The text below represents commentary, not legal advice.

Return to: Critical Copyright, and more.

Return to: Paul Edward Geller, attorney.

Published in the Revue Internationale du Droit d'Auteur (RIDA), no. 159 (Jan. 1994), p. 3, in the following English text, as well as in French and Spanish translations. Copyright © Paul Edward Geller 1994. This article is based on a longer essay, "Must Copyright Be For Ever Caught Between Marketplace and Authorship Norms?," which was published in Of Authors and Origins: Essays on Copyright Law (Oxford: Clarendon Press, 1994), p. 159. I thank the editors of this collection of interdisciplinary essays, Brad Sherman and Alain Strowel, as well as Arthur Bierman, Monique Blanvillain, Lorin Brennan, François Dessemontet, Ysolde Gendreau, Paul Goldstein, André Kerever, Jerome Reichman, Mark Rose, and Robert Rotstein, for their critical comments on prior drafts of this study. I am also grateful to the Max Planck Institute for Foreign and International Patent, Copyright, and Competition Law, in Munich, for the opportunity to undertake the basic research for this study there. I am responsible for translating all non-English texts which are not cited below in translation.

Toward an Overriding Norm in Copyright:
Sign Wealth

Paul Edward Geller

INTRODUCTION

        The term norms will here mean principles that lead to rules governing concrete cases. To adopt the suggestive words of the drafters of the French Civil Code, norms are "principles fertile in consequences" for lawmakers.(1)

        Distinct norms are at work in copyright law. On the one hand, marketplace norms require rules to maintain a reliable market in products of the mind. On the other, authorship norms dictate rules to empower authors to control the use of their self-expression by others. Different bodies of copyright law vary in their compliance with these distinct norms. Anglo-American copyright laws, in the usual course of affairs, rely on marketplace norms. European laws of author's rights, if faced with the choice, respect authorship norms.(2)

        Do we always have to choose between market forces and authors' interests? This alternative is often purely theoretical, since in most cases marketplace and authorship norms lead to much the same practical results. Nonetheless, in certain cases, here called difficult cases, each norm could separately lead to results inconsistent with those favored by the other. For example, while one norm directs the marketplace to allow all rights to be freely alienated by contract, the other vests authors with rights that sometimes resist such alienation. By overriding norms I mean those which help to resolve just such tensions, which inevitably arise as any body of law develops.(3)

        Anticipate, for the moment, the aims of marketplace and authorship norms. If reliable, markets in products of the mind would increase the wealth of socially available information. If empowered to control the use of their self-expression by others, authors would assure that such expression continues to signify their own insights even as the media carry it across society. An overriding norm, aiming at both informational wealth and significant expression, ought to help generate what I here propose to call sign wealth.

        In part I below, I will analyze premises that have typically distinguished marketplace and authorship norms. In part II, I will briefly critique these premises and attempt to restructure them in terms of an overriding sign-wealth norm. In part III, I will test this new norm by asking whether it might help to resolve tensions that arise in difficult copyright cases.

I. DISTINCT NORMS

        Values push and pull on lawmakers as they try to move from norms to rules. Lawmakers also take into account often-tacit models or exemplars of the factual conditions that the law has to influence. In subpart 1 below, I will consider values that have motivated copyright law from the start and, in subparts 2 and 3, specify certain shifts in these values, as well as in relevant models and exemplars.(4) In subpart 4, I will indicate how marketplace and authorship norms, diverging as a result of these shifts, lead to inconsistent results in difficult cases.

1. Enlightenment Values

        Copyright only became possible as certain values gathered force in modern times. Some five centuries ago, Europe began to be swamped with new information: Europeans discovered lost texts and new continents and started to use increasingly powerful research tools. Since that time, the printing press has precisely and rapidly reproduced such information, and the book trade has spread it along with nascent national cultures.(5)

       In Europe of the fifteenth century, a feudal order continued to dominate. The church and universities served as censors; as in other trades, guilds held sway in the book trade.(6) Starting in the sixteenth century, however, a mercantilist model of society began to take hold: sovereigns sought to build nation-states that they could manipulate like giant marionettes. The English king delegated police powers to the London association of printers and publishers while confirming its trade monopoly. The French royal authorities censored works while granting monopolies, called privileges, to publish books or stage plays.(7)

        In the eighteenth century, Enlightenment values crystallized. Writers freed themselves from patronage by turning to publishers to exploit their works.(8) At the same time, increasingly literate segments of the public demanded access to books free of state censorship and monopolies. Generally, state measures favoring specific works or media enterprises, at the expense of others, no longer fit within an emerging liberal framework of law. To maximize both intellectual and material wealth, the rising middle class thought it best to institute law that would leave individuals free to think and express themselves, to work and trade, as they saw fit.(9)

        In 1710, the first copyright law, the British Statute of Anne, was enacted. It already encapsulated Enlightenment values in stating its purpose: the "Encouragement of Learning".(10) In time, copyright laws recognized individual authors' rights to control not just the reproduction, but the public performance of their works. Authors could, following these laws, contractually convey such rights to media enterprises catering to public taste. Decisions to make, disseminate, and enjoy works were, as a result, allocated through open markets. In tandem with a free press, copyright thus helped to decentralize control over the media.(11)

        Whatever values they shared at the start, copyright laws began to diverge at the end of the eighteenth century. Enlightenment values, along with other copyright premises, have shifted in emphasis over time, and these shifts have varied from culture to culture. The following schematic analysis can at best only dramatize these shifts by emphasizing some of their aspects while skipping over many others.(12)

2. Marketplace Premises

        I just defined a marketplace norm as requiring rules to maintain a reliable market in products of the mind. In the eighteenth century, while developing marketplace theory in general, Adam Smith compared according an author a "monopoly ... of a new book" to granting a monopoly to a company "to establish a new trade with some remote and barbarous nation."(13) That is, copyrights were needed to encourage the making and marketing of works, since publishing ventures, like colonial expeditions to new lands, could vary from disastrous to profitable. This reasoning also looked to the Enlightenment goal of accelerating the progress of the human mind: as more works were disseminated, they would make more data and ideas more widely known.(14)

        This marketplace theory started from a somewhat mechanical model of creativity. The Renaissance set the ideological stage for copyright by cultivating the notion of the individually creative subject.(15) The Enlightenment, from the start, was dominated by Locke's model of the individual subject, according to which the mind did nothing more than rework experience piecemeal. This subject, at birth, was compared to "white paper, void of all characters"; from that point, its mind would fill up incrementally in receiving simple "ideas" of sensation and combining them into new, more complex "ideas".(16) Within this perspective, it was possible to envisage rationally allocating time and energy to experiment with different combinations of ideas and to communicate the information gained as a result. In particular, Diderot's and d'Alembert's Encyclopedia exemplified an optimally informational work produced by such incrementally creative activity.(17)

        Locke also laid the foundations for a general theory of marketplace norms. He imagined an uneven initial distribution of both surplus and scarce resources. His initial vision was virgin land ready to be colonized; in his words, "in the beginning all the world was America ...".(18) Locke proceeded to ask what type of social order would provide incentives and mechanisms for individuals to allocate scarce resources so that more goods and services would reach larger marketplaces. He responded with a view of law that the Enlightenment would adapt for its own purposes: through a "compact and agreement", individuals "settled the property which labour and industry began" by instituting rights that could be reallocated by contract.(19) That is, labor, reworking raw materials, gave rise to property that the law could confirm as a framework for increasingly specialized goods and services and increasingly frequent market exchanges. Locke, however, did not take the next step of making the argument that copyright would be essential to reliable markets in works of the mind.(20)

        The Enlightenment made this step toward copyright following an analogy too clear and simple for it to resist. Physical labor drew material wealth from virgin land; intellectual labor produced mental constructs starting from the tabula rasa of the mind at birth. Diderot, for example, made just this connection: as property rights prompted farmers to cultivate the land, copyright would induce authors to develop works of the mind.(21) At the same time, the model of the creative subject began to be integrated into another broader model, that of the marketplace conceived as a communication network: the market, allocating scarce resources such as the media, would also convey works to users. As a result, the argument in favor of incentives for authors began to slip into a plea that defended copyrights as incentives to encourage enterprises to invest in the media.(22)

        Marketplace norms then turn on the following premises: first, the Enlightenment values of maximizing the information available to society; further, the model of the subject incrementally creating works by assembling bits of experience; and, finally, the larger model of the marketplace as a communication network. These premises have come together in the argument that copyrights would furnish incentives for authors and media enterprises to make and market works that would compete with each other. Such rights would accordingly operate as "restrictions in competition in order to promote competition".(23)

3. Authorship Premises

        I started by defining an authorship norm as dictating rules to empower authors to control the use of their self-expression by others. Authorship theory began to take shape in the eighteenth century, as literary critics set about distinguishing ordinary authors from creators displaying genius. To quote Addison, most authors "formed themselves by rules", in contrast to "genius" that created "by the mere strength of natural parts", by virtue of "something wild and extravagant".(24) If copyright did no more than prompt the accumulation and dissemination of information incrementally brought together in works, it might not address genius as a potentially autonomous source of mental breakthroughs. Even Enlightenment thinkers, anticipating Romanticism, increasingly recognized that genius accelerated the progress of the mind thanks to its unexpected, but fruitful insights.(25)

        Authorship theory thus presupposed powerful, but rather imponderable processes as feeding creativity. Indeed, European doctrine could not have drawn its most typical consequences in the field of copyright without relying on some notion of an individual subject who would create thanks to novel insights.(26) Moving beyond the attempts of other Enlightenment thinkers to build a mechanical model of mental processes, Kant instead posited a subject whose thought was spontaneous rather than causally determined.(27) He supported this position by invoking "genius" as an exemplar of subjects capable of reaching mental breakthroughs: genius was not subject to predictable "rules", but rather developed its own rules to be imposed on experience, most notably in creating "original" works.(28) Romantic critics went on to surmise that, by virtue of the creative act, such authors would necessarily leave the marks of their unique personalities on their works. In particular, they found in Shakespeare a most telling example of such breakthrough creation.(29)

        Kant also formulated the first general theory of authorship norms. He began by observing that authors expressed themselves in what he called their "discourse".(30) His ideal was that of autonomous self-expression: he found this autonomy best illustrated when learned authors, in contrast to employees or functionaries bound by their duties and positions, freely communicated among themselves.(31) For Kant, as for Enlightenment thinkers in general, the marketplace continued to serve as the model of relevant communication networks: authors conveyed texts to publishers who, investing in publication, needed protection against pirate editions. According to Kant, to guarantee the autonomy of their self-expression, authors ought to be accorded exclusive rights to authorize whether, and how, their discourse was to be recommunicated to the public.(32) Vested with such rights, authors could delegate enough authority to their publishers to allow the latter to prevent third parties from republishing their texts without consent. Kant, however, only extended this argument for copyright to authors' discourse, that is, only to literary works but not to visual objets d'art.(33)

        Furthermore, Enlightenment debates on copyright raised a troubling question. Could the copyright in a work undercut the autonomy of expression itself in some cases?(34) If copyright covered ideas, some authors could assert it to preclude others from expressing ideas that they had already injected into the communication network in the form of their works. To escape this dilemma, certain notions had to be refined: thus Fichte distinguished between the "thought" and the "form" of a work, that is, in more current terms, between freely usable "ideas" and protected "expression".(35) Given this distinction, the Romantic critics could save the argument based on autonomy by positing that each creative subject's self-expression had to take on its own personal, unique form. Copyright in only such "original" expression of one subject would not preclude others from expressing themselves, each in their own way.(36)

        Authorship norms then took shape as the following three premises began to shift. First, Enlightenment values turned from collectively seeking information to stressing individual insight. Further, rather than holding on to models of incrementally creative subjects, Romantic critics invoked exemplars of breakthrough creativity in genius. Finally, the marketplace remained the overall model of relevant communication networks, but on the condition that subjects autonomously communicate works within such networks. According to the argument which ensued, in assuring such autonomy, copyright would permit authors to convey their self-expression to the public intact. Each creator could accordingly give of himself "in his work, which represents the best of himself" to humanity at large.(37)

4. Resulting Tensions

        Marketplace and authorship norms have just been contrasted. The differences between them in theory, however, are greater than they are in practice. Only in certain difficult cases do these distinct norms lead to different results, generating tensions in copyright legislation as well as case law.(38) I will now very quickly outline some cases where these tensions are felt.

        While a marketplace norm only allows for fashioning copyright narrowly, an authorship norm gives it a broader scope. On the one hand, marketplace norms do not authorize legislating rights stronger than necessary for inducing the making and marketing of works. The law of the United States enumerates a closed bundle of rights, further limited by the open-ended exception of fair use which, for example, the U.S. Supreme Court invoked in excusing certain cases of home copying.(39) On the other hand, authorship norms justify rights broad enough to make authors the masters of their self-expression, however this expression might be eventually used. The French and German laws conceptualize authors' rights in broad and flexible terms and limit them in restrictively construed, specific exceptions. The German Constitutional Court even faulted legislative exceptions as unfaithful to this approach because they were not narrow enough.(40)

        Each of these norms implies a different relation of priority between economic and moral rights. To maintain a reliable market in works, a marketplace norm avoids burdening the contractual transfer of economic rights. Anglo-American laws tend to codify previously inchoate moral rights in terms that permit authors contractually to waive invoking these rights against transferees.(41) By contrast, to empower authors to control the use of their works, an authorship norm leads to recognizing inalienable moral rights that authors may assert in the face of contracts to contrary effect. Consequently, French and German copyright laws both formulate such moral rights in broad terms that enable them to survive contractual transfers.(42) Suppose that an author transfers the economic right to adapt a work but later claims the moral right to stop adaptations that distort the work. In such cases, unlike a marketplace norm, an authorship norm might well lead courts to enforcing just such claims.(43)

        These distinct norms result in recognizing different persons or entities as holding copyright, that is, to use European terms, different subjects of rights. A marketplace norm permits legislators to allocate economic rights to those subjects they find most capable of having works created or put on the market. In the United States, the initial owners of rights are authors, often natural persons that independently create works, but also enterprises that employ creators and direct their work.(44) By contrast, an authorship norm permits legislators to attribute moral rights only to flesh-and-blood creators, they alone being capable of such self-expression as these rights are to protect. French law follows both approaches: it exceptionally accords economic rights to those legal entities responsible for making and marketing collective works, but it treats moral rights as "inalienable" since they are "attached" to the very "person" of the author.(45) German law follows only the latter approach: it vests copyright only in natural persons who actually create works, and these creators may not fully alienate economic rights, especially those concerning unknown media, nor moral rights.(46)

        Marketplace and authorship norms place different emphases on the criteria defining what copyright protects, that is, again in European terms, the object of this right. Since marketplace norms favor inducing investment, they allow some trial courts, above all those still steeped in the British legal tradition, to recognize copyright in objects that display such indicia of investment as "skill and labor". American law makes the standard of protection slightly more stringent: it covers only works produced with "some minimum level of creativity", but without requiring any manifestly "personal" input.(47) Authorship norms only concern authors' self-expression, and European doctrine often contemplates protecting works only if they bear some "imprint of personality".(48)

II. RESTRUCTURING COPYRIGHT THEORY

        The confines of this brief study do not allow for any proper critique of all the premises of marketplace and authorship norms. Even if a thorough critique were to put these norms radically into question, it would be foolhardy to abandon them and venture out, without further conceptual guidelines, in search of a wholly new norm for copyright. I will instead attempt to revise questionable premises of these norms in terms of an overriding norm, which hopefully should lead to more consistent results in difficult cases.(49) In subpart 1 below, I will offer some critical remarks to indicate how certain premises of marketplace and authorship norms, respectively, are inappropriate to many cases. In subpart 2, I will try to recast copyright values, so that they might better motivate a more comprehensive, overriding norm. In subparts 3 and 4, I will ask how to rethink our notions of creation and communication consistently with this new norm.

1. Some Critical Remarks

        The arguments for marketplace and authorship norms are at points one-sided. Neither one norm nor the other convincingly applies all across the full and empirically rich spectrum of works of the mind. Toward one end of this spectrum, there are incrementally created works, such as directories or technical drawings, indeed all kinds of banal writings, tunes, and pictures. Toward the other end, we recognize works that represent creative breakthroughs relative to inherited culture, although they are not necessarily fully novel in every detail. It is crucial to note that the types of works clustering toward one end or the other of this spectrum do not correspond neatly to the diverse ways in which authors participate in creation.(50) Some authors work alone, other authors collaborate closely in pairs or small groups, while still others work in large-scale industries. A solitary author, writing a formulaic novel or making an academic painting, might only incrementally vary a given genre. An industrially produced, capital-intensive work, like D.W. Griffith's film Intolerance, might represent a breakthrough.

        According to the theory of marketplace norms, copyright provides financial incentives for making works. To the extent they expect more profits on the market, authors and enterprises are supposed to invest more energy in incrementally putting works together. Nonetheless, in the last century, the British historian Macauley questioned the assumption that such close economic calculations always prompted creative acts. He asked, for example, whether extending the copyright term, and thus increasing the chances of capitalizing on any work, would have driven an author such as Doctor Johnson to create more works. "Would it have induced him to give us one more allegory, one more life of a poet, one more imitation of Juvenal? I firmly believe not. I firmly believe that... he would very much rather have had twopence to buy a plate of shin of beef at a cook's shop underground."(51) In effect, market prospects secured by economic rights seem to play hardly any practical role in the creation of many works: Emily Dickinson wrote her poems without daring to publish them, and Van Gogh painted his last pictures as he slipped into madness. Despite such cases, some Anglo-American commentators do attempt to generalize the argument for economic rights by collapsing the distinction between authors and enterprises. An often-cited study begins: "To simplify the analysis, we... use the term 'author' (or 'creator') to mean both author and publisher."(52)

        According to the theory of authorship norms, copyright is justified by assuring the autonomy of personal expression. It seems to be a matter of common sense that every author elaborates works in some inner mind's eye before expressing them. At the turn of the century, Kohler reformulated this notion in terms of some global creative vision (Weltschöpfungsidee) that must control the imaginary configuration (imaginäres Bild) of any breakthrough work.(53) European copyright doctrine has since tended to conclude that the author's personality, what it effectively posits as the motive force guiding any global creative vision, leaves its imprint on the expression of any resulting work. From here it is a short step to invoking some equation between author and work: for example, an Italian court explains that moral rights "protect the identity of the author, which is objectified in the work, that is, the external embodiment of this subject as it merges with the creative result...".(54) In practice following such an approach, authors are presumed to be damaged when their works undergo uses contrary to their intentions, much like the subject represented by a voodoo doll must suffer when that object is manipulated with ill intent. Or so goes the argument on which many European commentators rely to support moral rights: authors need them to control the uses of their works, in which their personalities are so intimately wrapped up. However, authors all too often create works without manifest personal involvement, such as the Dadaists' automatic writing, Duchamp's ready-mades, or Warhol's soup cans. T.S. Eliot had no difficulty admitting that a poem "is not the expression of personality, but an escape from personality."(55)

        Both marketplace and authorship norms assume the marketplace as a model of relevant communication networks. Traditionally, in the marketplace, authors convey rights to enterprises that in turn invest in media considered as scarce resources. Starting in the last century, publication has become progressively cheaper on a copy-by-copy basis as printing technologies have improved.(56) In the future, the cost of any given transmission of a work through a large-scale telecommunication network will go down as the investments made in developing the network are amortized. In addition, technology increasingly puts into the hands of private parties, not only copying and telecommunication equipment, but computers that allow authors to engage in self-publishing. These developments tend to create new communication circuits, for example, through electronic bulletin boards, that bypass media enterprises to reach the public. They put into question the traditional marketplace as a model of communication networks necessarily allocating works.(57)

        Do the market incentives of copyright work? My answer is: at best only in some cases of creation, but not in all cases, so that such incentives cannot be invoked as the sole basis for copyright. Do authors express themselves personally, at least often and obviously enough to justify copyright to defend the autonomy of such expression every time there is a work at issue? I would answer: they do not, for the simple reason that self-expression is not manifested in many cases of creation, so that the autonomy of such expression cannot form the only grounds for copyright. Furthermore, it is no longer possible to take the traditional marketplace for granted as the only valid model of every communication network in which copyright is supposed to apply. In seeking an overriding norm to govern the full range of possible cases, it therefore proves necessary to rethink our premises, at least on the points just broached.(58)

2. Recasting Copyright Values

        The values motivating copyright have shifted over the centuries. Legislators, in the eighteenth century, specified the purpose of copyright as the "encouragement of learning" or "the progress of science".(59) It would be interesting to know just why these Enlightenment lawmakers blithely contemplated instituting copyright as a means for pursuing this kind of goal. Under the sway of the then-pervasive view of thought, they perhaps tacitly assumed that the mind just worked that way: its essential function would be to generate information by virtue of simple and conscious references that its ideas quite naturally made to the things of the world.(60) Whatever became of this view, by the opening of the nineteenth century, copyright values had already begun to change: Romanticism turned to personal insight as the source of mental breakthroughs, although how these were to take place has remained imponderable. Nonetheless, copyright doctrine has continued to view the author's mind much like a private, consciousness-filled workshop. The law, however, including copyright law, does not deal with private mental constructs as much as with public behavior. It would therefore seem appropriate to reformulate copyright values in terms of what is publicly accessible.(61)

        Recall the tensions to which copyright laws become subject in following distinct norms. A marketplace norm would govern the market to furnish incentives for increasing informational wealth. An authorship norm would give authors the prerogative of assuring that their self-expression autonomously signified their insights. I propose to elaborate an overriding norm which would guide the law with an eye to enhancing these apparently divergent values: both informational wealth and significant expression. At the start of this century, de Boor observed that copyright comes into play when, "... as a result of creation by the mind, something arises as a possibility of communication (Mitteilungsmöglichkeit) that was not previously available and can be used, not only by the author, but by others as well".(62) Now, "creation" thus conceived, just like any new "possibility of communication" in which it would become manifest, could only take place in signs capable of being made publicly accessible. To keep analysis focused on these public things, I would propose the notion of sign wealth that would encompass the values of informational wealth and of significant expression. An overriding norm, aiming at sign wealth, ought not supersede either marketplace or authorship norms but rather subsume them both. This norm would impose complementary desiderata on copyright: making the variety of signs proliferate and access to them broaden. It will here be called the sign-wealth norm.

        Semiotics, the study of signs, provides a framework of analysis.(63) Manifold sign materials, such as sounds, colors, or shapes, make up signs. A simple digital language communicates by signs differentiated as zeros and ones. More complex, natural languages communicate by signs articulated in phonetic or visual units. A sign can only signify something in being interpreted, so that any semiotic framework of analysis has to include at least three types of elements: sign, something signified, and interpretation. Consider the simple case of pointing at something with a finger while using a name: the sign would be the gesture of pointing coupled with the name; what is signified, the thing being pointed at; and interpretation, further discourse or action with regard to that thing.(64) An interpretation can be quite mechanical, as when a warning light goes on, indicating that the engine of an automobile is over-heating -- but it is social interpretation which concerns us most especially. That is, when it is social, interpretation follows rules that allow large sets of signs to coordinate the behavior of the members of an entire group over time, these rules forming semiotic codes. Such rules may be artificially simple, as in the Morse code, while the complex codes of natural languages affect how verbal formulations generate rich, often ambiguous, and changing meanings. Canons, schema, and still-more fluid codes affect how harmonies, composition, and related uses of tones and colors, among other materials, generate sense in music and the fine arts. In these fields, it is necessary to avoid the temptation to tailor semiotic analysis on literary models without adapting it to musical and artistic works.(65)

        Literary, musical, and artistic works can serve as examples of the sign wealth constituting any culture. Cultures traditionally developed in varying degrees of isolation: over two millennia ago, even the Roman Empire, bringing together cultures around the Mediterranean, and the Han Dynasty unifying China did not engage in any dialogue with each other. Over time, the media have grown more powerful, pumping literature, music, and art from previously isolated cultures into a worldwide pool of sign materials, but without necessarily having them flow into a correspondingly enhanced sign wealth. As George Steiner warns, this trend toward worldwide communication rather threatens to end up in "a crisis in the organic coherence between language and its cultural content".(66) He anticipates this crisis with his notion of Babel, that is, too many languages, eventually a plethora of all kinds of sign materials, most of them experienced as nonsense. To move beyond Babel, it is necessary to enrich each language, so that it can translate meanings originally better conveyed and understood in other languages. This was the challenge of translating Shakespeare into German and French or, to take a more historic example, the Buddhist scriptures into Chinese. Of course, strictly speaking, one musical or artistic work does not "translate" another. It is nonetheless possible to speak of enriching our stock of musical or artistic signs.(67)

        It is then on this notion of sign wealth that I will rely to comprehend the diverse values motivating copyright law. Such values, and the norms they motivate, can be viewed more systematically from the perspective of intellectual property in general. Parallel to copyright law, which has just been seen as enhancing the wealth of cultural signs, patent law seems to be directed at increasing the wealth of technological know-how. Each of these laws focuses on a different process generating a distinct type of wealth: copyright law looks to "the authorship of works" and patent law to "invention", while both laws attend to allocating these types of resulting wealth through the marketplace. To this extent, the overriding norm specific to each of these fields of law will be distinct from that governing the other field, although these distinct norms do overlap in applying to hybrid products of mind, such as industrial designs, that fall between copyright and patents. I will nonetheless restrict myself to issues specific to copyright law rather than issues that concern such hybrid products in the penumbral zone between this law and patent law.(68)

3. Rethinking Creation

        Copyright doctrine has often confined itself to conceiving of creation in terms of author and work. Semiotic analysis leads to opening up this distinction between the subject and object of copyright by examining, not any author's mind alone, but rather the conditions under which works might contribute to sign wealth. It is a matter of analyzing possible legal relations, not merely between two elements, namely authors and works, but also between the following three types of elements: authors; works; and other persons, who form potential audiences for works.(69) For the moment, I will only briefly indicate how to reconceptualize creation with an eye to later asking what legal relationships between these three types of elements would tend to increase sign wealth.

        For copyright purposes, I propose to define works as global signs. That is, a work is distinguished by bringing together materials, like sounds, colors, or images, into a whole that has its effects as a sign with relative independence of context.(70) Thus an opera is made up of literary texts, stage directions, and music, all eventually brought together in an audiovisual production with an overture, recitatives, and arias. Each sentence spoken or sung in the opera is not necessarily a work in and of itself, since its linguistic meanings might change noticeably if it were heard in wholly different contexts. Take the sentence "Questo è il baccio di Tosca!" -- "This is Tosca's kiss!" -- as an example: stated by a maiden aunt doting over a young nephew, this sentence might express solicitous affection; uttered in the middle of the opera when Tosca knifes Scarpia, it signifies mortal contempt at least. By contrast, an aria of an opera does not radically change linguistic meanings or musical sense when it is heard as an isolated work in a recital and then in the context of the opera itself, although its overall sense is enriched in the latter case. To recapitulate, authors select sign materials and, to use one commentator's evocative metaphor, weave them together into the "fabric" of relatively self-standing global signs.(71)

        Defined as a global sign, a work need no longer be conceptualized either as some imaginary mental construct or as some material embodiment.(72) This proposed definition, like any attempt to define what copyright protects in general, is put to the test in specific cases of infringement. Consider the following hypothetical case: the author of a poem in English sues the author of a later poem in French for copyright infringement, alleging that it is a matter of the "same" poem which has only been translated into another language. The first author might win this suit in proving that the English poem was simply translated into the later poem in French; the second author could defend in showing that this later text was not the same poem translated in another language, but rather a clearly distinct work at most inspired by the earlier one. The success of this suit will turn on the "similarity" -- not between the fixed wording of one poem and that of the other, since each is written in a different language -- but between the sign content of one poem and then the other, notably between the meanings that each conveys. The trial court would have to find that, in this regard, both works were the "same", or sufficiently "similar", before determining what creative "substance", if any, to protect in the prior poem. The court would make this finding, not by reading both poems in the light of the authors' intents, but in the light of the English and French languages. Thus infringement analysis here looks to the "vocabulary" and "syntax" available to creators.(73)

        Such infringement analysis changes terms in moving from "literary" to "artistic" works.(74) Literary works are coded in terms of discrete, spoken or written symbols, notably words and phrases. Slightly altering how these signs appear, for example, changing type fonts, might affect their readability, but not necessarily their meanings. Artistic works are coded by using continuously manipulable sign materials, such as more or less intense colors or more or less accentuated chiaroscuro. The sign content of a painting can be spread across the space it occupies, so that slightly changing hue, luminosity, or compositional schema, almost anywhere in the painting, might change its overall impact. There are, of course, hybrid literary-artistic works -- for example, Chinese poetry is traditionally coupled with the art of calligraphy -- as well as other types of hybrid works such as operas or motion pictures. Musical works fall between literary and artistic works: music is coded in discrete symbols such as notes, as well as in instructions allowing of interpretation along continuums, such as those concerning the volume or speed of execution. Like literary works when faithfully translated, musical works remain more or less the "same" when faithfully orchestrated or performed and, therefore, likely to be infringed absent consent for such uses. Analysis can no longer follow the literary model when musical works become "different", for example, due to improvisations in the course of performance.(75)

        Creation is then to be rethought, no longer as the making of static things, whether mental or physical, but as the reworking of dynamic signs. Return to the case of the English poem allegedly translated into French: like all poetry, both poems are rich in ambiguities or, more precisely, plurivocal at many points. That is why the analysis of infringement concerns similarities between how these global signs contribute to sign wealth -- in particular, between such knots of multiple meanings as are typical of poetry -- but not between fixed traits constituting the identity of each work. Moreover, a work may vary in its embodiment but remain quite distinct from other works: for example, a concerto might call for improvised cadenzas, and an interactive video game can allow for changing its audiovisual interface. For copyright purposes, it is only necessary that the work take on meaning or sense with sufficient independence relative to context to avoid confusion with other works. Thus copyright may protect a work that, within different limits for different genres, remains more or less "open".(76)

4. Rethinking Communication

        Copyright doctrine tends to take the marketplace for granted as the model of relevant communication networks. This model, however, does not focus on how works contribute to sign wealth as they circulate within any communication network. Works are rather considered in their marketable form, as so many copies or performances that are to be protected because they represent rare commodities or services. For example, commentators ask about originality, that is, to what extent any given work has not been copied from others, or else novelty, that is, to what extent the work is statistically unique.(77) I will here indicate slightly different lines of inquiry that might better elucidate how works enhance sign wealth in being communicated.

        In revising the marketplace model, it is crucial to specify where, in any communication network, copyright comes into play. At the most basic levels of any network, the right of privacy entitles us to keep messages confidential, no matter what their contents or quality. This right is arguably so fundamental that the law may not obligate anyone to disclose confidential expression, whether creative or not, at more accessible levels of the network. Privacy claims become weaker, however, to the extent anyone voluntarily releases a message beyond intimate circles of communication to circulate within more public levels of the network. At such higher levels of the network, that include the open market, copyright permits its owners to control how others recommunicate works through the media. Copyright is nonetheless subject to the right of free expression: for example, anyone may freely recommunicate ideas conveyed in works. In the final analysis, according to this model, the media do not go beyond actively communicating works to passive consumers. A more complete model would include the feedback processes thanks to which some users start to create.(78)

        Consider the distinction Roland Barthes draws between mere "scriveners" (écrivants) and seminal "writers" (écrivains).(79) Scriveners rely on language as a storage room, from which they draw standard materials and tools without changing them. By contrast, writers, in working with language, for example, with vocabulary, syntax, and sound, at moments transform certain usages. This distinction can be reconsidered from the standpoint of a pair of legal notions that were just introduced: in the narrow sense of the term, a work displays "originality" to the extent it does not copy other works, and it displays "novelty" to the extent its shape is hitherto unknown. Even scriveners incrementally create original works: after drawing upon communication networks at hand for such elements as words, turns of phrase, grammatical structures, and literary forms, they recombine them into configurations that, to some extent, have not been copied. By contrast, writers achieve the breakthrough creation of novel works: they recast prior usages, fashioning new linguistic devices, often called new styles, with which they write texts that, at points, are unlike any other. Thus both scriveners and writers feed on language, but only the latter in turn feed language with their writing.(80)

        At the same time, such creations may well call for new interpretations. Umberto Eco speaks of works so novel that they drive us to reconstrue semiotic codes: he gives the example of a poem that readers could only understand if they tried "to rethink the whole language, the entire inheritance of what has been said, can be said, and could or should be said."(81) Such works could exemplify new poetic forms, new story lines, new harmonies or rhythms, new color palettes or compositional schemes, that is, in general terms, new devices for elaborating new global signs. Copyright, of course, protects such a work, but it does not cover these devices as such, even though they constitute the most important contributions that breakthrough creation can make to sign wealth. Indeed, subsequent authors are free to use such new devices as any novel work embodies for creating further works -- free to explore the semiotic potential that this work opens up. Copyright then protects the further works that these devices facilitate incrementally creating even if, as a result, these works are not novel, but only original. For example, Renaissance painters discovered "master keys" to picture depth, such as foreshortening and perspective. The artists of the following centuries freely used these keys in painting further, original pictures.(82)

        Communication then has to be rethought without remaining confined to the marketplace model. However, within any communication network, rights more important than copyright may also be asserted. As already noted, these rights flow from norms requiring respect for the human personality and freedom of expression. These are more generally overriding norms than any sign-wealth norm, in that they may apply in many fields of law rather than in just one specific field such as copyright. Of course, respect for each personality reserves realms of privacy in which to experiment with a variety of new signs without fear of being disturbed, and freedom of expression tends to make that variety of signs increasingly accessible. Thus, in most cases, but not in all possible cases, these more general overriding norms would coincide in aims with the overriding sign-wealth norm specific to copyright. It will prove necessary, in the next and final part of this study, to sort out these overlapping norms.(83)

III. RESOLVING TENSIONS IN PRACTICE

        Return to the difficult cases in which marketplace and authorship norms lead to inconsistent results. Often, to decide such cases left unsettled by certain norms specific to a given field, lawmakers look to more general norms. They may well invoke some ultimate norm of justice and equity, hopefully applicable throughout the law, or else follow those general norms just broached: those which require respect for the human personality and freedom of expression.(84) In this final part, I will only refer to these more general norms for the purpose of guiding the application of the specific overriding norm which I have proposed for copyright. In subparts 1, 2, and 3, respectively, I will apply this sign-wealth norm to the issues of the scope, the priority, and the subjects and objects of rights. In subpart 4, I will consider cases involving some mix of these diverse issues.

1. The Scope of Rights

        It is possible to distinguish between limits imposed on copyright from the inside by reason of norms that are specific to it and limits imposed from the outside by reason of more general norms. On the one hand, in certain difficult cases, marketplace and authorship norms, both specific to copyright, respectively lead to limiting this right by giving it narrow and broad scopes and, accordingly, broad and narrow exceptions.(85) On the other hand, beyond copyright, the norm requiring freedom of expression would suffice, for example, to justify giving each user of a work a right to take illustrative quotations, thus limiting copyright which might otherwise preclude such copying. Without losing sight of more general norms, I would propose to consider how the overriding norm specific to copyright, that aiming at sign wealth, might help us to harmonize the scope and exceptions of this right.(86)

        A sign-wealth norm tends to give broad protection to the timid muse that privately inspires new works. Consider the case in which a news magazine quoted excerpts from the memoirs which ex-President Ford of the United States had been preparing in private but had not yet released for publication. The U.S. Supreme Court ruled that these quotations, because they came from unpublished passages, did not fall into the exception of fair use but rather violated copyright, which not only protected the economic right to control publication of the memoirs but also the author's "personal interest in creative control" of his work.(87) It is nonetheless important to distinguish between functions that judges may exercise in this last regard: outside copyright, courts might have to coordinate rights to both privacy and to freedom of expression, including the right to quote works; inside copyright, courts protect privacy only as that special inner space from which authors contribute to sign wealth. A court could impose strong remedies to guarantee the rights of privacy and of freedom of expression, which are constitutionally basic to a liberal society, just as it could fashion more limited remedies for copyright which, after all, is only a specially legislated right.(88)

        When an author is no longer alive, a sign-wealth norm allows for narrowing the scope of copyright with regard to the disclosure of works. Of course, a deceased author could no longer exercise any creative function, any "creative control", relative to works that he had never disclosed to third parties. Nonetheless, in theory, French copyright law recognizes authors' moral rights as perpetual, granting certain heirs standing to assert such rights after the author's death. But, in practice, French law limits the moral right to control disclosure by empowering the courts to authorize the release of posthumous works to the public in cases in which authors' successors arbitrarily withhold such disclosure.(89) A court could not, however, order any such release if the heirs did not act arbitrarily, most notably in cases where authors had explicitly formulated the intention that certain works not be made available to the public after their death. Still and all, upon death, the author's privacy is no longer an issue, and sign-wealth seems to be enhanced by disseminating as many posthumous works as possible to the public. It would thus have justified a decision such as Max Brod's to publish Kafka's posthumous works in the face of that author's statement of intent to the contrary.(90)

        A sign-wealth norm leads to comparable decisions as works cross the threshold of privacy into any communication network feeding the public. A case arose on just such a threshold in the United States, when a researcher found certain personal letters by the reclusive novelist J.D. Salinger in university libraries. The researcher proposed to quote the letters to illustrate a biography of Salinger, who in turn brought a copyright action to enjoin publishing the quotes in the biography, a suit which he won.(91) Perhaps, in this matter, a pure privacy right would have provided appropriate grounds for bringing suit, at least against the recipients of the letters at issue and the university libraries, if not against the author of the biography and publishing company. In any event, the court's decision, based on copyright, clearly made it difficult to make good use of the letters in the biography and, as a result, to contribute to sign wealth. Further, there could have been no question of the author's exercising "creative control" over the letters at issue, which had been finished and sent off. Finally, this case law gives a dead author's successors the power to censor quotes from posthumous works.(92)

        Look, for a moment, to the realm of privacy where we often enjoy works. On the one hand, the general norm requiring respect for the human personality precludes copyright measures that might violate the privacy rights of the end-users of works. Avoiding any such intrusion into privacy, many copyright laws deal with the home copying of works by remunerating authors from levies imposed on the eventually public sale of recording media. On the other hand, neither the norm requiring freedom of expression nor any sign-wealth norm would allow for precluding private uses likely to generate new works, as long as these uses did not prejudice the public exploitation of the prior works being used. For example, young painters have long had the custom of freely making private copies of the works of old masters, in order to improve their skills and techniques to the point of helping them paint new works. Many copyright laws now tend to exempt the case of privately making research copies of computer programs, notably by decompilation, to develop new programs.(93)

2. The Priority of Rights

        While marketplace norms call for freely transferable economic rights, authorship norms require inalienable moral rights. Difficult cases arise when authors transfer all economic rights in their works but later invoke moral rights in ways that prevent transferees from fully exploiting these works.(94) Of course, if the exploitation in question included simple torts against their persons, authors could always bring suits to impose general civil liability for these torts, this quite independently of copyright. They could, for example, protect their personal reputations from defamation resulting from the public dissemination of their works in distorted form or without proper attribution of authorship. It is, however, not my concern to determine how, in such cases, the general norm requiring respect for personality, and therefore for personal reputation, might govern simple tort actions. My question is rather whether, and how, an overriding sign-wealth norm, which is specific to copyright, allows us to coordinate economic and moral rights.(95)

        A sign-wealth norm does not allow for granting authors rights to protect the integrity of their works at will. The mutilation or destruction of an objet d'art might well diminish sign wealth to the extent the object is unique or rare. The case is different when a work is transformed, not by tampering with its available embodiments, but by translation or adaptation that generates new embodiments. Now, if authors of prior works had absolute control over such attempts to rework sign materials from their works, they could eventually prevent others from deriving new works. One commentator, favoring such absolute control, reasoned that Bizet's opera Carmen ought not have its integrity violated by the motion picture Carmen Jones, where black performers acted out the story of the tragic opera in American settings.(96) By parity of reasoning, Prosper Mérimée or his heirs could have prohibited Bizet from adapting the story of Carmen into the opera: if generalized to its furthest consequence, such reasoning could preclude any work that later authors might make on the pattern of past works while deviating from their "spirit". This reductio ad absurdum suggests that, in a case where a transferee holds the economic right to derive a new work from a prior work, but where an author asserts the moral right of integrity against that derivative work, it might enhance sign wealth to allow that new work to be disseminated. This approach expands upon the various ad hoc measures to which courts almost worldwide resort to accommodate adaptation rights held by contractual transferees with integrity rights asserted by authors.(97)

        In effect, a sign-wealth norm requires drawing rather fine distinctions in such difficult cases. It allows for precluding technical changes that only impoverish the sign content of works, such as arbitrary cuts or interruptions and poor recordings. In many cases, however, some technical changes might be permissible to the extent that they are indispensable to making works accessible to the public, but only to that extent. For example, films made for big theater screens can still only be broadcast onto smaller television screens, where they at times lose some of their cinematographic impact, and it is only possible to have non-pay and non-state television by relying on advertisements. In one Italian case, the first trial court, noting the demands of such exploitation, consequently refrained from categorically prohibiting all advertisements that might interrupt a televised film, but it did contemplate prohibiting undue interruptions such as those made at high points in the story line of the film.(98) Live or recorded performances can constitute more delicate cases: in playing or acting a work, artists vary how it is heard or seen, so that such a performance can be classified as more creative than technically presenting a work but less creative than adapting it. It has been argued that old musical works should only be performed on such instruments as were originally available during the period when these works were composed. One response to this position is that systematically imposing such limitations could often impoverish our experience of the musical potential of many works. Sign wealth is, of course, enhanced when artists are free to follow different points of view in giving a variety of performances.(99)

        A sign-wealth norm does lead to reinforcing the right to credit for authorship. It is implicit, in the most widespread understanding of works, that human minds create them. The failure to attribute the authorship of these global signs to their actual creators might accordingly impoverish their very sign content. Until now, it has often been costly to provide full information concerning authorship, for example, during a television broadcast where time is at a premium. In the future, such information costs will drop as works are transmitted in compressed digital form, through telecommunication networks with broad channels, are electronically stored as needed, and enjoyed at the end-user's leisure. Furthermore, it will become easier, not merely to attribute authorship systematically, but to create new works by reworking prior works electronically and to indicate prior works thus transformed into any such new derivative work when telecommunicating that new work.(100) I have just argued that a sign-wealth norm would tend to allow exploiting derivative works pursuant to the transfer of economic rights that authorized such exploitation, this even when the authors of prior underlying works object in asserting their moral rights of integrity. Nonetheless, it is important to recognize that sign wealth might be best served if end-users, in enjoying a new derivative work, were in a position to refer back to any prior underlying work to judge for themselves whether the new work distorts the prior one. As if to facilitate such references, some scrupulous "post-modern" authors, in communicating works they create by reconfiguring the works of other authors, acknowledge the identities of these past works and these other authors.(101)

        Perhaps, above all in new media, a sign-wealth norm justifies but one moral right, which I will call a right to reference. On the one hand, this right would require each communication of a work to include references to those persons who have creatively contributed to the overall fabric of that work, as well as to their diverse roles as authors. On the other hand, this same right would require each communication of a work to include references to the prior works that these authors have consciously transformed in creating the overall fabric of the work at hand. Signatures of the authors, as well as reference numbers of the works themselves, could be electronically embedded into the fabric of digitally archived works. The right to reference would, of course, imply strictly prohibiting the removal of such electronic identification marks, even when reworking prior works to create new ones. Following such marks across networked digital archives, end-users could, starting from any given work, refer back to underlying works from which this one had been drawn. Such hypertext references would thus facilitate end-users' judging for themselves the aesthetic fate of works reworked over time. This is but one example of how such references might enhance sign wealth, especially if they are legally protected.(102)

3. The Subject and Object of Rights

        Copyright laws sometimes identify different subjects who initially hold rights and different objects that these rights protect. While a marketplace norm allows for allocating rights to both natural persons and to legal entities, an authorship norm only lets rights vest in flesh-and-blood creators. Further, these norms each lead to differently emphasizing the criteria of protected works: a marketplace norm requires at most minimal creativity; an authorship norm, some personal imprint. I submit that an overriding sign-wealth norm might help lawmakers moot these tensions.(103)

        A sign-wealth norm would seem to favor natural persons as subjects of rights, since the human race constitutes the largest and most diverse pool of potential creators on this planet. It would not, however, necessarily preclude vesting some rights in other parties: in particular, if legislators splintered economic rights in team works by vesting them in all the team members, such works would become difficult to market and, as a result, to make accessible. In truth, there need be no uniform solution to the problem of allocating economic rights in works as long as the policies leading to one solution in one jurisdiction are not imposed on others.(104) Multiple solutions allow for diverse contractual situations, in which a variety of works might be created, thus ostensibly supporting sign wealth. By contrast, it seems necessary to vest any moral right, including the right to reference just proposed, only in flesh-and-blood creators. Such a right responds to the all-too-human needs that, throughout history, have driven authors to create: not so much needs for money as glory.(105)

        The difficult issue raised in infringement cases is not what to protect, but how much protection to grant. Any norm, including a sign-wealth norm, in directing the trial court toward a resolution of this issue, delimits the effective object of copyright. Different copyright laws apply differently formulated, but functionally comparable rules to narrow the scope of protection in cases where there is only one way or, more often, too few ways of forming a work.(106) American copyright law applies a rule of merger: to the extent an idea can only be rendered concrete in one or too few expressions, it is said to merge with the expression of the work at issue, which to that extent is not protected. Many European laws seem to presuppose a comparable notion, which is sometimes formulated in terms of a space made up of creative options (Spielraum): to the extent there are fewer options for giving form to a work, there are fewer variations on this work likely to infringe it. It is, however, difficult to draw conclusions for sign wealth as long as it remains unclear what is meant by an "idea" or, for that matter, by "too few" ways to "express" the idea or by exercising "creative options".(107)

        To make things more concrete, replace "ideas" with sign materials and "expression" with translating, reconfiguring, and otherwise reworking such materials. From the standpoint of sign wealth, it is not just a matter of determining what materials to protect or not in the plaintiff's work, but also what procedures the defendant should be free to use in reworking such materials. Metaphorically speaking, where there are few ways to translate, reconfigure, or otherwise rework the sign materials constituting a given work, copyright only extends to a "thin" surface of the work, and the court may only protect it against making slavish copies. Some examples of such materials would be stock plots or scenes in plays or novels, standard symbols and legends on maps or charts, indeed any sign obtaining conventionalized meanings. Where, as in most literary, musical, or artistic fields, the possibilities for recasting sign materials are more varied, the protected "core" of the work is "thicker". In such cases, the analysis of copyright infringement is more highly differentiated.(108)

4. Mixed Cases

        Consider parodies as an example of a mixed case. Most copyright laws make exceptions of such works, while freedom of expression can also justify allowing them and many works like them. Nonetheless, parody cases are not always consistently decided, no more than many other difficult cases subject to mixed and shifting considerations. I would submit that, in such cases, an overriding sign-wealth norm should guide lawmakers toward more cogent results.(109)

        A sign-wealth norm might have helped the courts to see one apparently simple case in a different light. The conceptual artist Jeff Koons had a sculpture made by closely following a photograph of a man and a woman who were holding a string of little puppies. The trial court in the United States found that the sculpture infringed copyright in the photograph without any room for doubt and rejected the specific defense of parody raised in this case.(110) The artist Koons had frankly admitted deliberately imitating the sign materials which the photographer had brought together in the image of the couple and the puppies, and he ventured the defense that he could hardly have done anything else to parody the banality of such sentimental images current in the media. Other artists now worry that this case law precludes them from using a whole range of "post-modern" procedures for enhancing sign wealth. These are procedures for reconfiguring or recontextualizing specifically representative signs to critique generally held ideas.(111)

        Cases where the parody defense is raised illustrate all-too well and all-too often the judicial vicissitudes of analogous derivative works likely to increase sign wealth. French law expressly provides an exception for "parodies, pastiches, and caricatures"; however, French courts sometimes appear to be impatient with works that, while incorporating sign materials from prior works for reasons apparently akin to parody, do not strictly follow, to quote the statute, "the laws of the genre".(112) This narrow understanding did not allow, for example, for a play which presented a critical vision of the comic-strip character "Tintin", belittling him as overwhelmed by the petty details of life, but which did not infringe any particular comic-strip. Finding the play to be too serious to be a parody, the court held it to violate the "integrity of the work", but without clarifying what work was at issue: the comic-strips, which were not copied, or "Tintin", who was less assuredly a work. Thus the court might have precluded the play from enhancing sign wealth without protecting any other specific work at all.(113)

        Avoiding the technicalities of either open-ended or narrow exceptions, a sign-wealth norm instead allows for resolving these cases through infringement analysis. I have argued that copyright protects, not any given configuration of sign materials alone, but rather a global sign that pulls such materials together to generate meanings or sense. Thus a work would not violate copyright if, in incorporating sign materials taken from a prior work, like the image of the couple holding the string of puppies or the comic-strip character "Tintin", it radically changed their sense. That is, in order to avoid infringement, it is necessary for the parody or any other comparable work to constitute, in the apt words of one commentator, the "subversion of the sense" of the prior work.(114)

CONCLUSION

        In copyright, marketplace and authorship norms, each predicated on one-sided premises, lead to inconsistent results. I have proposed an overriding norm that might guide lawmakers in resolving the basic tensions that then arise in such difficult cases. Like all doctrinal thought-experiments, my argument for this sign-wealth norm has surely ignored many troublesome details with which the law must finally come to grips.

        Before closing, it might still be useful to look beyond such details. Imagine a new media system in the guise of a universal electronic library: all works, created from the distant past forward, would be archived somewhere among millions of digital memory devices. Further, we would navigate through these electronic archives via a worldwide telecommunications network: we could thus eventually find all works there, receive and rework any of them at multimedia computer terminals, and input resulting new works into the system. This system would seem to satisfy the desiderata of sign wealth: variety and access. But it would not necessarily be sure to satisfy us....

        Jorge Luis Borges described a "Library of Babel".(115) Since it contained all books, this Library had to include a "catalogue" of its contents, as well as books to decipher all the others coded in "lost languages." The librarians looked high and low for a "book which is the formula and perfect compendium of all the rest", wandering aimlessly through "infinite" galleries and stairways from one "elegant endowment of shelves" to another. But wouldn't our utopian electronic library end up reminding us a bit strangely of Borges' nightmarish Library of Babel? Users would navigate across our library seated at computer terminals, as the librarians wandered about that of Babel on foot. But how would any of them know for what, and where, to search in their universe of endless data? How would they sort out all these potential signs without any "catalogue" or key? How would they make sense of it all?

        Perhaps increasing the cumulative variety of works, while providing virtual access to them all, might not suffice to achieve sign wealth. It remains to be seen how to move from the mere proliferation of potential signs, no matter how diverse and available, to making sense of them in a new media world. It is possible that these desiderata of variety and accessibility, which I have proposed for sign wealth, are only good enough to keep our bearings in presently difficult cases. In that event, we have only approached an overriding norm for copyright without formulating it in fully adequate terms. It might not yet be ready to save us from being again caught in basic antinomies as we plunge forward into the future.

1. Portalis, "Discours préliminaire prononcé lors de la présentation du projet de la commission du gouvernement" (1er pluviôse, an IX), in Naissance du Code Civil 41-42 (F. Ewald, ed. 1989).

2. From here on, the term "copyright" will also refer to author's rights in the European sense. For a typology of diverse copyright laws, see Paul Edward Geller, "International Copyright: An Introduction", sec. 2[3], in International Copyright Law and Practice, vol. 1 (P.E. Geller and M.B. Nimmer, eds. updated 1993).

3. For an analysis of how norms emerge in historically dynamic bodies of law -- as distinguished from any Grundnorm such as Hans Kelsen posits as the basis for fictively static systems of law, in which "everything falls into place" -- see André-Jean Arnaud, Critique de la raison juridique -- 1. Où va la Sociologie du droit?, passage quoted at 336 (1981).

4. For the distinction between models or examplars and "paradigms" in the large sense of that term, see Thomas S. Kuhn, "Postscript -- 1969", in The Structure of Scientific Revolutions 174 (2d ed. 1970).

5. See Elizabeth L. Eisenstein, The printing press as an agent of change 62-129 (1979).

6. For the early book trade, see Lucien Febvre and Henri-Jean Martin, L'apparition du livre, ch. 5 (2d ed. 1971).

7. Compare L. Ray Patterson, Copyright in Historical Perspective, chs. 2 and 4-6 (1968) (England), with Marie-Claude Dock, Étude sur le droit d'auteur, bk. 2 (1963) (France).

8. Compare Harold A. Innis, "The English Publishing Trade in the Eighteenth Century", in The Bias of Communication 142 (1951) (England), with Robert Darnton, The Literary Underground of the Old Regime, ch. 1 (1982) (France).

9. See C. B. Macpherson, The Political Theory of Possessive Individualism: Hobbes to Locke, esp. ch. 6 (1962); Jürgen Habermas, Strukturwandel der Öffentlichkeit 142-160 (new ed. 1990).

10. 8 Anne, c. 19 (1710). For parallel developments, see Jane C. Ginsburg, "A Tale of Two Copyrights: Literary Property in Revolutionary France and America", 64 Tulane Law Rev. 101 (1990).

11. Compare L. Ray Patterson, "Free Speech, Copyright, and Fair Use", 40 Vanderbilt Law Rev. 1, 13-36 (1987) (Anglo-American law); Carla Hesse, Publishing and Cultural Politics in Revolutionary Paris, 1789-1810, esp. chs. 1 and 3 (1991) (France); Martin Vogel, "Urheberrecht in Deutschland zwischen Aufklärung und Vormärz", 77 Buchhandelsgeschichte 96 (1979) (Germany).

12. For further historical analysis of these shifts, see Benjamin Kaplan, An Unhurried View of Copyright, ch. 1 (1967); Ysolde Gendreau, "Genèse du droit moral dans les droits d'auteur français et anglais", 1988 (no. 1) Revue de la recherche juridique 41; György Boytha, "Die historischen Wurzeln der Vielfältigkeit des Schutzes von Rechten an Urheberwerken", in Die Notwendigkeit des Urheberrechtsschutzes im Lichte seiner Geschichte 69 (R. Dittrich, ed. 1991); André Kerever, "Copyright: The Achievements and Future Development of European Legal Culture", 26 Copyright 130 (1990).

13. Adam Smith, The Wealth of Nations 712 (E. Cannan, ed. 1937).

14. See Denis Diderot, Sur la liberté de la presse (sometimes identified as Lettre sur le commerce de la librairie) 62-65 and 79-88 (J. Proust, ed. 1964).

15. See Walter Bappert, Wege zum Urheberrecht 105-109 (1962).

16. John Locke, An Essay Concerning Human Understanding, vol. 1, 121 (bk. 2, ch. 1), 144-147 (ch. 2), 159-165 (ch. 6), and 213-217 (ch. 12) (A.C. Fraser, ed. 1959).

17. See Lucien Goldmann, "La philosophie des lumières", in Structures mentales et création culturelle 21, 23-25 (1970).

18. John Locke, Second Treatise of Government 29 (sec. 49) (C.B. Macpherson, ed. 1980).

19. Id. at 27-28 (sec. 45) (1980).

20. For the extent of his anticipation of copyright, see John Locke, "Observations on the censorship", in Peter King, The Life and Letters of John Locke 202-209 (1884, reprinted 1972).

21. See Diderot, supra note 14, at 41-43.

22. Compare Mark Rose, Authors and Owners: The Invention of Copyright 71-84 (1993) (England), with Dock, supra note 7, at 115-126 (France).

23. Michael Lehmann, "The Theory of Property Rights and the Protection of Intellectual and Industrial Property", 16 International Review of Industrial Property and Copyright Law (I.I.C.) 525 (1985).

24. Joseph Addison, "On Genius" (no. 160, 3 Sept. 1711), in Addison and Steele, The Spectator, vol. 2, 126-127 (D.F. Bond, ed. 1965).

25. See Denis Diderot, "Génie", in L'Encyclopédie (Diderot and d'Alembert, eds., 1757), excerpted in Diderot, De l'interprétation de la nature 142 (J. Varloot, ed. 1971).

26. See Martha Woodmansee, "The Genius and the Copyright: Economic and Legal Conditions of the Emergence of the 'Author'", 17 Eighteenth-Century Studies 425 (1984).

27. Immanuel Kant, Kritik der Urteilskraft 33-35 ("Einleitung", IX) (K. Vorländer, ed. 1924).

28. Id. at 161-164 (secs. 46-47).

29. See M.H. Abrams, The Mirror and the Lamp: Romantic Theory and the Critical Tradition, ch. 9 (1953).

30. Immanuel Kant, "Von der Unrechtmässigkeit des Büchernachdrucks", originally published in 5 Berlinische Monatschrift 403 (1785) and cited below as reprinted in 106 Archiv für Urheber- Film- Funk- und Theaterrecht (UFITA) 137 (1987).

31. Immanuel Kant, "Beantwortung der Frage: Was ist Aufklärung?", in Was ist Aufklärung? Aufsätze zur Geschichte und Philosophie 55, 57-58 (J. Zehbe, ed. 1985).

32. Kant, "Von der Unrechtmässigkeit des Buchernachdrucks", supra note 30, at 138-143.

33. Id. at 143-144. For Kant's aesthetic analysis favoring literature, see his Kritik der Urteilskraft, supra note 27, at 183-188 (sec. 53).

34. Compare Rose, supra note 22, at 85-91 (England), with Hesse, supra note 11, at 100-124 (France).

35. Johann Gottlieb Fichte, "Beweis der Unrechtmässigkeit des Büchernachdrucks. Ein Räsonnement und eine Parabel", originally published in 21 Berlinische Monatschrift 443 (1793) and cited here as reprinted in 106 UFITA 155, 158-159 (1987).

36. Compare Rose, supra note 22, at ch. 7 (England), with Woodmansee, supra note 26 (Germany).

37. Bernard Edelman, La propriété littéraire et artistique 39 (1989).

38. For a systematic comparative analysis, see Alain Strowel, Droit d'auteur et copyright: Divergences et convergences (1993).

39. Sony Corp. v. Universal City Studios, Inc., 464 U.S. 417, 447-456 (1984). For the statutory provisions, see Copyright Act (17 U.S.C.), secs. 106-107 (as amended 1992).

40. In re Kästner, 7 July 1971, 1972 Gewerblicher Rechtsschutz und Urheberrecht 481, English trans. in 3 I.I.C. 394 (1972).

41. For a critique, see Keith Schilling, "The New Moral Rights in English Law -- All Bark and No Bite?", 1990 (no. 9) Copyright World 26.

42. For a comparison, see Adolf Dietz, Das Droit Moral des Urhebers im neuen französischen und deutschen Urheberrecht, ch. 10 (1968).

43. For a critical comparison, see Bernard Edelman, "Entre copyright et droit d'auteur: l'intégrité de l'oeuvre de l'esprit", 1990 (no. 40) Dalloz (Chronique) 295.

44. See 17 U.S.C. secs. 101 in fine and 201.

45. See Loi no. 57-298 of 11 March 1957 (Loi de 1957), arts. 1, 6, 9, and 13 (as codified in 1992: Code de propriété intellectuelle [CPI], arts. L.111-1, L.121-1, and L.113-2 and -5).

46. See Gesetz über Urheberrecht und verwandte Schutzrechte, secs. 7, 31, and 39 (as amended 1990). For the doctrinal basis of the German approach to copyright transfers, see Eugen Ulmer, Urheber- und Verlagsrecht 114-118, 363-365 (3d ed. 1980).

47. For a comparative analysis, see Paul Edward Geller, "Copyright in Factual Compilations: U.S. Supreme Court Decides the Feist Case", 23 I.I.C. 802 (1992).

48. For this position, see Edelman, La propriété..., supra note 37, at 13-25.

49. For other recent and notable attempts at harmonization, see Paul Goldstein, "Copyright: The Donald C. Brace Memorial Lecture", 38 Journal of the Copyright Society of the USA (J. Copr. Soc'y) 109 (1991); Adolf Dietz, "Transformation of authors rights: change of paradigm", 138 Rev. int. dr. aut. 22 (1988); idem., "Copyright in the Modern Technological World: A Mere Industrial Property Right?", 39 J. Copr. Soc'y 83 (1991).

50. Compare Jane C. Ginsburg, "Creation and Commercial Value: Copyright Protection for Works of Information", 90 Columbia Law Rev. 1865, 1873-1893 (1990) (low- to high-authorship), with Dietz, "Copyright in the Modern Technological World", supra note 49, at 88-89 (distinguishing creativity from authorship).

51. Thomas Macaulay, Speech delivered in the House of Commons, 5 Feb. 1841, excerpted in Paul Goldstein, Copyright, Patent, Trademark and Related State Doctrines 4 (3d ed. 1990).

52. William M. Landes and Richard A. Posner, "An Economic Analysis of Copyright Law", 18 J. of Legal Studies 325, 327 (1989).

53. Josef Kohler, Das literarische und artistische Kunstwerk und sein Autorschutz 37-39, 83-86, and 138-139 (1892); idem., Kunstwerkrecht 27-30 (1908).

54. Germi c. Reteitalia e Rizzoli Film, Corte di Appello, Rome, 16 Oct. 1989, 61 Il Diritto di Autore 98, 104-105 (1990).

55. T.S. Eliot, "Tradition and the Individual Talent", in The Sacred Wood: Essays on Poetry and Criticism 55, 58 (1920).

56. See Innis, "Technology and Public Opinion in the United States", in The Bias of Communication, supra note 8, at 159-160 and 172-173.

57. For a more complex analysis, see Ithiel de Sola Pool, Technologies of Freedom, esp. at 196-217 (1983).

58. For another critique on many comparable points, see David Vaver, "Some Agnostic Observations on Intellectual Property", 6 Intellectual Property Journal 125 (1991).

59. See the Statute of Anne (cited supra note 10) and the U.S. Constitution, art. I, sec. 8, cl. 8. For the broad Enlightenment sense of "knowledge", which encompassed poetry, music, and the fine arts, see Jean le Rond d'Alembert, Discours préliminaire de l'Encyclopédie 49-51 (new ed. 1765, reprinted 1965).

60. For the analysis of these epistemological premises, see Michel Foucault, Les mots et les choses, esp. at 77-81 and 92-94 (1966).

61. For the critique of subjectivist methodology in copyright, see Ivan Cherpillod, L'objet du droit d'auteur 41-51 and 111-114 (1985).

62. Heinrich O. de Boor, Urheberrecht und Verlagsrecht 167 (1917).

63. For a general introduction, largely followed here, see Umberto Eco, A Theory of Semiotics (1976).

64. For the origins of this framework of analysis, see Charles S. Peirce, "Some Consequences of Four Incapacities", in Charles S. Pierce: The Essential Writings 85 (E.C. Moore, ed. 1972).

65. For attempts at such adaptation, see Diana Raffman, Language, Music, and Mind, esp. ch. 3 (1993); E.H. Gombrich, Art and Illusion: A Study in the Psychology of Pictorial Representation, esp. ch. 9 (2d ed. 1961).

66. George Steiner, After Babel: Aspects of Language and Translation 491-492 (2d ed. 1992).

67. For examples, see Morse Peckham, Man's Rage for Chaos, chs. 4-5 (1965).

68. For an analysis of this zone, see Jerome H. Reichman, "Legal Hybrids Between the Patent and Copyright Paradigms", in Information Law Towards the 21st Century 325 (W.F. Korthals, E.J. Dommering, P.B. Hugenholtz, and J. Kabel, eds. 1993).

69. For seminal analyses focusing on such three-sided legal relations, see W.N. Hohfeld, "Fundamental Legal Conceptions as Applied in Judicial Reasoning", 26 Yale Law J. 710 (part 2), esp. at 733 (1917); de Boor, supra note 62, at 26-72.

70. For a debate on the extent to which literary works take on meanings independently of context or readers, see Umberto Eco, with Richard Rorty, Jonathan Culler, Christine Brooke-Rose, Interpretation and overinterpretation (S. Collini, ed. 1992).

71. See Ulmer, supra note 46, at 119-125.

72. For a critical analysis, see Robert H. Rotstein, "Beyond Metaphor: Copyright Infringement and the Fiction of the Work", 68 Chicago-Kent Law Rev. 701 (1993).

73. See Paul Goldstein, Copyright: Principles, Law and Practice, vol. 2, ch. 8 (updated 1992); Geller, "Copyright in Factual Compilations ...", supra note 47, at 805.

74. For the source of the following analysis, see Nelson Goodman, The Languages of Art, esp. ch. 5 (1976). For a critical application to music, see Raffman, supra note 65, at ch. 6.

75. For a fuller analysis of the "deviations" that can "more or less" transform musical works, see Leonard B. Meyer, Emotion and Meaning in Music, chs. 6-7 (1956).

76. For the concept of an "open work", see Umberto Eco, Opera aperta 31-63 and 138-184 (1976).

77. For examples, see Melville and David Nimmer, Nimmer on Copyright, vol. 3, sec. 13.03[A][1][b] (updated 1992); Max Kummer, Das urheberrechtlich schützbare Werk, esp. at 30-57 (1968).

78. See Serge Proulx, "De la métaphore télégraphique à celle de la conversation: Représentations du pouvoir des médias et modèles de la communication", in Technologies et symboliques de la communication 283 (L. Sfez and G. Coutlée, eds. 1990).

79. See Roland Barthes, "Écrivains et écrivants", in Essais critiques 147 (1964); idem., Le plaisir du texte (1973). For a comparable distinction in the fine arts, see George Kubler, The Shape of Time: Remarks on the History of Things 39-53 and 62-82 (1962).

80. For a different attempt to relate originality and novelty, see Gunnar Karnell and Thomas Dreier, "Originality of the Copyrighted Work: A European Perspective", 39 J. Copr. Soc'y 289 (1992).

81. Eco, A Theory ..., supra note 63, at 274.

82. See Gombrich, supra note 65, at 359-360.

83. For a different analysis of the relations between such norms, see Ejan Mackaay, "An Economic View of Information Law", in Information Law Towards the 21st Century, supra note 68, at 43.

84. For analyses trying to apply norms of justice and equity to intellectual property, see Wendy Gordon, "An Inquiry into the Merits of Copyright: The Challenges of Consistency, Consent, and Encouragement Theory", 41 Stanford Law Rev. 1343, 1455-1465 (1989); François Dessemontet, "L'enrichissement illégitime dans la propriété intellectuelle", in Festgabe für Max Kummer 191 (H. Merz and W.R. Schluep, eds. 1980).

85. For these difficult cases, see supra text accompanying notes 39-40.

86. For an analysis that draws a large exception from more general norms, see Cherpillod, supra note 61, at 145-171.

87. Harper & Row, Publishers, Inc. v. Nation Enterprises, 471 U.S. 539, 555 (1985).

88. For reflections in this regard, see Jon O. Newman, "Not the End of History: the Second Circuit Struggles with Fair Use", 37 J. Copr. Soc'y 12 (1989); James L. Oakes, "Copyrights and Copyremedies: Unfair Use and Injunctions", 38 J. Copr. Soc'y 63 (1990).

89. See Loi de 1957, arts. 6, 19, and 20 (CPI, arts. L.121-1, L.121-2, and L.121-3). Compare Sté. Art Conception Realization c. la Veuve Foujita, 16 Nov. 1990, Cour d'appel Rennes, 148 Rev. int. dr. aut. 168 (1991) (publication allowed), with Sté. la Règle du Jeu c. Salzedo, 24 Nov. 1992, Cour d'appel Paris, 1st chamber, 155 Rev. int. dr. aut. 191 (1993) (publication inconsistent with the intent of deceased author, in this case, Roland Barthes, is precluded).

90. See Max Brod, "Nachwort zur Ersten Ausgabe", in Franz Kafka, Der Prozess 277 (3d ed. 1946).

91. Salinger v. Random House, Inc., 811 F. 2d 90 (2d Cir., 1987), cert. denied, 484 U.S. 890 (1987).

92. For further critical analysis, see Pierre N. Leval, "Toward a Fair Use Standard", 103 Harvard Law Rev. 1105, 1113-1122 (1990).

93. Compare Sega Enterprises Ltd. v. Accolade, Inc., 977 F.2d 1510, 1520-1530 (9th Cir. 1992) (judge-made exemption), with E.C. Directive of 14 May 1991 on the Legal Protection of Computer Programs, art. 6 (OJEC no. L 122/42) (statutory exemption).

94. For these difficult cases, see supra text accompanying notes 41-43.

95. For the need to distinguish general personality rights from any moral right specific to copyright, see Alois Troller, "Réflexions sur 'l'Urheberpersönlichkeitsrecht'", 73 Le Droit d'auteur 304 (1960).

96. Roger-Ferdinand, "L'affaire 'Carmen Jones'", 8 Rev. int. dr. aut. 3, 21 (1955).

97. See Geller, "International Copyright: An Introduction", supra note 2, at sec. 6[2][c].

98. Germi c. Reteitalia e Rizzoli Film, Tribunale, Rome, 30 May 1984, 56 Il Diritto di Autore 68 (1985), reversed Corte di Appello, Rome, 16 Oct. 1989 (cited supra note 54 and critiqued in the accompanying text). For an ironic commentary, see Maurizio Nichetti, The Icicle Thief (Bambu Productions, 1989). For the legislative solution in Italy, see Legge no. 223 of 6 Aug. 1990, art. 8.

99. For different points of view on performing old music, see Authenticity and Early Music: A Symposium (N. Kenyon, ed. 1988).

100. See Paul Edward Geller, "The Universal Electronic Archive: Issues in International Copyright", 25 I.I.C. (forthcoming 1994).

101. For an example, see David Sanjek, "'Don't Have to DJ No More': Sampling and the 'Autonomous' Creator", 10 Cardozo Arts & Ent. Law J. 607, 622-623 (1992).

102. For the advantages of such references generally, see George P. Landow, Hypertext: The Convergence of Contemporary Critical Theory and Technology, esp. ch. 2 (1992).

103. For these tensions, see supra text accompanying notes 44-48.

104. See Paul Edward Geller, "The Proposed EC Rental Right: Avoiding some Berne Incompatibilities", 1992 Euro. Intell. Prop. Rev. 4, at 5-6.

105. See Bappert, supra note 15, at 37-38, 113-114.

106. See Geller, "Copyright in Factual Compilations", supra note 47, at 804-807.

107. For further critical analysis, see Cherpillod, supra note 61, at part 1.

108. For comparable analyses, see Ulmer, supra note 46, at 119-141; Goldstein, Copyright: Principles..., supra note 73, at vol. 1, sec. 2.3, and vol. 2, ch. 8; Rotstein, supra note 72, at 741-752.

109. For a different analysis of mixed cases, but one leading to comparable results, see Wendy J. Gordon, "A Property Right in Self-Expression: Equality and Individualism in the Natural Law of Intellectual Property", 102 Yale Law J. 1535, 1592-1605 (1993).

110. Rogers v. Koons, 751 F. Supp. 474 (S.D.N.Y. 1990), affirmed, 960 F. 2d 301 (2nd Cir. 1992).

111. See Martha Buskirk, "Appropriation Under the Gun", 80 (no. 6) Art in America 37 (1992).

112. Loi de 1957, art. 41 (CPI, art. L. 122-5).

113. Wolf c. Mme. Vlamynck, Cour d'appel, Paris, 20 Dec. 1990, 1991 Dalloz (Jur.) 532, followed by a note by B. Edelman.

114. Bernard Edelman, "Le personnage et son double", 1980 (no. 30) Dalloz (Chronique) 225, at 229.

115. Jorge Luis Borges, "The Library of Babel", in Ficciones 79 (A. Kerrigan, ed. 1962) (quoted matter found in following pages).