form or be otherwise identified as copyright owners. [FN77] The intent to be joint authors is distinct from an intent to merge contributions. For example, a party could intend her contribution to merge with that of another party while intending that that party not be a co-owner of the completed work. [FN78] That the intent to be joint authors is distinct from an intent to merge contributions is further demonstrated by Judge Leval's explanation in Fisher v. Klein [FN79] that "there is this additional requirement of the shared intention that the contributions be merged into a unitary whole . . . . It is only where that dominant author intends to be sharing authorship that joint authorship will result." [FN80] In effect, the intent to be joint authors is intent for the legal conclusion, co-ownership, that results from a finding of intent for contributions to merge. [FN81]

The intent-to-be-joint-authors requirement owes its origin to a variety of sources. The leading case is perhaps the Second Circuit's 1991 decision in Childress v. Taylor, [FN82] in which the court found that because a playwright did not regard an actress as a joint author, even if the actress's contribution were sufficient, joint authorship would not result. [FN83] The Childress opinion contains a fuller and more explicit discussion of the intent-to-be- joint-authors requirement than had the few earlier cases that made the same requirement. [FN84] The Second Circuit stated the intent standard as "whether the putative joint authors regarded themselves as joint authors." [FN85] The district court in Childress had appeared to rely primarily on its finding that the actress' contributions failed to satisfy the contribution standard for joint authorship, and had not explicitly stated a requirement of intent to be joint authors, but in its recitation of the facts, had stressed that the playwright did not intend co-authorship. [FN86] The Second Circuit's Childress decision was antedated by the same court's decision in Weissmann v. Freeman, [FN87] in which the court stated an intent-to-merge requirement, but emphasized that the scientist did not intend to share credit for the work with her mentor, and concluded on this basis that "this work was not intended to be joint." [FN88] The Childress court relied on Weissmann, as well as on Fisher, for the proposition that intent to be joint authors is necessary. [FN89] The latter case was decided in 1990 by the Southern District of New York, which stated that where one creator's contribution is larger, "[i]t is only where that dominant author intends to be sharing authorship that joint authorship will result." [FN90] Also preceding Childress were several architecture cases, which, although they contained nothing explicit regarding intent to be joint authors, did support the view that it is the parties' expectations concerning ownership, rather than intent to merge contributions, that is significant. [FN91] These courts stated that in the relationship between an architect and a client, it is expected that the client will make a sketch or provide design features, the implication being that joint authorship is not expected, and therefore, does not result. [FN92]

Most other circuits considering the issue of the required intent for joint authorship appear to have adopted the Second Circuit's position. The Seventh Circuit, finding in 1994 in Erickson v. Trinity Theatre, Inc. [FN93] that actors were not the joint authors of three plays, required that the parties "have intended to be joint authors." [FN94] The Ninth and Eleventh Circuits have mentioned the intent for a work to be joint, but too briefly to permit any certainty as to what the courts meant. [FN95] By contrast, the District of Columbia Circuit, in its discussion of joint authorship in Community for Creative Non-Violence v. Reid [FN96] in 1988, required only that the parties intend "to merge their contributions into a unitary whole." [FN97]
Imposition of a requirement of intent to be joint authors results at times in overly limiting findings of joint works because in some genuine collaborations, one party did not regard the other as a co-author or intend for the other to own an interest in the resulting work. For instance, this appears to have been the case in Weissmann v. Freeman, decided by the Second Circuit. [FN98] Due to that court's requirement of an intent for joint authorship (that is, joint ownership), evidenced by joint credit, the court found that two scientists' collaboration on a joint work over a period of five years had terminated in 1985 when one scientist made additional changes to the work and placed on that most recent version her own name alone. [FN99] Had the court required only the parties' intent for her latest contributions to merge with the prior contributions of both parties, that standard would have been easily satisfied, since both parties expected additional versions of the syllabus (a paper reviewing past research on a given topic to be used as a handout at a lecture) to be prepared by one or the other of them. [FN100] The scientists in Weissmann had been passing the syllabus back and forth for five years making changes and additions to it. [FN101] The circuit court's finding that the latest version was a derivative work prepared by one of the prior joint authors was made possible by its imposition of a requirement that both parties intend that they share ownership. [FN102] The court relied on one scientist's use of her own name alone on the syllabus in 1985 to find that she lacked the intention for the work to be joint. [FN103] The court thus put it in the power of one joint author to unilaterally exclude the other from that status.

Similarly, the requirement of intent for joint authorship produced an overly restrictive concept of joint works in a case dealing with collaboration between a playwright and an actress. The play at issue in Childress v. Taylor may well have constituted a sufficiently genuine collaborative product for joint work status, since both the playwright and actress intended their contributions to merge, and since the actress's contributions may have satisfied the applicable standard. [FN104] The Second Circuit expressly did not decide whether the actress's contributions of selection of facts, scenes, and characters to be included in the play satisfied the copyrightability standard enunciated by the court, because the playwright did not intend that the actress receive credit or co-own the play and, therefore, did not intend joint authorship. [FN105] Should the playwright's failure to regard the actress as a co-author deny the actress that status if, in fact, she was one?

One contributor's unilateral intent that the other not own an interest in a work should not prevent finding a joint work, assuming appropriate contributions from both and the intent to merge their contributions. If it is necessary to exclude from joint authorship status certain contributors, such as the case of typical editors, [FN106] who may satisfy the requirements of sufficient contribution and intent to merge, a safer basis for such exclusion would be that neither contributor intends one of them to be a joint author.

Note that where one alleged co-author is a professional, this circumstance appears to bias some courts against joint authorship. [FN107] Such courts appear to assume that both parties expected that the non-professional would own no interest; they therefore conclude that the non-professional owns none. For example, where clients have provided input to architectural plans, courts have explained that it is "normal" for clients to do so, but that such contributions do not make the clients authors of the plans. [FN108] The implied premise in this argument appears to be that no one expects the clients to have joint ownership. Where, in fact, the client did not intend to be excluded from ownership, and where the client's contribution was sufficient and both parties intended merger, one party's professional status should not alone preclude joint authorship. An absolute bar to joint authorship does, in fact, appear to be contemplated by some courts where one contributor is a professional. For instance, the district court in Whelan Assocs., Inc. v. Jaslow Dental Lab., Inc., [FN109] analogizing the client-computer programmer relationship to that between a client and an architect, concluded that the "architectural drawings are not co-authored by the owner, no matter how detailed the ideas and limitations expressed by the owner." [FN110]

In addition to narrowing the intent requirement for the creation of a joint work, some courts have restricted the scope of joint work doctrine beyond that provided by the Copyright Act with respect to the nature of a joint author's contribution. [FN111] Neither the 1976 Act nor its legislative history makes reference to the type of contribution required for joint authorship. [FN112] More recently, however, legislation was proposed, but not enacted, which would have inserted the word "original" into the definition of joint works, thus requiring that each joint author make an original contribution. [FN113] This bill was apparently intended to require that each contribution be both qualitatively copyrightable (i.e. constitute expression) and quantitatively copyrightable (i.e. contain sufficient expression to sustain a copyright). [FN114] The bill, which was never reported out of committee, also required a signed writing for the creation of certain joint works, and would have amended the definition of works made for hire, as well as that of joint works. [FN115] Thus, Congress' abandonment of this amendment does little to clarify Congress' views regarding the nature of the contribution required from a joint author.

In the face of this vacuum left by Congress, courts have imposed various limits on the contribution necessary to qualify as a joint author. One view, attributed to Nimmer, and rejected by most courts that have considered it, is that any contribution which is more than de minimis is sufficient. [FN116] However, attributing this proposed standard to Professor Nimmer may be a misreading of his treatise. In the cited passage of the treatise, Nimmer appears to say that it is unclear how much is required from a joint author, but that the contribution must at least be more than de minimis. [FN117] Saying that any contribution which exceeds the de minimis level satisfies the requirement is very different from merely saying that a de minimis contribution is insufficient. Nonetheless, in addition to discussing the quantitative standard, Professor Nimmer expressed the view that a sufficient contribution could consist of ideas which "standing alone would not be copyrightable." [FN118] This position regarding the qualitative standard for a contribution to a joint work was brought to prominence by the District of Columbia's suggestion, in Community for Creative Non-Violence v. Reid, that it might be correct. [FN119]

In spite of the absence of any restriction explicitly stated in the statute or legislative history, most courts that have considered the nature of the necessary contribution have required that each contribution be independently copyrightable. [FN120] To the limited extent that the courts have attempted to supply authority supporting this requirement, they have done so by arguing that the term "author" in the Copyright Act or the Constitution implies a requirement of originality, so that each joint author's contribution must be a copyrightable work of authorship. [FN121] Certainly, in the case of a sole author, the requirement of originality has been supported by finding that the term "authors" in the Constitution implies satisfaction of a standard of originality. [FN122] On the other hand, it has been argued that a constitutional requirement of originality cannot apply to all uses of the term "author" in the Copyright Act, since the employer, as author of a work made for hire, need not have supplied any of the originality herself. [FN123] However, a distinction can perhaps be drawn between a joint author and an employer as author of a work made for hire, because in the latter case, the employer takes the copyright vicariously for the employee who did create a protectible work. In the case of a joint author, there is no one for whom a joint author can take the copyright vicariously, unless it be his joint author. Nonetheless, the argument that either the Constitution or the Copyright Act implies a requirement of originality for each joint author, although not illogical or unreasonable, is inconclusive.

A few courts have raised policy arguments to justify requiring a copyrightable contribution from each joint author. [FN124] One such policy argument is fairly persuasive, given that copyright law is intended to encourage creators to make use of others' ideas. [FN125] Where both an idea supplier and idea user intended the idea to be merged into the user's work, the statutory intent requirement would be met. If supplying an idea alone satisfies the contribution requirement, the idea originator might then claim joint authorship. Thus, use of others' ideas in the creation of works might subject creators to the risk of losing exclusive ownership of their works; this risk might make creators wary of using others' ideas. [FN126] This would run counter to a basic tenet of copyright law, that the progress of science is promoted by encouraging creators to use and build upon the ideas of others. [FN127]

It should be noted that the copyrightability standard for joint authorship does not by its terms appear to require much, if anything, more, quantitatively, than the more than de minimis standard. [FN128] The standards for copyrightability set forth by the Supreme Court in Feist Publications, Inc. v. Rural Telephone. Service Co. [FN129] are low. [FN130] Indeed, Justice O'Connor wrote for the Court that "copyright protects only those constituent elements of a work that possess more than a de minimis quantum of creativity." [FN131] Thus, the standard for copyright protection enunciated in Feist may be no higher, quantitatively, than the more than de minimis standard for joint authorship. The main impact of the copyrightability standard for joint works, in theory at least, is to exclude claims based upon the contribution of ideas alone.

In application, however, some courts which espouse the copyrightability standard, have in fact required more than a copyrightable contribution. [FN132] Requirements beyond a copyright able contribution may in some cases be too high and result in an excessively narrow view of joint works. Such appears to have been the case in the Seventh Circuit's decision in Erickson v. Trinity Theatre, Inc. regarding "The Theatre Time Machine", a play in which the defendant theatre claimed joint authorship with a playwright based on contributions made by actors. [FN133] In spite of the court's finding that "the evidence shows that the actors were involved in the development of the melodrama and improvisational scenes," the court concluded that the theatre had not satisfied the copyrightable contribution requirement because the actors could not identify the specific contributions which they had made. [FN134] Although the contributions of the actors to these scenes through the process of improvisation may well have been copyrightable, the contributions were rejected by the court because they were inseparable from the playwright's contributions. [FN135] Requiring a putative joint author to identify exactly which words, shades of meaning, and influences she contributed to a scene conflicts with the Copyright Act, which explicitly contemplates the possibility of inseparable contributions. [FN136]

A similarly questionable result was reached in Balkin v. Wilson, [FN137] in which the district court endorsed the copyrightability standard, but granted the plaintiff's motion for summary judgment on the issue of infringement. [FN138] The court granted this motion in spite of the fact that the defendant had contributed "precise technical specifications" to the songs at issue, that the parties had discussed together the "content" of the songs before their composition, and that the plaintiff described himself as the "mechanic" of the songs and the defendant as the "architect." [FN139] Although the defendant did not claim to have written "a single note of the music or a single word of the lyrics," could not his contributions of technical specifications, content, and structure be copyrightable? [FN140] Thus, the court may well have in fact required more than a copyrightable contribution. Consider also Cabrera v. Teatro del Sesenta, Inc., [FN141] in which the District of Puerto Rico found that the contributions of four collaborators to a play were not sufficient to support joint authorship because they were not "tangible." [FN142] The court rejected the collaborators' contributions despite the court's previous conclusion that their participation was "real, effective and crucial throughout the writing of the script," [FN143] and despite the court's ultimate conclusion that its interpretation of the law "may have produced what might be considered an unfair result." [FN144] By "tangible," the court appears to mean contributions fixed in a tangible medium by the collaborators themselves. [FN145] The court thus made a requirement beyond copyrightability, by requiring that each collaborator physically fix his contribution, rather than orally communicate his contribution to another collaborator for fixation.

The courts have thus taken, under the 1976 Act, a restrictive view of joint works. A narrow view of joint works broadens the scope of infringement since joint work doctrine is so often invoked as a defense to a claim of copyright infringement. In a majority of cases in which the existence of a joint work has been litigated under the current Act, joint authorship has been raised as a defense to an infringement claim. [FN146] Since a joint author has the right to independently use or license the entire work, she cannot be liable to another joint author for copyright infringement of that work. [FN147] Joint authorship is thus a complete defense to a claim of copyright infringement. Therefore, given the fact that joint work doctrine is so often raised as a defense to infringement, a restrictive view of joint works favors finding infringement and may contribute to the over-enforcement of copyrights. [FN148]

IV. Intent For Future Use of Internet Communications
We turn now to further implications of the courts' restrictive view of joint works, in the limited context of serial collaborations, that is, works to which contributions are made consecutively. [FN149] Such serial collaborations implicate the courts' narrow view of joint works elaborated above, particularly the requirement of an intent to be joint authors. [FN150] Serial collaborations have gained significance with the increasing popularity of the Internet, which permits widespread dissemination of works in a form in which they are easily manipulated by the recipient. [FN151]
How do serial collaborations fare given the restrictions imposed by courts on joint work doctrine? The courts' hostility to the doctrine, and their use of conclusory findings that contributions have not merged, will affect serial collaborations, but not in a fashion different from their effect on any other purported joint work. [FN152] The subordination of joint work doctrine to work-made-for-hire doctrine prevents finding a joint work in many serial collaborations, such as a studio musician adding an accompaniment or a film editor rearranging footage. However, this subordination will presumably have little impact in the Internet setting, where a recipient could rarely be construed as having been hired by the author of transmitted material. [FN153] The second and subsequent contributors to serial collaborations may well make independently copyrightable contributions, as required by most courts recently considering the issue, although the manipulation of this standard to avoid joint authorship may prevent any given serial collaboration from being found a joint work. [FN154] But it is in the courts' requirement of intent to be joint authors that we find the greatest potential bar to joint authorship in the case of a serial collaboration. It is this question which we now examine in some detail.

Where a serial collaboration is alleged to be a joint work, but a joint work is not found, how is the resulting work to be classified? [FN155] Interestingly, courts rejecting the possibility of a joint work do not always consider the alternatives. [FN156] If the intent requirement for a joint work stated in the Copyright Act is applied (i.e., the intent for contributions to merge), [FN157] a serial collaboration may be a joint work if the first author produced her work with the intent that others might add to it. If, on the other hand, courts require an intent to be joint authors, a serial collaboration will be a joint work only where there is evidence that both authors intended co-ownership, such as express statements, copyright registration consented to in both names or joint credit consented to by both. [FN158] Thus, where intent to be joint authors is required, a serial collaboration is less likely to be found a joint work. Where a serial collaboration is not a joint work, it is either an infringing derivative work or a licensed derivative work, assuming sufficient use of material from the first work and the absence of fair use and other statutory exemptions. [FN159] In the absence of an express license by the first author, the question of infringement by the second author will turn on the existence of conduct by the first author creating an implied license. [FN160] Without such a license, the second contributor to the serial collaboration has infringed.

The foregoing scenario has increased in importance with the rapid growth of the Internet in recent years. [FN161] The Internet particularly implicates serial collaborations because it provides widespread availability of material protected by copyright, because it provides the material in a form in which it can be easily manipulated and adapted, [FN162] and because a culture of free use has prevailed in some Internet contexts. [FN163] A handful of judicial decisions have confirmed that copyright infringement can occur via the Internet, [FN164] and if the recommendations of the Working Group on Intellectual Property's final report are ultimately enacted by Congress in some form, liability for transmission via the Internet will be that much clearer. [FN165] A case in point may help illustrate the significance of serial collaboration via the Internet. In 1994, the Estate of Elvis Presley demanded that the creator of an Elvis Presley Home Page stop transmitting a "Cyber Graceland Tour" containing Elvis Presley songs and photographs of Graceland in which the estate held the copyright. [FN166] If the creator of the tour received the sound clips or photographs via the Internet, this was a good example of Internet serial collaboration. A first author distributed material via the Internet and a second author contributed the tour arrangement to produce the final work, which, assuming sufficient use of material and the absence of exemptions, is either an infringing or impliedly licensed derivative work, or a joint work.

To examine how the serial collaboration scenario described above might play out in the Internet context, I collected from five listserv lists (or listservs) all examples of express language employed in 1994 by contributors reflecting intent regarding future use of their messages to the lists. [FN167] The listservs selected were cni-copyright, dinosaur, music research, nerdnosh, and weird-L. [FN168] I thus chose one listserv on which most contributors would be knowledgeable about copyright law, and might therefore be likely to address the issue of future use, and four listservs in which contributors would be likely to make creative contributions, and might therefore contemplate future use of their works. My search of the archives of the five listservs for 1994 produced numerous examples of express language regarding future use in messages to cni-copyright, the listserv used mainly by copyright experts, but very few in contributions to the creative listservs. [FN169] It is curious that relatively few attempts are apparently made on listservs to clarify what future uses are intended or permitted by the copyright owner. It may be that many people who are not knowledgeable about copyright law believe that postings to listservs are not protectable by copyright, and that there is, therefore, no need to address the issue. [FN170]

Where no express language regarding future use is included in a message to a listserv, which appears to be true in the vast majority of cases, any license for making a derivative work or expression of the intent required for a joint work would have to be implied from the circumstances. It has been suggested that contributors to listservs do expect future use of their works. [FN171] If the expectation on listservs is that others will copy, in whole or in part, to comment or to make additions or other changes in the case of a creative work, this expectation could be construed as either an implied license for such use, or as the intent required for a joint work. As between these two outcomes, which is more advantageous to the first author? If the resulting work is a joint work, the first author can use and license that work, and is entitled to an accounting from the second author for any profits. [FN172] If, on the other hand, the first author is found to have granted an implied license for creation of a derivative work, she receives none of the above benefits. Consider here the impact of requiring, for creation of a joint work, on the one hand, intent for contributions to merge, and on the other hand, intent to be joint authors. [FN173] An expectation that one's contribution to a listserv might well be modified by another list subscriber, together with the decision to make the contribution under those circumstances, could be construed as intent for the contributions to merge. Such an expectation is most unlikely, however, to be construed as an intent to be joint authors together with some unknown individual. Thus, under these circumstances, requiring an intent for contributions to merge would permit the joint work outcome, whereas requiring an intent to be joint authors would probably preclude that result. Bear in mind that finding a joint work is more beneficial to the first author than finding an implied license for a derivative work. Of course, in the absence of an expectation that others will modify postings to listservs, any such modification would, assuming sufficient use and the lack of an applicable exemption, constitute an infringing derivative work.
We now examine the impact of the express language regarding future use which I found included in messages on the five listservs. Among the express language regarding future use were waivers of copyright, [FN174] such as "Unless otherwise indicated, I make no claim of copyright in messages I write for transmission over the Internet" and "No copyright claimed, reproduce at will should you inexplicably care to." [FN175] Such waivers appear intended, and should be effective, to place the material in the public domain. [FN176] Note that the first waiver quoted above was followed immediately by, "I hereby grant permission to all others to reproduce, retransmit or copy such messages, subject only to the request to retain the signature line or to give appropriate credit for the use of any such material." [FN177] This addendum is inconsistent with a waiver, since if there is no copyright, no permission is needed, and thus the addendum renders the waiver, as a whole, ambiguous. [FN178] Where a waiver is effective, and the material is therefore in the public domain, any subsequent work employing that material would be a derivative work based on public domain antecedents, and would not be a joint work.
Some express language on the five listservs regarding future use was ambiguous as to whether the intent was for reuse limited to unmodified uses of a work in its entirety, or for any reuse including modification. Consider the following four examples: 1) "I grant permission for all my postings to this list to be forwarded to other lists and freely circulated on the nets." 2) "I hereby give permission for the use of any material contained in this message but ask that the user give me appropriate credit." 3) "I also should have said that anyone can feel free to re-post elsewhere for non-commercial purposes. Attribution is appreciated, but not necessary." 4) "Feel free to spread it far and wide, but remember you saw it here first." [FN179] If such language is construed to include intent for modifications, the expression of that intent could constitute either a license to make derivative works, or expression of intent for the author's contribution to merge with others' contributions. The result in the latter case, assuming sufficient contributions, could therefore be a joint work. However, such language does not express intent to be joint authors, and thus, where intent to be joint authors is required, no joint work could result.

Unambiguous language prohibiting modifications was less commonly employed on the five listservs I examined. Language such as, "This message may be freely distributed, but only if unmodified and not used for commercial purposes," [FN180] precludes the possibility of a joint work under either intent requirement, and, assuming sufficient use and the absence of exemptions, renders any derivative work infringing. Even more rare on the listservs I examined was language expressly prohibiting any use. [FN181] Such language would, of course, preclude a joint work under either intent requirement, and leave any derivative work an infringement, again assuming sufficient use and the absence of exemptions. The scarcity of language expressly prohibiting any use or prohibiting modifications offers some support for the view that listservs are an environment in which creators do expect future use of their works, which has implications, as discussed above, for the vast majority of cases, in which no express language regarding future use is employed.

In sum, we have seen that serial collaborations are less likely to be found joint works under the courts' current requirement of intent to be joint authors, than under a requirement of intent for contributions to merge. [FN182] Where, as may be true at least in some contexts over the Internet, there is an expectation of future use, the requirement of intent to be joint authors is disadvantageous to the first author, because it will probably produce a finding of an implied license for derivative works, rather than a finding of a joint work. This result may be one more argument, in addition to those adduced in Part III above, against requiring an intent to be joint authors.

What advice for users of the Internet does the preceding analysis of express language used on the five sample listservs suggest? Waivers of copyright, in order to be effective, should be unambiguous. Where the author's intent is to permit only certain uses, those uses should be enumerated. Where reproduction, whether in whole or in part, is to be permitted, but no other modification, such modification should be expressly excluded. Finally, in a context such as the Internet, in which material may be widely available and readily adapted, and in which a culture of free use may prevail, it is important for authors to expressly state their intentions with respect to future use.

V. Conclusion
In constructing a definition of joint works, Congress chose, from the precedents available to it in 1976, the requirement that putative joint authors have the intention, at the time of production, that their creations contribute to one work. The limitations imposed by courts beginning in the late 1980s on the definition of a joint work, in the form of requirements of intent to be joint authors and independently copyrightable contributions, were not then an integral part of the judicial definition of joint works. These limitations form part of a narrow view of joint works taken by courts under the current Copyright Act. This restrictive view has led, in at least a handful of cases, to decisions that have denied creators who should be joint authors that status. We may expect to see more such cases in a variety of contexts; in particular, it is not unlikely that the Internet will provide a fruitful source of such controversies. Based on a limited investigation performed by the author, it appears that the failure of contributors to the Internet to expressly state their intentions regarding future use of their works will contribute to such conflicts. Finally, to the extent that intent for future use is shared by Internet contributors, a requirement of intent to be joint authors benefits the second author of serial collaborations at the expense of the first.