The Idea-Expression Dichotomy in Copyright Law

There is hardly a single principle of copyright law that is more basic or more often repeated than the so-called idea-expression dichotomy.[FN1] The doctrine is followed dutifully as an unquestioned [p. 323] principle in hundreds of cases: the "ideas" that are the fruit of an author's labors go into the public domain, while only the author's particular expression remains the author's to control. This principle, sometimes described as having constitutional origins, was developed by the common law, and has now been incorporated into the copyright act itself.[FN2]

Some commentators who have studied the idea-expression dichotomy in the greatest detail have criticized it, arguing that continued recognition of the dichotomy is neither justified nor helpful in deciding cases.[FN3] Yet, not only have the courts continued to embrace [p. 324] the idea-expression dichotomy, they have extended it to explain related copyright problems, which over the years had also led to confusion.[FN4] Nevertheless, the cases generally do not analyze the principle in detail; and rarely (if ever) is the doctrine actually decisive of particular cases. Perhaps this should not be surprising, since the doctrine is so general in its statement as to defy particular application. It is not a doctrine that could be used predictably to put a particular work either into the public domain or within the author's exclusive rights; instead, it seems to be an ex post facto characterization that justifies an outcome based upon other, more concrete, factors. Thus, if the outcome in a particular case is to be infringement, the work is deemed to be protectable expression; if the outcome is to be noninfringment, then the work is described as an "idea."[FN5]

Given the difficulty of defining the terms of the doctrine, some courts and commentators have developed an "abstractions" test[FN6] or a "patterns" analysis,[FN7] which purports to place a given work along a continuum between idea and expression. Although it is impossible to state precisely when a particular work has crossed the threshold from one end to the other, the courts are nonetheless supposed to struggle to apply the terms. These terms may not be precise, but they are no more imprecise than general concepts of negligence or fault, which the courts must also deal with on a case-by-case basis.[FN8] The general principle embodied in the idea-expression dichotomy is supposed to help in focusing the inquiry and resolving particular cases in accord with fundamental copyright values.

If the characterization, the abstraction, or the continuum is too vague to determine particular cases, however, then this principle is not as important as it might first appear. It is helpful to have a model that explains why certain decisions have to be made, and why not all cases will be easy to decide. But if other factors are at work in the idea-expression cases, then those specific factors should be examined more closely rather than blindly repeating the principle as dogma.

[p.325] This discussion of the idea-expression dichotomy is divided into four sections. Part II traces the history and development of the doctrine; Part III analyzes and criticizes the doctrine; Part IV discusses the constitutional considerations involved in the doctrine; and Part V analyzes the relationship between the idea-expression dichotomy and other basic principles of copyright law.

II. The Idea-Expression Dichotomy: History and Development

Some of the early copyright cases, from which the idea-expression dichotomy developed, purported to establish a true dichotomy between the idea and the expression of a copyrighted work. Copyright in the early days protected only against literal copying, and not against a more abstract taking of a copyrighted work. Judge Learned Hand explained the idea-expression dichotomy by means of a less precise "abstractions" test. He stated: "Upon any work . . . a great number of patterns of increasing generality will fit equally well, as more and more of the incident is left out."[FN9] At its most concrete level, a work was protected by copyright; but at some level of abstraction, it became more of an "idea," to which the protection of copyright did not extend. The abstractions test was significant because copyrighted works were given protection on some level of abstraction, although that protection did not extend to the greatest level of abstraction. Judge Hand's abstractions test thus has expanded the eligibility for copyright by protecting those works that might be somewhere along the continuum between the expression and the idea of a work.

Some recent computer copyright cases have expanded the scope of the idea-expression dichotomy even further, indicating a tolerance for protection of copyrighted computer programs at such an abstract level that one might wonder whether there is any vitality left in the idea-expression dichotomy at all.[FN10] The courts in these cases did not purport to lay down a special rule for computer programs; rather, they framed their analysis in terms of generally applicable copyright principles.[FN11] In order to understand the historical shift in the idea-expression dichotomy—from a presumption leaning toward the "idea" (and thus, nonprotection) to a presumption leaning toward the "expression" (and thus, protection)—it is first necessary to trace the idea-expression dichotomy to its roots.

[p. 326] A. Formal Origin: Baker v. Selden

The idea-expression dichotomy in America is said to have originated in the United States Supreme Court case of Baker v. Selden.[FN12] The case was based in large measure upon others that were somewhat suspect, or that have been at least partially repudiated in this country as well as in England. Even if the reasoning in Baker may be attributed to a faulty or rejected reading of the law, the case has never been overruled; in fact, it has been so enthusiastically incorporated into the fabric of copyright law as perhaps to render its origins irrelevant.[FN13] In its explanation and development of what has come to be known as the idea-expression dichotomy, the case is shrouded in the limited view of copyright that was typical of its day. The weakness (or obsolescence) of the Supreme Court's argument does not necessarily compel a rejection of its holding. A careful analysis of the case, however, sheds light upon the role of the doctrine, and perhaps imposes upon those advocating the doctrine the burden of finding a more contemporary justification for its existence.

In Baker, Selden [FN14] obtained a copyright in a book which disclosed his particular bookkeeping system.[FN15] The book contained forms "consisting of ruled lines, and headings, illustrating the system and showing how it is to be used and carried out in practice."[FN16] Baker published a book on accounting, which "use[d] a similar plan so far as results are concerned; but ma[de] a different arrangement of the columns, and use[d] different headings."[FN17] The complainant alleged Baker had used Selden's system of accounting, but did not allege that Baker had actually copied the particular forms contained in Selden's book.[FN18] The Supreme Court concluded that Selden had no [p. 327] exclusive rights under the copyright laws to prevent the use of the accounting system.[FN19] The Court noted that "there is a clear distinction between the book, as such, and the art which it is intended to illustrate."[FN20] The former could be copyrighted, the latter could not.[FN21] The Court's conclusion can be characterized in four different ways:(1) a "use" test; (2) a "merger" theory; (3) a "categorical" approach; and (4) an "exclusive rights" test.

1. The Holding

a. The Use Test

The Court distinguished between the use of a work and the statement or explanation of it, implying that infringement depended upon the purpose for which a copy is made.[FN22] This approach, to the extent that it focuses upon the purpose of the actual user in making the copy (whether for use or for explanation), has been criticized.[FN23] Although the nature of the use might be relevant in determining otherwise relevant issues of fair use under section 107 of the Copyright Act,[FN24] or other specific limitations under other sections, the copyrightability of the subject matter in the first instance should not [p. 328] otherwise depend upon the nature of the use by a particular user.[FN25]

To the extent that the Supreme Court focused upon the use that the creator makes of the work, or to the extent that the only purpose of a given work is in its use rather than its explanation, the distinction between explanation and use offered in the Baker case has been followed by later courts.[FN26] Even in this limited context, however, the continued vitality of the Baker case has been questioned.[FN27]

[p. 329] b. The Merger Theory

The Court in Baker invoked a merger theory by focusing upon the utilitarian or practical nature of the original work. The Court concluded that "where the art it teaches cannot be used without employing the methods and diagrams used to illustrate the book, or such as are similar to them, such methods and diagrams are to be considered as necessary incidents to the art, and given therewith to the public."[FN28]

If this theory is the primary justification for the Baker outcome, Baker might be limited to particular kinds of works—architectural works, fashion designs, charts and forms—which by their nature merge idea and expression. Thus, the Baker holding need not be viewed as one that would apply to the broader category of works that are capable of more than one type of expression.

The merger theory has also been criticized. Professor Nimmer, for example, has argued that "it is factually erroneous to conclude that there is any system or method which can be performed by the use of only one particular form of written expression.[FN29]

c. The Categorical Approach

By invoking a categorical approach, the Supreme Court in Baker concluded that "blank account books are not the subject of [p. 330] copyright"[FN30]—thereby excluding blank account books from copyright protection, whether the books are copied for purposes of use or statement. On an analytical level, this conclusion must be drawn from some broader principles (such as the idea-expression, "works of utility," or merger principles). The explanation uses an ad hoc approach, however, so that categories of works other than blank account books are not necessarily affected. Another valid reason for finding that blank forms are not copyrightable is that they do not reflect sufficient originality to constitute copyrightable material.[FN31] The Court in Baker, however, did not address this argument.

d. The Exclusive Rights Test

The Supreme Court stated that the copyright in the book, although effective to protect the book, "did not confer upon [the author] the exclusive right to make and use account-books, ruled and arranged as designated by [the author] and described and illustrated in said book."[FN32] This explanation seems to address itself more to the scope of the exclusive rights that might otherwise be granted copyright protection than to the threshold question whether blank account books could ever, under any circumstances, be copyrighted.

According to Professor Nimmer, this limited explanation is the true holding of Baker, with the other explanations representing expansions by later courts.[FN33] Professor Nimmer stated that "[s]ince the defendant's forms . . . were not substantially similar to plaintiff's, the decision could have rested simply on the basis of absence of similarity in the expression of an idea, even though there may have been similarity in the idea of a bookkeeping system."[FN34] Professor Nimmer also noted that many of the courts invoking Baker, particularly for the merger rationale, "have found that the writings in issue were in themselves by no means substantially similar."[FN35] Although he criticized the Baker opinion and certain interpretations of it, and argues that it and other cases could have been decided as easily by focusing upon substantial similarity, Professor Nimmer nonetheless has acknowledged the importance of the idea-expression dichotomy generally, and does not seem to view the fallacy of the cases as an opportunity for questioning the validity of the doctrine itself.[FN36]

[p. 331] 2. Precedential Authority

The precedents cited by the Supreme Court in the Baker case are of dubious vitality today [FN37] The Court first cited Clayton v. Stone,[FN38] which held that daily price quotes were not copyrightable.[FN39] The court in Clayton observed the constitutional prerequisite that copyright "promote the progress of science,"[FN40] and concluded that "[t]he term 'science' cannot, with any propriety, be applied to a work of so fluctuating and fugitive a form as that of a newspaper or price-current, the subject-matter of which is daily changing, and is of mere temporary use."[FN41] The court stated that the Copyright Act "was not intended for the encouragement of mere industry, unconnected with learning and the sciences."[FN42] Such a rule, if read over broadly, would exclude a large number of works from copyright. Although there have been recurring reminders of the begrudging approach exemplified in Clayton, it has been established that newspaper articles and price-quotes are not per se excluded from copyrightability, at least not on the ground that they are "temporal" or merely the result of "industry" rather than "learning."[FN43]

[p. 332] The Court next cited the British case of Cobbett v. Woodward,[FN44] which held that a claim to copyright in a catalogue of furniture sold by the plaintiff was not valid.[FN45] The court in Cobbett found under English law a "mere advertisement" for the sale of particular articles was not a proper subject of copyright.[FN46] The Baker opinion does not even mention that Cobbett was specifically overruled in Maple & Company v. Junior Army and Navy Stores.[FN47] Further, that Cobbett was not to be decisive of American law was made clear in the leading case of Bleistein v. Donaldson Lithographing Company,[FN48] in which Justice Holmes, writing in a deceptively offhanded manner, dismissed the argument that advertisements were not sufficiently part of the "useful arts" to qualify for copyright protection.[FN49]

The third case cited in Baker was Page v. Wisden,[FN50] an 1869 English case, which held that a cricket-scoring sheet was not a fit subject for copyright. Although the English court did make some statements which might imply that cricket-scoring sheets are generally ineligible for copyright,[FN51] the case actually was decided on the particular [p. 333] facts, primarily the lack of any novelty in the particular scoring sheet: in one example, there was evidence of the use of earlier similar scoring sheets, which evidence presumably would not have been necessary if scoring sheets were categorically excluded from protection.[FN52] The Page court also seemed to be influenced by an earlier suit before the same judge, in which the Page plaintiff was a defendant and had himself claimed that the scoring sheet lacked originality.[FN53]

The Court's use of Drury v. Ewing[FN54] is somewhat confusing. In Drury, the circuit court of appeals held that a chart for cutting dresses was copyrightable under the existing copyright law, and that the defendant's chart was substantially similar and thus infringing.[FN55] Although the Supreme Court could have disregarded this lower court opinion, or limited it to the procedural setting (in which the issue of copyrightability of the charts had already been conclusively conceded by the defendant), the Court in Baker nonetheless refused to decide whether it would follow the case on the same issue. Instead, the Supreme Court purported to distinguish it. The Supreme Court stated that

such designs could only be printed and published for information, and not for use in themselves. Their practical use could only be exemplified in cloth on the tailor's board and under his shears . . . . Surely the exclusive right to this practical use was not reserved to the publisher by his copyright of the chart. Without undertaking to say whether we should or should not concur in the decision in that case, we think it cannot control the present.[FN56]

The Supreme Court in Baker seemed to conclude that the dress pattern case was easier to decide, because it could only be described as "information" (and thus protected). After all, the case did not involve a suit against a tailor who used the pattern to make clothes, but a suit against a woman who had made copies of the chart, and was selling them in competition with the plaintiff. [FN57] Such an observation does not, however, distinguish the Drury case from the Baker [p. 334] setting. Defendant Baker was not sued for actually implementing the ledger system on his own account books, just as Ewing was not accused of actually cutting the cloth patterns; rather, both of them were sued for competing with the original works by making "informational" use of them and describing their use to others. There may be distinctions between the cases, but they do not seem to be along the informational/use lines the Court described.

Although the cases cited by the Supreme Court in Baker do not compel the conclusion that the Court reaches, there may be something more compelling about the several examples that the Court used to illustrate what is not the proper subject of copyright. The Court never used the words "idea" and "expression" to describe the distinction it is drawing between unprotectable and protectable works. It developed instead the dichotomy between the informational use of a work (to convey the content or "art" of the work) and the use of the work in practice. The Court applied this distinction to state that the author of a work on medicine, ploughs, perspective, or mathematical science has rights in the precise form of describing the subject, but no legal rights, absent a patent,[FN58] in the actual operation of the art communicated. Yet, even conceding all of the examples, the outcome in the Baker case is still not compelled. There is no indication that Baker was a mere passive user of the system described by Selden's book; instead, Baker was making his money by selling books that served the same informational function as that served by Selden's book. Whether there was infringement should not have depended upon the existence of a distinction between the communication of information and its use, but rather upon the extent to which Baker's communication of the information was in fact similar to Selden's.

B. Early Uses of the Doctrine:1853-1924

Although Baker was the seminal Supreme Court case that seemed to be based, at least in part, on what might be thought of as the idea-expression dichotomy, other American courts were also citing [p. 335] the doctrine during the latter part of the nineteenth and the first part of the twentieth centuries. Before the courts embarked upon analyses of the facts of each case (usually to determine the substantial similarity between the two works involved), the doctrine was usually cited as a basic principle of copyright law.[FN59] Curiously, there were still some cases that articulated the principle in a way that was significantly different from what it has become today. These cases stated that copyright was not really in either ideas or expression, but only in the embodiment of the former in the latter.[FN60] Some cases then emphasized that the protection of the copyright law was focused not upon the abstract work, but upon the existence of a physical manuscript that was capable of being copied. For example, in the case of Johnson v. Donaldson,[FN61] the court described the basic principle as follows:

[p. 336] The law of copyright originated in the recognition of the right of another to be protected in the manuscript which is the title of his literary property. . . . It does not rest upon any theory that the author has an exclusive property in his ideas, or in the words in which he has clothed them. If each of two persons should compose a poem identically alike, he who first composed it would have no priority of title over the other, nor would he acquire priority by first publishing it. The law of copyright would protect each in his own manuscript, but would not prevent either from using his own.[FN62]

The latter part of this quote would sound familiar to modern students of copyright as a statement of the basic principle that copyright only protects against actual copying.[FN63] The earlier part of the quote, by contrast, focuses attention upon the physical manuscript, which represents the embodiment of the idea in a particular expression. The statement purports to disavow that copyright covers "expression" as an abstract concept apart from a physical manuscript, as much as it disavows the protectability of ideas. Similarly, in Stowe v. Thomas,[FN64] the court read the scope of copyright narrowly, limiting it to a literal transcript of the language in which the conceptions of the author are embodied in a tangible form.[FN65] Accordingly, the court found that a translation of Harriet Beecher Stowe's Uncle Tom's Cabin was not an infringement of the copyright in the original book, because the translated work was expressed in different words.[FN66]

[p. 337] Another famous case involving the protectability of copyrighted works is White-Smith Music Publishing Company v. Apollo Company.[FN67] Here, the United States Supreme Court held that a copyright in a musical work was not infringed by the making of a piano roll which was to be used in a player piano to play the music depicted in the copyrighted sheet music.[FN68] Although the holding of the case speaks more directly to the issue of what constitutes a copy for purposes of infringement, this matter also relates to a conception of the copyrighted work in its most narrow, physical sense. The Court stated that "[t]he statute has not provided for the protection of the intellectual conception apart from the thing produced, . . . but has provided for the making and filing of a tangible thing, against the publication and duplication of which it is the purpose of the statute to protect the composer."[FN69] Thus, this view of the idea-expression dichotomy, or a concept inextricably intertwined with it, denies protection of ideas and protects expression only in the most narrow, physical, perceptible sense.

The narrow holding of the White-Smith case was overturned legislatively by the almost immediate passage of the 1909 Copyright Act,[FN70] which protected musical works against duplication by phonorecords[FN71] and also created a compulsory license to avoid a monopoly in the sound recording industry.[FN72] Additionally, the reasoning of the Supreme Court in White-Smith and similar cases was rejected in 1976 by a definitional structure that made clear that a copyrighted work is not merely the physical, tangible copy, but exists on a more abstract level.[FN73] The point can be illustrated by contrasting [p. 338] the definition of "literary works" (one of the categories in which copyright can exist) with the definition of "copies." Whereas literary works are somewhat abstract "works . . . regardless of the nature of the material objects . . . in which they are embodied, [copies are] material objects . . . in which a work is fixed."[FN74] As explained by the House Report:

The definitions of these terms in section 101, together with their usage in section 102 and throughout the bill, reflect a fundamental distinction between the "original work" which is the product of "authorship" and the multitude of material objects in which it can be embodied. Thus, in the sense of the bill, a "book" is not a work of authorship, but is a particular kind of "copy."[FN75]

Under the approach adopted by the Stowe and White-Smith cases, the idea-expression dichotomy was a true dichotomy. Anything beyond the exact language or physical form in which a work was embodied or expressed was not within the scope of copyright; only exact takings were prohibited. Thus, there were no shades of ambiguity between the idea and its expression.

There were other cases during this early period in which courts held that even a nonliteral copying could constitute copyright infringement.[FN76] These cases implied that to some extent the idea of a work—or at least something more than the literal expression of it—could be the subject of copyright. One of the more broadly worded cases was King Features Syndicate v. Fleischer,[FN77] in which the United [p. 339] States Court of Appeals for the Second Circuit held that the making of a three-dimensional doll embodying the image of a copyrighted cartoon character was an infringement of copyright.[FN78] The court stated that "[t]he appellees did not take all of the copyrighted matter, or all its principal characters, but took one, the idea of the horse 'Sparky.'"[FN79]

The court in Fleischer purported to be protecting ideas. In any event, it certainly adopted a much broader concept of protectable expression than did the Stowe case. The court in Fleischer created a problem, however, that did not exist under the simplistic approach of the Stowe case: if something more than the exact expression is to be protected, where should the line be drawn? The answer is ostensibly given in an opinion written by Judge Learned Hand, a recognized expert in copyright law.

C. The "Abstractions" or "Patterns" Tests

In Nichols v. Universal Pictures Corporation,[FN80] Judge Hand attempted to set the standard by describing an "abstractions" test. After a lengthy recitation of the plots of the two plays involved in the suit, Judge Hand stated:

It is of course essential to any protection of literary property, whether at common-law or under the statute, that the right cannot be limited literally to the text, else a plagiarist would escape by immaterial variations. That has never been the law, but, as soon as literal appropriation ceases to be the test, the whole matter is necessarily at large, so that, as was recently well said by a distinguished judge, the decisions cannot help much in a new case.[FN81]

[p. 340] Judge Hand thus presumably rejected cases like the Stowe case, which applied such a literal test.[FN82] Yet, in recognizing a principle of law which concededly "cannot help much in a new case," Judge Hand opted for almost no rule at all. This may be the price paid for extending protection beyond mere literal copying.

Judge Hand also observed that there are at least some relatively easy cases: for example, if a plagiarist lifts a whole scene or a part of the dialogue, "[t]hen the question is whether the part so taken is 'substantial,' and therefore not a 'fair use' of the copyrighted work."[FN83] [p. 341] When a plagiarist takes not a literal portion, but "an abstract of the whole, decision is more troublesome."[FN84] Judge Hand then announced his "abstractions" test, which is meant to explain the nature of the idea-expression dichotomy:

Upon any work, and especially upon a play, a great number of patterns of increasing generality will fit equally well, as more and more of the incident is left out. The last may perhaps be no more than the most general statement of what the play is about, and at times might consist only of its title; but there is a point in this series of abstractions where they are no longer protected, since otherwise the playwright could prevent the use of his "ideas", to which, apart from their expression, his property is never extended. Nobody has ever been able to fix that boundary, and nobody ever can. In some cases the question has been treated as though it were analogous to lifting a portion out of the copyrighted work; but the analogy is not a good one, because, though the skeleton is a part of the body, it pervades and supports the whole. In such cases we are rather concerned with the line between expression and what is expressed. As respects plays, the controversy chiefly centers upon the characters and sequence of incident, these being the substance.[FN85]

Judge Hand stated that the taking of either plot[FN86] or characters[FN87] might be sufficient to constitute a copyright infringement. He held, [p. 342] however, that in the instant case, both as to incident and character, "the defendant took no more—assuming that it took anything at all—than the law allowed."[FN88]

Judge Learned Hand seemed to say that the concepts of substantial similarity and idea-expression should be applied in different types of infringement cases, and that the tests are distinguishable. Substantial similarity is appropriate for determining whether portions taken directly from the original work are infringing; but this test is inapplicable, however, when the plot outline or generalized character is taken. In such cases, a different test—that of idea-expression—must be used to determine the extent to which the copyrighted work represents an idea or its expression. The difference between these two tests seems to be that the idea of a play or a character can be determined by reference to how well delineated it is in the original: "the less developed . . ., the less they can be copyrighted; that is the penalty an author must bear for marking them too indistinctly."[FN89] The test of substantial similarity works best in the situation in which something has been literally taken, and the observer can weigh the substantiality of the amount taken by comparing the two works.

This distinction simply does not work.[FN90] In fact, the lower court had found relatively easily, without resort to the abstractions test that Judge Hand propounded, that there was no infringement in the Nichols case[FN91]—a holding affirmed by the United States Court of Appeals for the Second Circuit.[FN92] In explaining that undeveloped [p. 343] characters—or, by extension, plots—would not be protected if they were not sufficiently delineated, Judge Hand used as examples Sir Toby Belch and Malvolio from Shakespeare's Twelfth Night.[FN93] In fact, the two Shakespeare characters are quite well delineated, and one might suppose that they would not be considered mere ideas as developed by Shakespeare. The problem with Judge Hand's abstractions test is that it does not look at the characters as developed by Shakespeare, but instead asks us to make successive abstractions of those characters, which on the most abstract level—as the riotous knight or the foppish steward—will not be protected. If the abstraction at issue is not determined by the characters or plot as developed, how does one determine the level of an abstraction upon which to focus? Presumably, the copying work is examined, and determine whether the character or plot, at the level of abstraction at which it is copied, represents idea or expression (or where along the idea-expression continuum it lies). But if the copying work must be examined anyway, how is the test distinguishable from the substantial similarity test, which has as its essence a comparison of the two works? Judge Hand implied that comparison, as used in the substantial similarity test, was not appropriate where a separate portion is not taken; some different test was required. [FN94]

Perhaps the major significance of the abstractions test is not in its ability to work as a tool for deciding cases—indeed, Judge Hand himself admitted that each case must be decided on its own facts, and developed the abstractions test more as a model for guidance in deciding cases than a rigid rule for determining the outcome of the case.[FN95] Rather, the significance of the doctrine rests in its rejection [p. 344] of the Stowe approach to copyright protection.[FN96] If a simplistic approach is adopted, an abstractions test would not be necessary; but "as soon as literal appropriation ceases to be the test," there are no simple answers.[FN97] The Nichols case signaled a willingness by Judge Hand, and the United States Court of Appeals for the Second Circuit, to take that more difficult road. After Nichols, there are hardly any more cases like Stowe, in which copying is an infringement only if it is literal.[FN98]

Although the facts of the Nichols case may not have required the development of a new test, the test that Judge Hand developed did pave the way for later cases which clarified the continued expansion of copyright law to protect against nonliteral copying.[FN99] The most celebrated such case was Sheldon v. Metro-Goldwyn Pictures Corporation,[FN100] in which Judge Learned Hand, speaking for a [p. 345] unanimous court, found copyright infringement where there was no identifiable language that was actually copied.[FN101] The district court had dismissed Sheldon's claim that defendant's movie, based partly on a celebrated murder trial in England, had infringed its copyright in a play based upon the same real-life incident.[FN102] The United States Court of Appeals for the Second Circuit, in reversing, found that there was substantial similarity in the characters and "parallelism of incident."[FN103] The court criticized the defendants (and implicitly, the lower court) for overemphasizing the extent to which the play was based upon public domain materials.[FN104] Judge Hand found that at [p. 346] several key points, what the defendant took from the play was not similar to what was contained in the previous materials but instead was material original with the plaintiff.[FN105]

Another prime example of how the abstractions test came to be used as a means of expanding rather than contracting the universe of works eligible for copyright is the case of Mazer v. Stein.[FN106] In Mazer, the United States Supreme Court recited a classic statement of the idea-expression dichotomy and characterized the Baker case as standing for that dichotomy.[FN107] Because the Court upheld Stein's claim of infringement, however, the Court necessarily viewed the infringing work as having taken the expression and not the mere idea.[FN108] The dichotomy was recited primarily to justify the expansion of copyright to include even utilitarian works: that was, the expansion was possible precisely because, as the idea-expression dichotomy reminds us, the resulting copyright is limited. In this sense, an apparent limitation on the scope of copyright law has been used to accomplish an expansion of the subject matter of copyright.[FN109]

[p. 347] The genius of the abstractions test is that it appears to answer the major criticism of the idea-expression dichotomy: namely, that it is obsolete because it was developed at a time when Stowe was a correct interpretation of the copyright law.[FN110] Because the abstractions test seems to accommodate the very expansion of copyright law that had led to the criticism, it appears less objectionable than the older formulation of the idea-expression dichotomy. Yet, if the result of the abstractions test is to perpetuate the arguably outmoded doctrine, then this test is justified only to the extent that new reasons are given which substantiate the continued need for the doctrine. Judge Learned Hand does not supply these new reasons. The genius of the case masks the mischief that it does.

D. Statutory Development

1. Copyright Office Regulations

Even prior to the Copyright Act of 1976,[FN111] the Copyright Office issued regulations purporting to describe certain material that is not copyrightable.[FN112] These regulations draw upon and expand upon several related doctrines, including the idea-expression dichotomy, [p. 348] the works of utility doctrine, the requirement of originality, and the exclusion of certain categories that are thought to represent de minimis contributions.[FN113] This section focuses upon that portion of the current regulation which purports to recite the idea-expression dichotomy.[FN114] The pertinent regulation concerning registration of claims provides that" [t]he following are examples of works not subject to copyright and applications for registration of such works cannot be entertained: . . . (b) Ideas, plans, methods, systems, or devices, as distinguished from the particular manner in which they are expressed or described in a writing . . . ."[FN115]

It is clear that the Copyright Office is empowered to establish regulations "not inconsistent with law for the administration of the functions and duties made the responsibility of the Register under this title."[FN116] Although entitled to some weight in deciding issues of law, the regulations are not binding on the courts.[FN117] Yet the Register of Copyrights presumably is in the best position to weigh the validity of certain classifications of copyrightable works, and must of necessity make a preliminary determination in deciding whether to register a work.[FN118]

The Copyright Office has explained the regulation in the following terms: "Ideas, Methods, or Systems cannot be copyrighted. Copyright protection is not available for: ideas or procedures for doing, making, or building things; scientific or technical methods or discoveries; business operations or procedures; mathematical principles; formulas; or any other sort of concept, process, or method of operation."[FN119] Interestingly, by its reference to "ideas. . . for doing, [p. 349] making, or building things,"[FN120] and to scientific, technical, business or mathematical methods, the Copyright Office's explanation seems to emphasize that aspect of the idea-expression dichotomy that incorporates the works of the utility doctrine. This aspect of the idea-expression dichotomy does not resemble the abstractions test that was propounded by Judge Learned Hand, but is a more carefully and narrowly drawn description of the idea-expression dichotomy. If it is applied only in the context of works of utility, the doctrine will not be used to foreclose ideas that have been embedded in otherwise expressive works. Abie's Irish Rose, the play that was the subject of the Nichols case (which led to the abstractions test),[FN121] would not have been, and was not, refused registration under the Copyright Office regulations. Since there are clearly expressive aspects of the play, it is eligible for copyright even though in its most abstract outline it is not protected against nonliteral "copying." Although the Copyright Office has not directly repudiated the use of the abstractions test, it certainly has not adopted any regulations that are designed to reflect that test, and the existing regulations appear to allow much less room for the exercise of judgment.[FN122] It may be inferred that the abstractions test is inappropriate at the registration or "subject matter" stage, and that it is more a test of substantial similarity—requiring a copying work for its application.[FN123]

[p. 350] 2. Codification in Section 102(b)

The idea-expression dichotomy was codified for the first time in 1976. Section 102(b) of the Copyright Act of 1976 provides: "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."[FN124]

There is extremely little legislative history on subsection (b), indicating perhaps that it was not viewed as particularly controversial. The House Report states that it was merely a codification of existing law, and was not intended to change the test in any way.[FN125] Since it is contained in section102 (dealing with copyrightable subject matter), rather than being contained in or following section 106 (dealing with the scope of the exclusive rights in copyright), Congress presumably intended to relate the idea-expression dichotomy to the nature of a work, and not simply to the test of substantial similarity.[FN126]

There is some indication, however, that section 102(b) was intended to emphasize the narrow scope of protection in order to allow expansion of copyright to new technologies. [FN127] The language of section [p. 351] 102(b) was added in the 1969 version of the copyright bill.[FN128] The primary relevant testimony that would seem to explain why the section was added deals almost exclusively with the issues raised by inclusion of computer programs within the copyright scheme.[FN129] The section was particularly designed to assure that copyrightable programs would create a monopoly only in the particular implementation of a method, not in an entire method for accomplishing a particular task.[FN130] The basic concern prompting the inclusion of section 102 was one involving a work of utility, a concern to which the dichotomy seems best suited.[FN131] Ultimately, section 102 would allow the expansion of copyright law specifically to cover a new subject matter.

E. Recent Cases

Despite the criticism of commentators,[FN132] the idea-expression dichotomy, which was seemingly approved by Congress in 1976, has also shown a staying power in the case law. The United States Court of Appeals for the Ninth Circuit, in Sid & Marty Krofft Television Productions, Inc. v. McDonald's Corporation,[FN133] concluded that "the distinction accurately conceptualizes the fundamental elements in an artistic creation and balances the competing interests inherent in the copyright law."[FN134] Thus, the necessity of the doctrine rests upon the [p. 352] danger that copyright law would otherwise be extended too far. As the court stated:

Application of [the substantial similarity test] would produce some untenable results. For example, a copyright could be obtained over a cheaply manufactured plaster statue of a nude. Since ownership of a copyright is established, subsequent manufacturers of statues of nudes would face the grave risk of being found to be infringers if their statues were substantially similar and access were shown. The burden of proof on the plaintiff would be minimal, since most statues of nudes would in all probability be substantially similar to the cheaply manufactured plaster one.[FN135]

The Ninth Circuit seemed willing to concede that the limited scope of protection, covering only original elements of expression, theoretically militates against an overextensive application of the copyright laws. Yet, the court perceived that a substantial practical burden is placed upon manufacturers of products: in a copyright suit with a clear prima facie case, the burden would be upon the defendant to prove that the work was his own or to prove a lack of originality in the plaintiff's work. The court was unwilling to impose this burden upon the marketplace.[FN136] The court concluded that a "limiting principle is needed," and that "the classic distinction between an 'idea' and the 'expression' of that idea" provides the needed limitation.[FN137]

The court cited Mazer v. Stein[FN138] as an example of the idea-expression dichotomy,[FN139] and explained the necessity of the doctrine by the example of the nude figures incorporated into a utilitarian article that was involved in that case.[FN140] As later developed more fully in Part V(B)(1)(b), the necessity of a limiting doctrine in the case of a utilitarian work does not require that the doctrine by any [p. 353] more extensive than the works of utility doctrine itself.[FN141] Krofft did not involve a work of utility; it was therefore worth asking if the same limiting doctrine was really necessary in the case that was actually before the court.[FN142] Because the court in Krofft concluded that the copyright law was infringed by a nonliteral taking of the characters involved in the suit,[FN143] the court obviously thought that the expression of the characters itself had been taken, so that subtle distinctions between idea and expression were not necessary to a determination of the case.

Indeed, Krofft represents about the farthest judicial extreme in finding that the nonliteral taking of expression, which arguably borders upon the taking of an idea only, can constitute an infringement of copyright. The court in Krofft, following Arnstein v. Porter,[FN144] found that an intrinsic(or relatively subjective) test of infringement is the relevant test.[FN145] Therefore, even if the works are not substantially similar (which the court did not concede), the test is one of "total concept and feel," which, in the given case, appropriately was found by the jury to be substantially similar.[FN146] Krofft, if it is seen as an idea-expression case, makes clear that protection may be granted even for works which tend toward the idea side of the idea-expression continuum.

The Krofft case can best be categorized as a false idea-expression case, since it focuses analysis upon the substantial similarity test of infringement, or proposes a rule that effectively interrelates the two doctrines.[FN147] The case does not ultimately rest upon a characterization of the fantasy characters in isolation; rather, it necessarily involves a close comparison of the two works, and is dependent upon the impression created in the minds of the intended viewers that the works are substantially similar.

In Durham Industries, Inc. v. Tomy Corporation,[FN148] the United States Court of Appeals for the Second Circuit also purported to find a need to continue using the "imprecise tool" of the idea-expression dichotomy, because of its apparently unique ability to balance societal interests in the use of copyrighted works with the author's—and society's—interests in protecting the work.[FN149] In Durham [p. 354] Industries, however, the court managed to find and apply several other theories that seemed to fulfill the balancing function. First, some of the mechanical toys were ineligible for protection because they were slavish copies of licensed Disney characters and added no original non-utilitarian elements.[FN150] Second, regarding some of the other mechanical toys involved in the case, the court found that the works of utility doctrine was sufficient to bar protection of the mechanical or utilitarian features of the works.[FN151] Finally, the court emphasized the flexibility of the substantial similarity doctrine, which required more than trivial or incidental similarities to be actionable.[FN152] Applying these tests (as well as the elusive idea-expression dichotomy), the court had little difficulty finding that, although pairs of plaintiff's and defendant's toys were "mechanically identical and structurally similar," they were "distinguishable in many respects," or "in every articulable respect."[FN153] Thus, although the case was permeated with idea-expression language, the Second Circuit found other, narrower bases for preserving the desired balance.

[p. 355] F. The Computer Cases

In recent years, computer software copyright cases have led to some of the most interesting applications of the idea-expression dichotomy. The major recent cases have generally held that computer programs are protectable notwithstanding the dichotomy.[FN154] Thus, while these cases seem to approve of the idea-expression dichotomy generally, the dichotomy is not sufficient to bar computer software programs from copyright protection. Furthermore, some of the cases enunciate a standard for the idea-expression dichotomy that is so liberal as practically to nullify its impact. From the early idea-expression cases, in which expression was taken only by a near-literal copying—the presumption being that any other taking was only of idea, and thus not a copyright infringement—the courts seem to have come 180 degrees to the point where almost any taking is of expression, so long as there is at least some other way of expressing the idea. Thus, in the historical development of the idea-expression dichotomy, the courts have moved from a characterization that is strongly shifted toward the idea (which is unprotected) to one that is strongly shifted toward the expression (which is protected).

The recent computer copyright cases can be divided into two distinct periods: the early phase, which dealt primarily with copyrightability of computer programs in various formats (eventually leading to a consensus that programs are protectable notwithstanding the form in which they are fixed); and the more recent cases, which concern primarily the scope of copyright protection.[FN155]

[p.356] 1. The First Phase: Concerns with Copyrightability

A short discussion of the application of the idea-expression dichotomy to computer programs was contained in the final report of the National Commission on New Technological Uses of Copyrighted Works.[FN156] The Commission distinguished between "programs," [p. 357] which were protectable, and "processes," which were not. The Commission observed that the Baker case "is often misconstrued as imposing a limit on the copyrightability of works which express ideas, systems or processes," and instead adopted Professor Nimmer's approach that Baker "in no event justifies the denial of copyrightability to any work."[FN157] Rather, Baker "stands for the proposition that using the system does not infringe the copyright in the description."[FN158]

The Commission separately identified the "idea-expression identity" exception to copyrightability of works.[FN159] This exception provides that "copyrighted language may be copied without infringing when there is but a limited number of ways to express a given idea."[FN160] The Commission concluded, however, that "[t]he availability [p. 358] of alternative non-infringing language is the rule rather than the exception,"[FN161] so that most computer programs would remain protected notwithstanding the presumed exception.

The leading case that clearly established that computer programs were copyrightable was Apple Computer, Inc. v. Franklin Computer Corporation,[FN162] in which the United States Court of Appeals for the Third Circuit held that Apple Computer's operating system programs, in object code as well as source code, which were embedded in ROM (Read Only Memory) chips as well as stored on disk, were "literary works" subject to copyright.[FN163] The court found that Franklin Computer's programs were copied from Apple Computer's, and therefore constituted a copyright infringement.[FN164]

One of the arguments raised by Franklin Computer was that the operating system programs could not be copyrighted because they represented idea and not expression. The court discussed the Baker[FN165] [p. 359] case, as well as Nichols[FN166] and other idea-expression cases;[FN167] the court of appeals concluded that the line between idea and expression "must be a pragmatic one. . . ."[FN168] The court adopted a "suggestion" contained in Dymow v. Bolton[FN169] and focused on "whether the idea is capable of various modes of expression. If other programs can be written or created which perform the same function as an Apple Computer's operating system program, then that program is an expression of the idea and hence copyrightable."[FN170] Curiously, the court equated the typical idea-expression case with one involving the merger of idea and expression: "In essence, this inquiry is no different than that made to determine whether the expression and idea have merged, which has been stated to occur where there are no or few other ways of expressing a particular idea."[FN171]

By jumping to the merger argument, the court, much like the CONTU Report when it focused upon the "idea-expression identity,"[FN172] emphasized not whether the work as copied constituted idea or expression, but shifted the focus to whether the idea of the copyrighted work was capable of expression in other forms.[FN173] The court in Apple Computer then remanded the case to the trial court for a determination whether some or all of Apple Computer's operating system programs represented the only means of expression of the idea underlying them.[FN174]

In discussing how the trial court should apply the test on remand, however, the court in two ways seemed to tilt the determination in favor of Apple Computer. First, it rejected Franklin Computer's major argument—that there were a limited number of ways "to arrange operating systems to enable a computer to run the vast body of Apple-compatible software."[FN175] The court found that the purpose of making a 100% compatible machine was a "commercial and [p. 360] competitive object which does not enter into the somewhat metaphysical issue of whether particular ideas and expressions have merged."[FN176]

Second, the court seemed to tilt toward Apple Computer in characterizing the "idea" of the operating system programs extremely broadly. As its only example, the court suggested that "[t]he idea of one of the operating system programs is . . . how to translate source code into object code. If other methods of expressing that idea are not foreclosed as a practical matter, then there is no merger."[FN177] Such a broad statement of the idea almost assures that there will be other means for achieving the same result. It is no wonder that Franklin Computer settled shortly after the case was decided![FN178]

2. The Second Phase: Scope of Protection

The leading case focusing upon the scope of protection of copyrighted computer programs is Whelan Associates, Inc. v. Jaslow Dental Laboratory, Inc.[FN179] In Whelan, the defendant Rand Jaslow had not literally copied plaintiff's program, but had taken only the "structure, sequence, and organization" of files, screen outputs, and five particularly significant subroutines.[FN180] The United States Court of Appeals for the Third Circuit focused upon the test for substantial similarity and the idea-expression dichotomy; [FN181] according to the Court's analysis, the defendant did copy the structure, sequence, and organization of plaintiff's program, and thus infringed.[FN182]

The court accepted as "axiomatic that copyright does not protect ideas, but only expressions of ideas."[FN183] The court reviewed Baker and concluded that "the line between idea and expression may be drawn with reference to the end sought to be achieved by the work in question."[FN184] The court then described the "idea" of a utilitarian work:

[p. 361] In other words, the purpose or function of a utilitarian work would be the work's idea, and everything that is not necessary to that purpose or function would be part of the expression of the idea. Where there are various means of achieving the desired purpose, then the particular means chosen is not necessary to the purpose; hence, there is expression, not idea.[FN185]

The court then characterized the purpose of the computer program before it as assisting "in the business operations of a dental laboratory."[FN186]

The court of appeals considered the rhetorical question asked by Judge Higginbotham in Synercom Technology, Inc. v. University Computing Company[FN187]—"if sequencing and ordering [are] expression, what separable idea is being expressed?"[FN188]—and concluded: "Our answer has already been given, however: the idea is the efficient organization of a dental laboratory . . . . Because there are a variety of program structures through which that idea can be expressed, the structure is not a necessary incident to that idea."[FN189] It is almost as if the court found that the structure of the program was not the idea, because at least one more level of abstraction existed beyond the program's structure. But that definition of idea—"the efficient organization of a dental laboratory"—is so broad as to immunize from attack under the idea-expression dichotomy almost any more specific expression of that idea. This is certainly a long way from Stowe in the historical march from the expression end of the continuum to the idea end![FN190]

The Whelan case probably was correctly decided. A proper appreciation of the case, however, does not depend so much upon the test that the court in Whelan developed for applying the idea-expression dichotomy, as it does upon the particular facts of the [p. 362] case. The defendant had admitted having access to the plaintiff's program, and had even been under contract with the plaintiff to promote and sell the plaintiff's program.[FN191] Whereas the plaintiff was a professional programmer who had spent a considerable amount of time developing the program to manage the defendant's dental laboratory, the defendant was a relative newcomer to computer programming;[FN192] it was therefore plausible to assume that the defendant would not have been capable of writing an advanced non-infringing program.

Subsequent courts have appeared to split on whether to follow the liberal approach of Whelan. Some courts have found, on an analysis of the programs before them, that the copying program did not infringe because it took only the idea and not the expression of the program.[FN193] For example, in Plains Cotton Cooperative Association v. Goodpasture Computer Service, Inc.,[FN194] the United States Court of Appeals for the Fifth Circuit declined to embrace Whelan, instead deferring to Judge Higginbotham's opinion in the Synercom case.[FN195] The defendants in Plains Cotton had "presented evidence [p. 363] that many of the similarities between the GEMS and Telcot programs are dictated by the externalities of the cotton market. . . . The record supports the inference that market factors play a significant role in determining the sequence and organization of cotton marketing software. . . ."[FN196] The court of appeals declined "to hold that those patterns cannot constitute 'ideas' in a computer context."[FN197] As noted by the editors of the Patent, Trademark & Copyright Journal, the reasoning of the court is not inconsistent with Whelan; rather, the case seems to turn upon a factual finding about the extent of the copying.[FN198]

[p. 364] On the other hand, some district courts have clearly embraced the Whelan reasoning.[FN199] At least one district court case has even extended it. In Broderbund Software, Inc. v. Unison World, Inc.,[FN200] [p. 365] a California district court specifically applied the broad Whelan idea-expression test and rejected the narrow Synercom test by which only literal copying represented infringement.[FN201] After finding another computer program that accomplished a similar result using different screen formats, the court concluded that screen formats represented expression and not idea.[FN202] As in Apple Computer and Whelan, if [p. 366] there were different ways of accomplishing the same result—here, by the existence of one other version—then the screen formats were protectable as expression and not idea.[FN203]

3. Application of the Idea-Expression Dichotomy to Computer Programs

There appears to be some disagreement about the appropriateness of applying traditional copyright principles, particularly the idea-expression dichotomy, to some of the new technologies—especially computer programs. One of the major premises of the Whelan case was that computer programs were subject to general copyright principles such as the idea-expression dichotomy;[FN204] despite some difficulties in application, the dichotomy was no more difficult to apply in the context of computer programs than in other contexts. [FN205]

[p. 367] A much narrower view of the idea-expression dichotomy was advocated in a recent study published by the Office of Technology Assessment[FN206] (OTA Report). Although conceding that the 1976 Copyright Act "treats all information-based products and services the same for purposes of copyrightability,"[FN207] the OTA Report stated that "[t]he central problem of copyright law's continued accommodation to new technologies lies in the indiscriminate application of the doctrine of idea and expression to three fundamentally different categories of works: works of art, works of fact, and works of function."[FN208] The OTA Report argued that works of art have been amenable to an analysis of idea and expression under the abstractions test because the intrinsic value of such works "depends heavily on the particular expression adopted, . . . [and such works] have always had a fixity and completeness to them."[FN209] By contrast, "works of function are seldom valued for their intrinsic or aesthetic qualities,"[FN210] and "electronic technologies permit a new dynamism that makes [p. 368] proprietary boundaries for many works of art indistinct, elusive, and subject to constant change."[FN211]

The difficulties of applying the traditional doctrines to computer programs stem from the "hybrid" nature of such programs. Computer programs, according to the OTA Report, "are hybrid functional works insofar as they employ words and symbols to implement and control a process."[FN212] Thus, with computer software programs

the difficulty is protecting their descriptive aspect—the symbols used in the program—without at the same time protecting their functional aspect—what the symbols do in a computer. Because programs possess both a symbolic and functional nature, copyright may either protect too little if the copyrightable expression is limited to the literal program code, or too much if the copyrightable expression extends beyond the program code.[FN213]

The OTA Report reviewed some of the cases indicating that courts are choosing the broad definition of expression to include "the logic, design, structure, performance or even the output of the computer program."[FN214] The OTA Report concluded by observing:

In theory, none of these rulings is permitted under traditional copyright principles. This is not because the courts have misinterpreted copyright law, but because copyright law cannot be successfully applied to computer programs. Unlike artistic or factual works, which are subject to the "abstractions" test between idea and expression, copyright in computer software is an either/or choice. . . . The "clear distinction" made in Baker v. Selden, "between the book, as such, and the art it is intended to illustrate," collapses in a computer program, because the program embodies both "the book" and "the art."[FN215]

The idea-expression dichotomy is viewed logically as a limitation on the copyrightability of works.[FN216] If one did away with the dichotomy, then one might expect that more, not fewer, works would be amenable to protection under copyright law.[FN217] How is it, then, that [p. 369] the OTA Report, after criticizing the applicability of the idea-expression dichotomy to computer programs, seemed to conclude that the protection of computer programs is thereby made less appropriate?[FN218] The Report stated, after all, that it is "wrong" to include "the logic, design, structure, performance or even the output" of a program within the protection of the copyright law.[FN219]

Apparently, the OTA Report viewed the abstractions test of the idea-expression dichotomy not as a limitation on the protectability of works, but as an accommodative principle that allows the copyright law to be expanded to new fields.[FN220] The abstractions test, according to the OTA Report, does not accommodate very well in the case of functional works.[FN221] Therefore, there is not enough flexibility remaining in the clear distinction test of Baker v. Selden,[FN222] and the expansive protection road that Judge Hand was willing to follow for other works where the accommodative principle was available[FN223] should not be followed for computer programs.

Yet, what kind of accommodative principle is the abstractions test if it does not help to resolve the difficult cases as well as the straightforward ones? One might have expected that in the face of new technologies, the idea-expression dichotomy and the abstractions test would be embraced, not rejected, as an accommodative principle.[FN224] Perhaps the OTA Report's emphasis upon the role of the abstractions test in resolving the idea-expression dichotomy for works of art overstates the usefulness of that principle even in that context.[FN225] If the traditional principles do not work in the case of [p. 370] computer programs, perhaps it is not because computer programs raise truly unique issues but rather because the principles never worked well in any of the contexts in which they were proposed.

In questioning the applicability of the abstractions test to computer programs, and in drawing the distinction between the application of the idea-expression dichotomy for works of art and works of function, the OTA Report seems to shift its emphasis away from the idea-expression dichotomy in isolation, and toward the works of utility doctrine.[FN226] In another context, the OTA Report considered the possibly accommodating application of the "fair use" doctrine in resolving some of the tough computer copyright issues.[FN227] If further accommodation of competing interests is needed, then other, perhaps legislative accommodations, may be needed.[FN228] For example, copyright protection in computer programs could be granted for only a very short time—say, five years. Or protection might be granted while imposing a compulsory licensing scheme to allow for commercial uses of such copyrighted works.

In any event, as conceded by the OTA Report, there is a whole range of doctrines and legislative responses that might help to resolve the difficult issues raised by the copyrightability of computer programs. The idea-expression dichotomy was never a good tool for deciding these issues, and any suggestion that this doctrine barred computer programs from being copyrighted "is now dormant."[FN229] If such a conclusion is reached for computer programs, then perhaps a similar verdict should be rendered for the idea-expression doctrine generally.

III. Analysis and Critique of the Idea-Expression Dichotomy

Earlier critiques of the idea-expression dichotomy have focused upon the lack of an adequate definition of what constitutes an idea, [p. 371] and have suggested that different judges have in mind different concepts in applying the doctrine;[FN230] or have argued that the doctrine simply does not reflect the marketplace that in fact pays handsomely for ideas.[FN231] These other critiques provide a full development of the definitional and market problems.

A. Excluding Categories of Works

The problem with the idea-expression dichotomy goes beyond the typical difficulties in defining the terms. To explain this criticism, it is first necessary to review some basics of copyright. As developed in subsection (1) below, it is quite clear that copyright protects against exact copying, as well as against other particular uses of works listed in the exclusive rights of copyright. The idea-expression dichotomy was developed to apply to, and indeed only makes sense when applied to, persons who infringe by making nonliteral copies of copyrighted works. Yet, the dichotomy presumably forecloses certain works or certain categories of work from federal copyright law entirely—excluding them not only from protection against nonliteral copying, but also from protection against exact copying. Subsection (2) presents a scheme that would grant federal protection for most types of works, although the scope of protection might vary depending upon the category of work involved. Almost all works otherwise eligible for federal copyright, including some works that might be described as "ideas," should be protected against exact copying. What is appropriate in the context of the idea-expression dichotomy, is also appropriate as a general premise: that is, a decrease in emphasis upon rules of law (such as that of idea-expression) that purport to foreclose entire categories of works from copyright protection; and an increase in emphasis upon rules (such as that of substantial similarity) that focus both upon the facts of particular cases, and upon a comparison of two particular works.[FN232]

[p. 372] 1. Protection Against Exact Copying

The federal copyright law[FN233] provides five exclusive rights in a copyrighted work: the rights (1) to reproduce the work, (2) to make derivative works of it, (3) to distribute it, (4) to perform it publicly, or (5) to display it publicly.[FN234] The first right, to "reproduce the copyrighted work in copies or phonorecords,"[FN235] clearly covers the "pirating" or exact duplication of copyrighted works (for example, by technological means such as photocopying, tape recording, or video recording[FN236]). Although the term "reproduce" is not defined in the act, the term "copies" is. Copies are "material objects . . . from which the work can be perceived, reproduced, or otherwise communicated, either directly or with the aid of a machine or device."[FN237] Thus, what we might call "exact copying" of copyrighted [p. 373] works is clearly prohibited by the act, unless the case is covered by an exemption or limitation in the other sections of the act.[FN238]

[p. 374] Additionally, it is an infringement of copyright for another person to make a nonliteral copy of a protected work, such as by incorporating it into a new and independent work using a portion that is "substantially similar"[FN239] to the copyrighted work.[FN240] As copyright expanded to cover nonliteral infringement,[FN241] major principles of copyright law were required to determine which nonliteral copies should constitute infringements and which should not. For example, the doctrine of substantial similarity, which is used to determine if a person has "copied" a work, really has application only in the case of a nonliteral copy: if someone copies a work exactly, then we need not look to the substantiality of the copying, since it is identical.[FN242] One of the subtests of substantial similarity—proof of access—is hardly likely to be a problem if the copying is so exact as almost to preclude the possibility of independent creation.[FN243] The [p. 375] doctrine of fair use might arguably have more applicability in the case of nonliteral copying than in the case of exact copying, in which the contribution to society is arguably less.[FN244] The idea-expression dichotomy, while relevant in weighing the infringing nature of a nonliteral copy, would not be particularly helpful in the case of an exact copy—by definition an exact copy is copied literally, and thus clearly includes expression as well as idea.[FN245]

In short, many of the traditional tests of copyright infringement—substantial similarity, access, idea-expression dichotomy, and fair use—seem to have been developed precisely to handle the case of the nonliteral copier. The case against the person making an exact copy on most of these principles is almost a per se case, as compared to that of the person making the nonliteral copy, for whom a rule of reason—and a case by case determination of liability—may be more appropriate.

2. Proposed Federal Scheme of Protection

There may be nothing wrong in principle with excluding a category of works from protection under the federal act. Yet, if a work is excluded from federal copyright based upon its general character—as lacking sufficient originality, not being fixed in a tangible medium [p. 376] of expression, or, for present purposes, because it constitutes an unprotectable idea—then the work presumably is ineligible for federal protection against exact copying as well as against nonliteral copying. To the extent that the exclusion is based upon a rationale (such as the idea-expression dichotomy) that is directed primarily to the problems of nonliteral copying, it is unfortunate that protection against exact copying is sacrificed as well. The problem is not just a recent one,[FN246] although it is particularly exacerbated now that the technology for exact copying—for example, audio and video tape recorders, photocopying machines, and computers—has come within the price range of the typical consumer.

There are three possible responses to such a categorical approach to exclusion of works from the federal copyright law. First, an observer might conclude that lack of protection is tolerable or desirable, and that the preemptive effect of the federal law logically should preclude the states from extending protection to fill the gap. Second, one might conclude that the federal copyright law does not prohibit the states from granting protection against "misappropriation" of a work that is otherwise ineligible for federal protection. Third, one might try to construe the federal copyright act (or amend it if necessary) to include the category of work involved, but limit the scope of protection if necessary to cover only (or at least) exact copying. Among these approaches, I recommend the third, for the following reasons.

The first approach—disqualification from all protection—seems inappropriate because it withdraws protection on too broad a scale: the penalty is too great when based upon a principle that seems addressed primarily to the case of the nonliteral copier. The potential harm that results from exact copying, which has the greatest likelihood of displacing an existing market for a work, is much greater than the potential harm from nonliteral copying. Exact copying would seem to be the type of activity least deserving of any first amendment or "marketplace of idea" claims,[FN247] since it contributes nothing to the stock of ideas in the marketplace, but only appropriates the profit or advantage from one party in favor of another. Although there may be legitimate fair-use contexts that would protect even exact copying,[FN248] such contexts require an examination of all relevant [p. 377] fair use factors—the "nature of the work" as idea should not alone be determinative.

The second approach, which would preclude federal relief of ideas, but allow the states to grant limited protection under a misappropriation theory, has a surface appeal. Such an approach, however, also has serious conceptual drawbacks. To understand the problem, it is necessary to review the cases that attempted to preserve state record piracy laws against preemption for sound recordings made prior to adoption of the federal Sound Recording Amendment of 1971.[FN249]

Some of the lower courts, in refusing to follow the broad preemptive reach of the Supreme Court's Sears-Compco cases[FN250] (primarily in the context of state record piracy laws), raised arguments in favor of state misappropriation laws and against federal preemption of the field.[FN251] In particular, some lower courts held that the Sears-Compco cases prevented the states from passing laws involving the copying of copyrighted works, but not their "misappropriation" of such works.[FN252] In this sense, misappropriation meant the same [p. 378] thing as the "exact" copying discussed previously—exact duplication without the independent creation of a new work. Unfortunately, these cases represented a fundamental misunderstanding of federal copyright law.[FN253] They implied, curiously, that federal copyright law protected against nonliteral copying, but not against misappropriation, or exact copying: if federal copyright covered both types of copying, there would be no basis for distinguishing between them for preemption purposes. Yet, the protection under federal copyright law for those works that are within the scope of federal copyright law protects as clearly against exact copying as against nonliteral copying.[FN254]

The approach illustrated by these record piracy cases, although typical of judicial attempts to expand state coverage where federal protection seems lacking,[FN255] presumably would not work in the case of the idea-expression dichotomy. Because ideas are specifically dealt with in the federal copyright law by way of exclusion in section 102(b), they are not, arguably, an appropriate subject for state protection based upon an "omitted category" approach under section [p. 379] 301(b)(1).[FN256] The states would have to base protection upon the other prong of section 301, which provides that preemption does not apply to "activities violating legal or equitable rights that are not equivalent to any of the exclusive rights within the general scope of copyright as specified by section 106."[FN257] Indeed, states traditionally have protected ideas, in appropriate situations, under theories of contract, implied contract, trade secret, or related doctrines.[FN258]

One of the problems with judicial attempts to fill the gaps in federal protection is that one of the purposes of a unified federal copyright law—to replace the old "dual system" of common law and federal statutory copyright[FN259]—is strained, if not defeated. The strain is eliminated if we adopt the third approach suggested above—construction of the federal act to include ideas, but limited, if necessary, in the scope of protection granted. Under this approach, at least a minimal role—protection against exact copying—can be played by the federal copyright law even for categories of works, such as ideas, that for one reason or another are excluded from the broadest federal protection.[FN260] The federal copyright law should be structured or interpreted to include the broadest range of categories of works, including works that may appear to be ideas;[FN261] if limitations in rights are appropriate,[FN262] they should be reflected in the federal act as limitations upon the scope of protection.

[p. 380] This approach parallels other suggestions and developments that recently have taken place in the field of copyright and related fields of intellectual property protection. Professor Gorman proposed, in the context of recognizing copyright in fact works,[FN263] that the scope of copyright might be "thin," but that "it ought not . . . be treated as nonexistent."[FN264] He concluded that "[t]he better approach [for legal and business forms], as with maps and directories, would be to grant copyright for forms in the generality of cases, so as to provide an incentive for their creation and to permit a finding of infringement for some of the more 'piratical' takings."[FN265]

In the context of character protection, Professor Kurtz made a similar suggestion.[FN266] She concluded:

By focusing on the copyrightability of a character, courts have blurred the distinction between the concepts of infringement and copyrightability. This focus shifts the court's attention from comparing the characters in the allegedly infringed and infringing work, to analyzing whether a character is sufficiently distinctive or well-developed to deserve protection. This requires judges to become literary and artistic critics, a dangerous undertaking for those whose education and experience are in the law. . . .
. . . .
Much of the confusion and lack of clarity in the standards used to protect fictional characters under copyright can be avoided by recognizing that the "copyrightability" of a fictional character, whether literary, pictorial or audio-visual, is not the issue.[FN267]

In trademark law, there has been a slight trend away from foreclosure of federal protection for package designs, distinctive [p. 381] names for ingredients, or location of labels on garments, and toward an analysis of the particular facts of each case, examining whether the asserted basis for trademark protection has in fact achieved secondary meaning.[FN268] A similar trend can perhaps be seen in the preemption field in the movement from the Sears-Compco[FN269] cases (which in 1964 foreclosed state protection of works that were ineligible for federal patent or copyright protection) to Goldstein v. California[FN270] (which allowed for state protection of categories left unattended by federal law[FN271]). Although the result was a categorical approach to the scope of federal law, it nonetheless represented a liberalizing trend away from a blanket foreclosure of state relief and toward a more ad hoc approach.

In the context of computer programs, the proposal parallels to some extent a development that has occurred in the patentability of computer programs. Early (for computers) Supreme Court cases had held that computer programs were not patentable as such.[FN272] The cases raised questions whether computer programs could ever be a proper subject for patents. The questions were resolved in the United States Supreme Court case of Diamond v. Diehr,[FN273] in which the Court held that a process for molding raw, uncured synthetic rubber into cured products was not ineligible for patent protection merely because a computer program was involved as a key step in the invention. As contemplated in Diehr, the issue in computer patent cases is not whether the process including a computer calculation can be patented, but whether the other tests of patentability and infringement have been met.[FN274]

Extending upon the case of Diamond v. Chakrabarty,[FN275] the Patent Office recently has been granting patents even in computer programs that have no "postsolution activity" as contemplated in Diehr.[FN276] Chakrabarty had allowed patent protection for a human-made micro-organism on the premise that "the patentee has produced [p. 382] a new bacterium with markedly different characteristics from any found in nature and one having the potential for significant utility. His discovery is not nature's handiwork, but his own; accordingly it is patentable subject matter. . . ."[FN277]

By similar reasoning, the Patent Office has granted patents for several complex mathematical equations that require the number-crunching abilities of a computer.[FN278] For example, a patent was granted to AT&T in 1988 for a "pioneering algorithm used by large organizations to decide the best way to allocate resources."[FN279]

Thus, the expansion of copyright law to cover computer programs, and perhaps even the "algorithms" embodied in them, corresponds to a similar development that has occurred in the field of patents, and raises the same basic question: whether the traditional prohibition of copyright or patent protection for "ideas" is really necessary, or ever was.

B. Merger Theory

Although there has been considerable difficulty in defining and applying the idea-expression dichotomy in general, there has been even more than the usual difficulty in applying it within a particular setting—where idea and expression are said to merge, or become a "unity." In the language of Baker, "Where the art [that the book] teaches cannot be used without employing the methods and diagrams used to illustrate the book, or such as are similar to them, such methods and diagrams are to be considered as necessary incidents to the art, and given therewith to the public."[FN280]

Whereas the focus of the idea-expression dichotomy is upon whether the work constitutes idea or expression, the merger doctrine focuses upon whether t