*79 WHEN AUTHORS WON'T SELL:

PARODY, FAIR USE, AND EFFICIENCY IN COPYRIGHT LAW

Originally published at: 62 U. Colo. L. Rev. 79 (1991)

Alfred C. Yen [FNa]

Copyright 1991 Alfred C. Yen, all rights reserved

I. INTRODUCTION

This article considers whether the fair use treatment of parody [FN1] in copyright law [FN2] is economically efficient. [FN3] The inquiry is undertaken *80 because it provides a good illustration of how the failure of economists' assumptions prevents efficiency theorists from completely explaining the results of traditional copyright intuition. [FN4] Awareness of this phenomenon is important in light of the rising popularity of efficiency theories in copyright, [FN5] and it also sheds light on the future of *81 copyright analysis. [FN6]

Courts have traditionally viewed copyright as an instrument for encouraging creative activity. [FN7] By giving authors the ability to prevent others from reproducing, distributing, performing, displaying, or basing new works on original material, [FN8] copyright creates a property right which authors may exploit commercially. [FN9] The possibility of realizing such financial gains gives authors incentives to create new works from which the public may benefit. [FN10]

Of course, granting these benefits is not without costs, for the creation of new works is heavily dependent on the ability of new authors to borrow from already existing materials. [FN11] If copyright law outlawed all borrowing from existing works, the creative process would surely grind to a halt, thereby depriving the public of the very benefits copyright should secure. Copyright law therefore limits the duration of an author's copyright [FN12] and allows a significant degree of borrowing from every work even while copyright subsists. [FN13]

*82 This basic structure often leads courts to determine the extent of an author's copyright rights by engaging in an intuitive cost­benefit analysis. Courts ask whether, on balance, granting the author's claim will foster more creativity than it hinders. [FN14] Judicial intuition necessarily plays a strong role in this calculation because courts seldom have hard evidence about either the quantity or quality of the creative activity that copyright fosters. [FN15]

In general, this intuitive cost­benefit analysis does a reasonable job of explaining the basic shape of copyright law. After all, most people would agree that the public is better off with some sort of copyright law than with none at all. [FN16] Most would also agree that doctrines that limit copyright's reach are necessary to keep copyright from choking the very creativity it should foster. [FN17] However, when it comes to copyright's details, the clarity of intuitive cost­benefit reasoning declines. It is difficult, if not impossible, to reach firm conclusions about whether copyright should prevent individuals from copying the structure of computer programs, [FN18] imitating an author's style, [FN19] or *83 videotaping television programs off of the air. [FN20] These problems arise, at least in part, because courts have neither articulated a firm vision of the public welfare nor devised a method for maximizing it. Instead, we are left with a collection of results in search of truly coherent principles.

This problem explains the growing use of economic efficiency in copyright, [FN21] for efficiency theory provides both a definition of the public welfare and an unambiguous method for improving it. As an initial matter, the efficiency theorist defines the public welfare in terms of a single variable, usually money. [FN22] This leads quickly to a method for improving and maximizing public welfare. Since dollars are quantifiable and measurable, the optimal copyright regime may be expressed as that regime which maximizes the amount by which copyright's benefits exceed its costs. [FN23]

The prospect of applying efficiency to copyright is tempting indeed. If efficiency's deduction could somehow replace copyright's intuition, copyright would take on an intellectual rigor rarely seen in law. Empirical cost­benefit calculations would identify improvements to copyright law. Instead of making ad hoc intuitive judgments about how much copyright best serves the public interest, copyright analysts could deductively prove that a given copyright doctrine actually maximizes society's welfare. [FN24] When combined with copyright's undeniable link to economics, these possibilities create pressure to immediately adopt efficiency as copyright's primary analytical tool. [FN25]

However, brief reflection reveals that this adoption should not *84 take place without careful deliberation. Among other things, one must wonder about the economists' assumption that social welfare is best expressed as money. People certainly do not want to sell everything they have. One would certainly be unlikely to sell one's kidneys. In copyright, authors often form emotional attachments to their works which are nonpecuniary in nature. Expressing these interests in money terms would seem, at the very least, to raise the risk that efficiency will mischaracterize the social value of these interests [FN26] and may therefore prove hostile to copyright doctrines which enjoy wide support. [FN27] Thus, efficiency should not replace intuition until we are certain that efficiency will support well­established copyright doctrines. [FN28]

The foregoing sets the stage for this article's inquiry into whether copyright's present fair use treatment of parody is efficient. First, analysts have suggested that the fair use doctrine and its application to parody promote efficiency. [FN29] Second, parody's fair use treatment [FN30] is the product of a traditional intuitive cost­benefit analysis. [FN31] The comparison of this reasoning to efficiency analysis therefore illuminates the general use of efficiency as a positive theory of copyright. Third, we can be confident that copyright law should treat parody as a fair use. A long string of cases commits the courts to the principle that parodists are entitled to borrow more from an author's work than other users. [FN32] Congress has specifically supported this trend by listing parody *85 as one of the uses which courts should defend under legislative codification of the fair use doctrine. [FN33] Scholars have contributed numerous articles which advocate the maintenance and extension of parody as a fair use. Thus, efficiency ought to account for parody's present treatment if it is a truly effective theory of copyright. [FN34]

Accordingly, the article begins by developing the traditional cost­ benefit analysis which supports parody's fair use treatment. The purpose behind this effort is to explain, but not necessarily endorse, the reasoning behind the present treatment of parody by copyright law. Next, the article develops an efficiency based model for the fair use doctrine and attempts to use that model to explain copyright's treatment of parody. The article pays particular attention to whether the traditional cost­benefit analysis is truly analogous to an efficiency analysis as well as whether parody's fair use treatment is in fact efficient. The article discovers that efficiency explains some, but not all, of the decisions that have been made by the courts, and that it is precisely efficiency's reliance upon money that frustrates the article's attempt. The article concludes with some observations about the future of copyright analysis.

II. THE TRADITIONAL COST­BENEFIT ANALYSIS

Presently, authors [FN35] own five exclusive rights in their copyrighted works. They alone control reproduction, distribution, public performance, and public display of the copyrighted original. [FN36] The code also grants them control over the preparation of derivative works from the *86 copyrighted original. [FN37] Individuals who exploit these rights without the author's permission may face either civil [FN38] or criminal [FN39] prosecution for copyright infringement. [FN40] Together, these provisions suggest that authors should be able to control production of parodies based on their copyrighted works.

As an initial matter, it is clear that parodies infringe the author's right to control creation of "derivative works." Since parodies always borrow and recast the plot, words, and style of a preexisting original, they necessarily fall under the definition of derivative work. [FN41] Furthermore, such a finding of infringement would seem to serve the basic goal of copyright. Parodists, like many authors, work in large part because they are paid for their creations. In fact, a quick look at the reported parody cases shows that the defendants in those cases generally achieved broad commercial dissemination of their parodies, thereby reaping generous remuneration. [FN42] Treating parodies as infringements would simply force parodists to use the money they earn to buy the necessary rights from authors. In turn, this would increase authors' financial incentives to create new material. [FN43]

Of course, courts have not applied copyright law to parody in this fashion. Instead of consistently treating parodists as copyright infringers, courts have given parodists considerable license to borrow freely from copyrighted works. [FN44] The parodist need only satisfy two threshold *87 requirements. First, her work must contain some criticism of the original work. [FN45] Second, her work must not function as a substitute for the underlying work. [FN46] Once these requirements have been satisfied, the parodist may freely borrow the material that is "reasonably necessary to conjure up the original." [FN47] The explanation for this result *88 lies in the courts' intuitive application of the so­called "fair use doctrine." [FN48]

In its presently codified form, [FN49] that doctrine excuses otherwise infringing use of a copyrighted work as "fair" if the court resolves four factors in the defendant's favor: 1) the purpose and character of the use, 2) the nature of the copyrighted work, 3) the amount of the copyrighted work used, and 4) the effect of the use on the potential market or value of the copyrighted work. [FN50] Not surprisingly, these factors are generally evaluated by referring to copyright's basic purposes. [FN51] Courts tend to find a given use fair when they believe that a finding of infringement would be inconsistent with copyright's promotion of social welfare. [FN52] Thus, a cost­ benefit analysis of the social consequences *89 associated with parody's fair use treatment should explain the present state of the law. [FN53]

Construction of such an analysis requires the assessment and comparison of three consequences associated with the fair use treatment of parody. First, the public would gain access to a popular form of humor. [FN54] Second, the public would benefit from the critical perspectives of the parodist. [FN55] Third, authors would lose the opportunity to commercially exploit their works through parody. [FN56]

*90 Our analysis begins by estimating the extent to which society can enjoy the benefits of parody without granting parodists favorable fair use treatment. As noted previously, parodists often achieve strong commercial rewards for their work. [FN57] The availability of such financial resources suggests that parodies would be widely available even without generous fair use treatment. After all, producers who turn novels into movies are not driven out of business by having to buy rights from novelists. It would therefore seem that parodists would also be able to buy similar rights from the authors whose works will be used. Fair use treatment for parody would therefore do little to change the composition of public discourse or advance the public interest. [FN58]

However, closer examination reveals that the sale of movie rights is vastly different from the sale of parody rights. When an author parts with his movie rights, he gains more than money. If the movie is successful, the author's reputation grows. Consumers who have never read the author's work will form favorable opinions of it. They may even buy the book itself. By contrast, the sale of parody rights exposes the author to risks that few authors desire. If the parody is successful, the author's reputation does not grow. Instead, the author and his work become the target of the parodist's humor. [FN59] This result makes it highly unlikely that an author will sell parody rights to his work at any price. As an initial matter, parodists who inquire about the availability of parody rights generally meet the reply that the rights are simply not for sale. [FN60] Furthermore, authors who sue parodists often seem as concerned with stopping unflattering references to their work as they are with any financial loss. [FN61]

The general refusal of authors to sell their parody rights makes it highly likely that parodies would be rare without fair use treatment. The public would therefore enjoy few, if any, of parody's benefits in *91 the absence of treating parody as a fair use. [FN62] However, this does not necessarily mean that fair use treatment for all parodies is warranted, for different types of parodies bring different benefits to the public.

The most obvious link between parody and the public interest is parody's humor. The public generally consumes parodies because they generate a good laugh. Humor is certainly an important part of society, and it provides some reason to encourage the creation of parodies. [FN63] However, parody's link to humor is not, in and of itself, a terribly good reason to encourage its creation. If nothing else, there is the danger that fair use treatment for humorous use could become an invitation to open ended misappropriation of copyrighted material. Too much unauthorized borrowing might erode the financial incentives that presently encourage the authors of original source material and offset whatever public benefits stem from parody's humor. [FN64] Moreover, since good sources of humor would exist with or without parody, one might argue that parody's fair use treatment gives society relatively little in return for putting copyright's basic structure at risk. [FN65]

Fortunately, a successful parody does much more than just give the public a good laugh. It also provides a very unique criticism of literary and social foibles. The twisting and ridicule which forms the heart of good parody enables the public to experience first hand the silliness of literary and social conventions. This combination of laughter and criticism is unmatched by any other artistic genre, and it carries *92 the potential to enrich public discourse and precipitate social change. Effective parodies therefore offer the public a very special and unique benefit in exchange for any resulting risk to copyright's underlying system of incentives. [FN66]

The foregoing suggests that our cost­benefit assessment depends on the sort of parody that the fair use doctrine encourages. If fair use spurs the creation of parodies whose value lies strictly in humor, the benefits to society appear equivalent to those provided by other forms of humor. However, if fair use stimulates the creation of parodies that criticize literary and social conventions, society gains a unique benefit. Thus, the case for treating parody as a fair use is much stronger when the parody in question involves criticism of some sort. [FN67] Comparing the losses associated with parody against the gains reinforces this conclusion.

As the reader will recall, the fair use treatment of parody prevents the author from exploiting her own work through this medium. [FN68] This ostensibly reduces the financial return the author can reap from her work, thereby harming the financial incentives of copyright law. However, it must be noted that authors can reap financial incentives only from those uses that they reasonably could be expected to exploit. [FN69] It therefore stands to reason that copyright need not protect uses that authors have no intention of exploiting, for protecting those uses would have no effect on copyright's financial incentives.

Normally, allowing individuals to borrow freely from copyrighted works erodes the incentives for the creation of new original material. For example, if television producers were allowed to turn novels into *93 sitcoms for no charge, the number of novels written would presumably decrease, thereby harming the public interest. Copyright law avoids this result by forcing producers to use the money raised through advertising to pay novelists for the rights to use their works. These payments stimulate further writings. [FN70]

However, as noted above, the sale of parody rights is vastly different from the sale of other derivative rights. Authors anxious to avoid being humiliated will seldom, if ever, voluntarily expose their own work to a critical parody. [FN71] By contrast, if a proposed parody were merely humorous, but not critical, there is no particular reason why an author might not license the use. [FN72] This dichotomy has powerful consequences for parody's possible erosion of copyright incentives. As an initial matter, it shows that fair use treatment for parodies that are critical of the original will not generally affect the financial returns which spur creative activity. [FN73] Conversely, it shows that parodies whose value lies primarily in humor are poor candidates for fair use treatment because their limited value [FN74] is offset by a very real risk to copyright incentives. [FN75]

The foregoing provides a good intuitive cost­benefit explanation of why courts treat parody as they have under the fair use doctrine. *94 The threshold requirement that a parody must criticize the underlying original [FN76] makes sure that parody's fair use treatment reaps a unique and positive benefit for the public in return for any possible harm to copyright's system of financial incentives. Similarly, the requirement that parodies not serve as substitutes for originals [FN77] ensures that those incentives are not seriously harmed by the fair use doctrine. In situations where both of these conditions have been met, it seems clear that society has much to gain [FN78] and little to lose from granting parodists special permission to borrow from copyrighted works. Thus, courts allow parodists whose works pass the threshold conditions to borrow whatever amount is "reasonably necessary to conjure up the original." [FN79] This explanation sets the stage for our examination of whether economic efficiency really does capture conventional copyright analysis.

III. PARODY AS EFFICIENCY

A. Concepts of Efficiency

Two closely related concepts of efficiency are critical to our analysis. The first concept is Pareto efficiency. [FN80] A Pareto efficient state of affairs is one in which no individual can be made better off without inflicting a loss on at least one other person. Thus, a situation P is Pareto superior to situation Q if all individuals in society are as well or better of in P than they are in Q. [FN81]

The second concept is Kaldor­Hicks efficiency. A Kaldor­Hicks efficient state of affairs is one in which no individual or group of individuals can be made better off without inflicting losses on others which exceed the amount of the gains. Thus, situation X is Kaldor­Hicks superior to situation Y as long as the gains of those who are better off in situation X can fully compensate those who are worse off than they were in situation Y. [FN82] Brief reflection reveals how both of these concepts could be used by a court analyzing the fair use doctrine.

A fundamental proposition of modern economics is that the self­ *95 interested actions of individuals creates a Pareto efficient state of affairs. This proposition flows from the assumption that rational [FN83] individuals will pursue what they subjectively [FN84] believe is in their self interest. This pursuit leads to a series of mutually beneficial transactions with other individuals which continues until all possibility of such exchange is exhausted. When this result is reached, no individual can improve her welfare without harming the welfare of another. The result is therefore Pareto efficient. [FN85] More importantly, economic theory also equates the attainment of Pareto efficiency under perfect market conditions [FN86] with a socially optimal state of affairs. [FN87]

For example, suppose that the author of a comic strip is interested in making an animated film based upon his comic strip characters. The cartoonist correctly perceives that the public would pay a *96 total of $500,000 to see the film and that it would cost him $450,000 to make the film. Suppose further that a movie studio is also interested in making the animated film and that they could produce it for a cost of $440,000. In this situation, the process of voluntary exchange will improve social welfare.

Since the studio can produce the movie for only $440,000, it stands to improve its welfare by $60,000. [FN88] Of course, the studio cannot reap this profit without the cartoonist's cooperation. The studio will therefore try to buy the necessary rights from the cartoonist for a price as high as $59,999, for any price in this range will still allow the studio to produce the movie profitably. At the same time, the author realizes that the most he can profit by making the movie alone is $50,000. [FN89] He therefore will sell the necessary rights to the movie studio if the studio offers him more than $50,000.

The foregoing analysis implies that the movie studio will buy the necessary rights from the cartoonist for a price between $50,001 and $59,999. Such an exchange would improve the welfare of both the studio and the cartoonist while satisfying the public's demand for the movie. This result represents the improvement of social welfare. The author is better off because the money offered by the studio is more than the author could earn by making the movie himself. The studio is better off because the exchange produced a profit that the studio could not have realized alone. The public is also better off because it values the experience of watching the film at something greater than $500,000.

The foregoing economic observations suggest that courts can promote the public interest by interpreting laws to promote Pareto efficiency. The basic judicial posture is simple. Courts should presumptively enforce the status quo because voluntary exchange leads to Pareto efficiency, which in turn represents a social optimum. [FN90] Thus, any interference with the product of such exchange must be inefficient and presumptively undesirable. [FN91] This reasoning governs as long as perfect market conditions exist.

If, however, perfect conditions do not exist (i.e. there is a "market *97 failure"), the court's task becomes more complicated, for the outcome of voluntary exchange under imperfect conditions is not necessarily socially optimal. In the example given previously, suppose now that it costs the studio $10,000 to conclude any sale of the necessary rights. [FN92] In this situation, the sale from cartoonist to studio will not happen. Since the studio is willing to spend up to $59,999 to acquire the movie rights, the $10,000 transaction cost prevents the studio from offering the author a sufficiently high purchase price. [FN93] Although the movie will still be made, [FN94] society's total wealth suffers because neither the cartoonist nor studio is as well off as either would have been in the absence of transaction costs. In particular, the studio reaps no profit while the cartoonist's profit is diminished. [FN95]

The foregoing example shows that courts confronted with imperfect markets should not necessarily enforce the status quo. Since the market no longer presumptively maximizes social welfare, courts must independently determine whether the status quo mirrors the result of the free market. Welfare economists generally solve this problem by making two key assumptions. First, economists generally assume that individuals value all gains and losses according to a single scale, usually money. [FN96] Second, they assume that money can always be redistributed among individuals without cost. [FN97]

Together, these assumptions enable the economist to promote Pareto efficiency by using the Kaldor­Hicks concept. Since all gains and losses may now be measured in dollars, the economist can perform a cost­benefit analysis to evaluate any proposed change in the status quo. She simply measures and compares how much the affected parties are willing to pay or accept to cause or allow the change. [FN98] If the "winners" could pay a subjectively adequate sum of money to the *98 "losers" and still improve their welfare, [FN99] the economist recommends that the change be made. [FN100] The economist need not worry about whether the compensation is actually paid because of the assumption that government can costlessly make the necessary cash transfers from the winners to the losers. [FN101] This process can continue until no Kaldor­Hicks superior moves remain. [FN102]

*99 The foregoing suggests a method by which a court could use the fair use doctrine to promote Pareto efficiency. First, a court should presumptively enforce an author's proprietary rights as defined by the copyright statute. Second, the court should only consider a fair use defense if it can identify a failure of perfect market conditions. Third, the court should declare the use fair (i.e. permit the borrowing) only if society's gains through fair use outweigh any losses imposed on authors. [FN103] The possibility of using efficiency to explain copyright intuition may therefore be tested by examining the correspondence between this analysis and the traditional analysis constructed in the previous section.

B. The Efficiency Explanation of Parody as Fair Use

Our analysis begins on a promising note, for the intuitive cost­benefit treatment of parody bears strong similarities to the first step of the Pareto efficiency approach. As noted previously, courts initially ruled that parodies infringed the rights of the copyright holder. [FN104] Thus, parodists were presumptively forced to buy rights from authors like any other derivative work user. [FN105] This result conforms to the basic presumption that voluntary exchange promotes Pareto efficiency and leads to a social optimum.

Of course, courts eventually changed this status quo in favor of a *100 rule that allows a parodist to borrow more from a work than is generally contemplated under copyright law. [FN106] This development rested upon the observation that parody was valuable "both as entertainment and as a form of social and literary criticism." [FN107] Courts paid special attention to this fact because it implied that the fair use treatment of parody would advance social welfare in a special and unique way. [FN108] This allowed courts to distinguish parody from other types of derivative uses that were undeserving of fair use treatment and set the stage for the intuitive cost­benefit analysis at the heart of the fair use doctrine. Interestingly, parody's unique social value also suggests that the economist's basic presumption about voluntary exchange may not be justified. Application of the fair use doctrine to parodies therefore seems economically plausible.

As noted earlier, the economist generally assumes that individuals who want to create derivative works will raise sufficient money to buy the necessary rights from the author as long as those rights are more valuable in the borrower's hands than in the author's. [FN109] This assumption seems particularly strong when commercial exploitation of the underlying work is involved because the public generally indicates its estimation of the derivative work's value by the price that the public is willing to pay to see the derivative work. [FN110] However, a successful parody does more than merely entertain. It also gives its consumers a humorous and critical perspective on both the work being parodied and society at large. The strength of this contribution increases as more people view the parody, and it is difficult, if not impossible, to achieve this through any other art form. To the extent that social opinions change as a result of this process, even those who did not view the parody benefit from its existence. Society thereby reaps a tangible benefit by having a more educated and intellectually active citizenry. [FN111]

This tangible but diffuse benefit associated with parody violates the general assumption that a derivative work's social value is reflected in the price the public is willing to pay to see the work. [FN112] Parodists *101 sell and consumers buy parodies primarily for their entertainment value. [FN113] Only rarely will someone buy a parody expressly for its criticism of the work being parodied. Furthermore, consumers probably never buy parodies because they want to subsidize the education of the citizenry at large. Thus, the money a parodist can offer an author for the purchase of parody rights is likely to understate the true social value of his work. [FN114] Since the true social value of parody is not reflected in the price paid for parodies, the unregulated market cannot be presumptively Pareto efficient. The economist would therefore no longer necessarily enforce the status quo, for there can be no assurance that the market allocates parody rights in a socially optimal fashion. [FN115] This explains why an economist would agree with the judicial decision to consider fair use treatment for those parodies that have value as literary criticism. [FN116]

The failure of the parody market's presumptive efficiency leads the economist to perform a cost­benefit calculation to see if the status quo should be changed under the Kaldor­Hicks criterion. As the reader will recall, the heart of the Kaldor­Hicks analysis is that a proposed change in the status quo promotes efficiency only if the gains of the winners are sufficient to compensate the losers. [FN117] In the case of parody as fair use, the "winners" are parodists and the members of society who benefit from the criticism associated with parody. Their gain may be expressed as the money paid to parodists by consumers *102 plus the diffuse benefit society gets from parody's criticism. [FN118] This amount must be compared against the authors' losses, which may be expressed as the money value of their lost chance of licensing parodies plus any emotional or reputational harm suffered as a result of parody's ridicule. [FN119] A comparison of this analysis and the traditional reasoning that supports parody as a fair use reveals some important similarities and differences.

The traditional cost­benefit reasoning that supports parody's fair use status involved the intuitive evaluation of three items. On one hand, the courts considered the "gains" associated with parody's value as entertainment and as criticism. [FN120] On the other hand, the courts considered any "losses" inflicted by parodists on authors by reason of lost sales of the authors' works. [FN121] The courts then made sure that the gains outweighed the losses by allowing fair use only when parody's value as entertainment and criticism were high and the authors' lost sales were low. [FN122]

At first blush, it might appear that this intuitive cost­benefit analysis may be directly translated into an efficiency calculation. Closer inspection, however, reveals that this is not the case, for the economist performing a Kaldor­Hicks analysis evaluates four, and not three, items. Like the traditional analyst, the efficiency theorist considers parody's value as entertainment, its value as criticism, and the lost sales of the author. However, the efficiency theorist also must consider the losses parody's ridicule inflicts upon authors. This addition is required because the economist deals only with the subjective desires of the individual. [FN123] In the case of parody, it is the author's desire to avoid ridicule that is generally responsible for her reluctance to sell rights to parodists. Forcing authors to endure the criticism of parody without compensation therefore imposes a loss on authors. Failure to account for such harm is fundamentally inconsistent with the pursuit of Pareto efficiency.

The addition of the author's emotional loss in the Kaldor­Hicks efficiency calculation raises considerable doubt about whether parody's fair use treatment is efficient at all. As the reader will recall, the *103 Kaldor­Hicks test allows the fair use treatment of parody only if society gains more than authors lose. [FN124] Under a traditional intuitive analysis, it seems that this is the case because the conventional analyst ignores any non­pecuniary or emotional losses that parody inflicts on authors. The gains from parody are tangible, and authors lose few, if any, of the financial incentives which encourage their work. [FN125] However, an economist applying the Kaldor­Hicks test must compare the gains from parody against the authors' desire to be left free of parody's critical bite. This comparison would normally be performed by determining if the gains of society (the winners) were large enough to pay authors (the losers) subjectively adequate compensation. [FN126] This comparison has two possible outcomes with radically different consequences for an efficiency explanation for parody.

One possibility is that authors do not sell parody rights because parodists are unable to raise enough money to successfully bribe the author. If this were the case, then efficiency would remain a plausible explanation for parody's fair use treatment. If the courts attached a high enough dollar value to society's gains from parody, they would correctly determine that, under perfect conditions, parodists would be able to offer sufficient bribes to buy the necessary rights from authors. [FN127] When coupled with the assumption of costless compensation, fair use treatment for parody would then promote Pareto efficiency. [FN128]

The second possibility is that authors are simply unwilling to sell parody rights for any sum. This would occur if authors valued their emotional tranquility in a manner similar to their limbs. Thus, an author who is offered a large sum of money for her parody rights would respond in the same way that she would if offered a large sum of money for her legs. She would simply state, "They are not for sale." [FN129] If this turns out to be the case, the fair use treatment of parody can no longer be described as promoting Pareto efficiency. This is because a proposed change in the status quo is Pareto efficient only if *104 all members in society are subjectively as well as or better off than they were before the change. [FN130] If authors are forced to endure the barbs of parodists with no hope of compensation other than money, they will be subjectively worse off because they never wanted to sell rights to parodists at any price.

The foregoing analysis shows that the plausibility of the efficiency explanation for parody's fair use treatment boils down to an empirical question: Do authors subjectively prefer a certain sum of money to freedom from the emotional stress caused by parodies? Unfortunately for efficiency theorists, intuition and empirical evidence strongly suggest that the answer is "No."

The intuitive evidence comes from the traditional justification for parody's fair use treatment. Authors, like all individuals, do not like to be ridiculed. Given the fact that a person's reputation or emotional *105 tranquility cannot be restored with money once it has been destroyed, the chance that an author (or any individual) would allow the destruction of those things via parody is very slim. [FN131]

The empirical evidence comes from work done in the context of libel actions by Professors Bezanson, Soloski and Cranberg. [FN132] In their work, these scholars surveyed 164 plaintiffs in libel cases. [FN133] Among other things, the victims were asked how the libel had harmed them, what the media could have done to satisfy the plaintiffs' grievances, and what the plaintiffs asked the media to do. [FN134]

Not surprisingly, most plaintiffs claimed that libel had harmed them emotionally, financially, or both. [FN135] More importantly, the plaintiffs overwhelmingly stated that they did not consider money damages an adequate remedy for the harm they had suffered. [FN136] 72.9% stated that an apology or a retraction would suffice, while another 20% indicated that nothing at all would suffice. This trend was reflected in the demands the plaintiffs made to the media as well. Only 0.8% of the plaintiffs requested money damages, while 79.7% asked for retractions, apologies, or corrections. [FN137] These statistics shed a good deal of light on the possibility of using efficiency to explain parody's fair use treatment.

In particular, the questions posed to the libel plaintiffs address issues that are similar to those involved in a copyright action against a parodist. A libel victim sues because statements made to the public portray the victim in an unflattering light. The victim suffers through the ridicule and scorn heaped upon him by those who have seen the *106 libelous statement. Similarly, an author sues a parodist because she resents the public portrayal of her work in an unflattering light. She suffers because those who have seen the parody hold her in lower esteem and laugh at her work. These similarities indicate that the answers given by libel plaintiffs are good evidence of how authors value their reputation and emotional tranquility against money. The statistics show that the overwhelming majority of those injured through criticism simply do not think that money compensates them for their losses. Indeed, the fact that 20% of the libel plaintiffs believed that no satisfactory remedy existed suggests that authors might never be subjectively compensated for the losses they suffer at the hands of parodists. This in turn makes it difficult, if not impossible, for an economist to defend parody's fair use treatment as efficient. [FN138]

IV. CONCLUSION

As the reader will recall, the power of efficiency analysis stemmed primarily from a single assumption, the reduction of value to money. [FN139] This assumption was important because it gave the efficiency analyst a single scale by which to measure the consequences of changing any status quo. In the context of parody, this assumption failed, and it became the reason that the article failed to construct an efficiency explanation for parody. This can best be seen by examining again the differences between the traditional and efficiency approaches to the parody question.

The traditional explanation for parody's fair use treatment rested in large part on the courts' refusal to count the harm of criticism on the author's emotional welfare in its intuitive cost­benefit analysis. [FN140] By contrast, the efficiency analyst was forced to take this harm into account. If authors had valued this emotional harm in terms of money, the construction of an efficiency explanation for parody's fair use treatment might well have been successful. Even though the author would have placed a high monetary value on her emotional tranquility, it is entirely possible that society would have valued its access to parody even more highly. Treating parody as a fair use would have been acceptable under the Kaldor­Hicks principle, for the efficiency theorist was entitled to assume that the money transfer needed to promote Pareto efficiency would be costlessly made. [FN141]

*107 Of course, it appears that authors did not value their emotional tranquility in terms of money. Since authors were apparently unwilling to sell their parody rights at any price, no sum of money could ever properly compensate the author for any losses suffered at the hands of parodists. This failure of compensation made justifying parody's fair use treatment under the Kaldor­Hicks criterion impossible. [FN142]

On a more general level, the foregoing suggests that future attempts to construct efficiency explanations for copyright doctrine are likely to fail whenever authors value their copyright rights for non­monetary reasons. In these situations, authors will consistently refuse to sell others the rights to use their works because money will not be an adequate substitute for any interests harmed by the contemplated use. For example, individuals sometimes use copyright to hinder biographers' use of their writings. This may be done because the biographer is hostile. [FN143] Other times the author may simply want privacy. [FN144] Similarly, an artist might sue those who use their works because they object to the defendants' alteration of the original artistic message, and not because they intend to exploit the work in the same manner themselves. [FN145] To the extent that courts have decided cases like these in favor of the defendants, the author's refusal to exchange her interest for money makes it impossible for the efficiency analyst to explain the result.

The importance of this for future copyright analysis is clear. No matter how hard we try, a complete efficiency explanation of copyright is unlikely because authors sometimes do not consider money an adequate substitute for their copyright rights. We must therefore develop non­efficiency copyright theories and apply them to explain at least some of copyright's intuitively supported results. This sort of inquiry leads in two possible directions. One possibility is that analysts will clearly identify the various considerations that make up copyright intuition. Once this has happened, the relative importance of these factors can be established. Efforts such as these will hopefully make *108 copyright intuition more predictable. [FN146]

Second, and perhaps more importantly, analysts might look to non­utilitarian theories of property for a different perspective on how far an author's copyright rights should extend. As this article has shown, efficiency's failure to explain certain facets of copyright law stems from its assumption that all preferences are equally worthy of legal protection. Those preferences may be violated only if subjectively adequate compensation is actually or theoretically given to the "loser." Since authors apparently do not always prefer money to certain fair use treatment of their works, this leaves the efficiency analyst without an adequate explanation for why copyright grants such fair use treatment anyway. By contrast, non­utilitarian theories based on individual rights may succeed where efficiency fails because they do not necessarily assume that all preferences are equally worthy of legal protection. These theories could therefore explain copyright's fair use treatment of works such as parody by directly evaluating whether the interests at stake fall within the definition of individual rights found in natural law, personality theory, or constitutional law. [FN147]

Of course, there is no guarantee that any of these efforts will ultimately provide a comprehensive theory of copyright. However, since we now know that copyright cannot be totally explained through efficiency, we also know that at least part of copyright's future rests in these relatively unexplored areas.

FNa B.S., M.S., Stanford University; J.D. Harvard Law School. Assistant Professor of Law, Boston College Law School. The author would like to thank Frank Upham, Ralph Brown, Wendy Gordon and Dave Sunding for their comments. Thanks are also owed to Tracy Tanaka and Kelly Cournoyer for their able research assistance. This article was made possible by the author's receipt of a Dr. Thomas Carney Law Faculty Research Grant from Boston College Law School and a Faculty Research Expense Grant from Boston College. Copyright 1990 by Alfred C. Yen. All rights reserved.

FN1. This article uses the term "parody" to encompass any humorous expression that is based on a preexisting work of authorship. By defining "parody" in such broad terms, the article includes works which might more accurately be considered satires, burlesques, or travesties. This is done to follow the courts' practice of treating these varied, but related, art forms as "parodies" under the copyright law. See Loew's Inc. v. Columbia Broadcasting Sys., 131 F. Supp. 165, 176 n.31 (S.D. Cal. 1955), aff'd sub nom. Benny v. Loew's Inc., 239 F.2d 532 (9th Cir. 1956), aff'd by an equally divided court sub nom. Columbia Broadcasting Sys. v. Loew's, Inc., 356 U.S. 43 (1958) ("It will be noted that the words 'burlesque', 'parody', 'satire' and 'travesty' are often used interchangeably."). For discussions of parody and other humorous art forms, see PARODIES: AN ANTHOLOGY FROM CHAUCER TO BEERBOHM­­AND AFTER 557­68 (D. MacDonald ed. 1960) (defining parody and setting forth a brief history); M. ROSE, PARODY//META­FICTION: AN ANALYSIS OF PARODY AS A CRITICAL MIRROR TO THE WRITING AND RECEPTION OF FICTION 17­55 (1979) (defining parody and distinguishing it from related forms of art). For brief introductions to the relationship between parody and copyright, see Goetsch, Parody as Free Speech­­ The Replacement of the Fair Use Doctrine by First Amendment Protection, 3 W. NEW ENG. L. REV. 39, 39­42 (1980); Netterville, Copyright and Tort Aspects of Parody, Mimicry and Humorous Commentary, 35 S. CAL. L. REV. 225, 225­29 (1962).

FN2. Copyright law gives the author, among other things, exclusive rights to the preparation of works which are based on her original material. 17 U.S.C. ß 106 (1988). Under this scheme, parodists would normally have to obtain permission from the author of any work they wanted to parody. Goldstein, Derivative Rights and Derivative Works in Copyright, 30 J. COPYRIGHT SOC'Y 209, 234 (1983) ("Parodies and satires represent derivative uses no less than dramatizations, abridgments or other arrangements of the underlying work."). See also Benny v. Loew's Inc., 239 F.2d 532 (9th Cir. 1956) (holding parodist liable for violation of copyright holder's exclusive rights), aff'd by an equally divided court sub nom. Columbia Broadcasting Sys. v. Loew's, Inc., 356 U.S. 43 (1958). However, the courts have frequently excused the otherwise infringing actions of the parodist under the so­called fair use doctrine. See 17 U.S.C. ß 107 (1988) (codification of the fair use doctrine); Fisher v. Dees, 794 F.2d 432 (9th Cir. 1986); Elsmere Music, Inc., v. National Broadcasting Co., 482 F. Supp. 741 (S.D.N.Y.), aff'd, 623 F.2d 252 (2d Cir. 1980); Berlin v. E.C. Publications, Inc., 329 F.2d 541, 545 (2d Cir.) ("For, as a general proposition, we believe that parody and satire are deserving of substantial freedom . . . . ") (emphasis in original), cert. denied, 379 U.S. 822 (1964). For further discussion of the fair use doctrine, see infra notes 49­53 and accompanying text.

FN3. Efficiency theorists generally refer to two different, but related, concepts of efficiency. The first concept, Pareto efficiency, describes a state of affairs in which no person can be made better off without inflicting a loss on at least one other person. J. COLEMAN, MARKETS, MORALS AND THE LAW 97­98 (1988). The second concept, Kaldor­Hicks efficiency, describes a state of affairs in which some individual has been made better off by a redistribution, to a point where he could hypothetically fully compensate those whose condition was worsened, with a net gain in welfare. Id. at 98. For further discussion of these concepts and their relation to copyright, see infra notes 80­103 and accompanying text.

FN4. Cf. Cooter, Book Review, Law and the Imperialism of Economics: An Introduction to the Economic Analysis of Law and a Review of the Major Books, 29 UCLA L. REV. 1260, 1264­65 (1982) (contrasting deductive nature of economic reasoning with inductive nature of traditional legal reasoning).

FN5. See, e.g., Landes & Posner, An Economic Analysis of Copyright Law, 18 J. LEGAL STUD. 325 (1989) (presenting a mathematically oriented efficiency model of major copyright doctrines); Gordon, Fair Use as Market Failure: A Structural and Economic Analysis of the Betamax Case and its Predecessors, 82 COLUM. L. REV. 1600 (1982) (presenting efficiency based model of the fair use doctrine); Menell, An Analysis of the Scope of Copyright Protection for Application Programs, 41 STAN. L. REV. 1045 (1989) (analyzing extent of copyright protection for computer programs under an efficiency analysis); Fisher, Reconstructing the Fair Use Doctrine, 101 HARV. L. REV. 1659 (1988) (using efficiency to shape fair use doctrine). The rising importance of efficiency as a copyright norm can also be seen in a recently published treatise by a leading copyright scholar, Professor Paul Goldstein. P. GOLDSTEIN, COPYRIGHT: PRINCIPLES, LAW AND PRACTICE (1989). Although Professor Goldstein does not explain all of copyright in efficiency terms, efficiency certainly plays a major role in many key areas of his work, particularly in his discussion of fair use. Id. at 194­95 (describing the major approaches to fair use as solutions to the problem of transaction costs).

The popularity of efficiency analysis should come as no surprise. The Constitution grants Congress the power "[t]o promote the Progress of Science and useful Arts, by securing for limited Times to Authors and Inventors the exclusive Right to their respective Writings and Discoveries." U.S. CONST. art. I, ß 8, cl. 8. Courts generally construe this clause as evidence that copyright rests on a strictly economic foundation. See Sony Corp. of Am. v. Universal City Studios, Inc., 464 U.S. 417, 429 (1984); Mazer v. Stein, 347 U.S. 201, 219 (1954). See also Breyer, The Uneasy Case for Copyright: A Study of Copyright in Books, Photocopies, and Computer Programs, 84 HARV. L. REV. 281, 284­91 (1970) (arguing that economics is the only defensible basis for copyright law). However, this impressive array of opinion has not eliminated other theories of copyright. Some analysts have claimed that natural rights theory provides copyright's true theoretical basis. Kauffman, Exposing the Suspicious Foundation of Society's Primacy in Copyright Law: Five Accidents, 10 COLUM. J.L. & ARTS 381 (1986). Others have argued convincingly that economics alone provides an incomplete basis for copyright law. See Gordon, An Inquiry into the Merits of Copyright: The Challenges of Consistency, Consent and Encouragement Theory, 41 STAN. L. REV. 1343, 1351 (arguing that "'wealth maximization,' as an aggregative criterion that disregards the possibility of independently derived individual rights, cannot serve as an acceptable foundation for the initial assignment of entitlements."); Yen, Restoring the Natural Law: Copyright as Labor and Possession, 51 OHIO ST. L.J. 517 (1990) (pointing out indeterminacy of economic principles and apparent presence of natural law reasoning in copyright). This dissatisfaction with a solely economic copyright theory is further reflected in recent explanations of copyright which draw little inspiration from economic reasoning. See Litman, The Public Domain, 39 EMORY L.J. 965 (1990); Hughes, The Philosophy of Intellectual Property, 77 GEO. L.J. 287 (1988). See also Brown, Eligibility for Copyright Protection: A Search for Principled Standards, 70 MINN. L. REV. 579 (1985) (advocating the combination of economic and authors' rights approaches). In light of these arguments, this article bases its analysis in economics for the purpose of inquiry, and not as an endorsement of the view that copyright analysis should be conducted solely as a matter of economics. For further criticism of efficiency analysis, see infra note 102.

FN6. See infra notes 146­147 and accompanying text.

FN7. See supra note 5.

FN8. 17 U.S.C. ß 102 (1988) (extending copyright protection to "original works of authorship"); 17 U.S.C. ß 106 (defining five exclusive rights of copyright holders).

FN9. Cf. Calabresi & Melamed, Property Rules, Liability Rules, and Inalienability: One View of the Cathedral, 85 HARV. L. REV. 1089, 1092 (1972):

An entitlement is protected by a property rule to the extent that someone who wishes to remove the entitlement from its holder must buy it from him in a voluntary transaction in which the value of the entitlement is agreed upon by the seller.

FN10. See Sony Corp. of Am. v. Universal City Studios, Inc., 464 U.S. 417, 429 (1984): The monopoly privileges that Congress may authorize are neither unlimited nor primarily designed to provide a special private benefit. Rather, the limited grant is a means by which an important public purpose may be achieved. It is intended to motivate the creative activity of authors and inventors by the provision of a special reward, and to allow the public access to the products of their genius after the limited period of exclusive control has expired.

See also THE FEDERALIST No. 43, at 279 (J. Madison) (E.M. Earle ed. 1976) ("[t] he utility of [the copyright] power will scarcely be questioned. . . . The public good fully coincides in both cases with the claims of individuals").

FN11. See Emerson v. Davies, 8 F. Cas. 615, 619 (C.C.D. Mass. 1845) (No. 4,436):

In truth, in literature, in science and in art, there are, and can be, few, if any, things, which, in an abstract sense, are strictly new and original throughout. Every book in literature, science and art, borrows, and must necessarily borrow, and use much which was well known and used before. . . . No man writes exclusively from his own thoughts, unaided and uninstructed by the thoughts of others. The thoughts of every man are, more or less, a combination of what other men have thought and expressed, although they may be modified, exalted, or improved by his own genius or reflection. See also Litman, supra note 5, at 1007­13; Landes & Posner, supra note 5, at 332.

FN12. 17 U.S.C. ß 302 (1988) (generally limiting duration of copyright to the life of the author plus fifty years).

FN13. Three major doctrines accomplish this result. The most prominent doctrine is the idea/expression dichotomy, which expressly withholds copyright protection from the ideas contained in every work. 17 U.S.C. ß 102(b); Baker v. Selden, 101 U.S. 99 (1880); Nichols v. Universal Pictures Corp., 45 F.2d 119 (2d Cir. 1930), cert. denied, 282 U.S. 902 (1931). The useful article doctrine works in a similar fashion by denying copyright protection to utilitarian aspects of copyright works. See Esquire, Inc. v. Ringer, 591 F.2d 796 (D.C. Cir. 1978), cert. denied, 440 U.S. 908 (1979) (denying copyright to overall design of lighting fixture); Carol Barnhart, Inc. v. Economy Cover Corp., 773 F.2d 411 (2d Cir. 1985) (denying protection to mannequins used to display clothing). Lastly, the fair use doctrine operates by excusing otherwise infringing activity when a finding of infringement would be inequitable or inconsistent with the general purposes of copyright. 17 U.S.C. ß 107. For further discussion of the fair use doctrine, see infra notes 49­53 and accompanying text.

FN14. Whelan Assocs., Inc. v. Jaslow Dental Laboratory, Inc., 797 F.2d 1222, 1235 (3rd Cir. 1986), cert. denied, 479 U.S. 1031 (1987), clearly states this view:

[P]recisely because the line between idea and expression is elusive, we must pay particular attention to the pragmatic considerations that underlie the distinction and copyright law generally. In this regard, we must remember that the purpose of the copyright law is to create the most efficient and productive balance between protection (incentive) and dissemination of information, to promote learning, culture and development.

See also Menell, supra note 5, at 1065.

FN15. See Litman, supra note 5, at 998. Concrete information about the amount and kind of works which best serve the public interests are even harder to come by.

FN16. But see Breyer, supra note 5 (raising doubt about value of copyright).

FN17. See Litman, supra note 5, at 1015­23; M. NIMMER & D. NIMMER, NIMMER ON COPYRIGHT ß 13.05 (3d ed. 1989) (stating that fair use doctrine "permits courts to avoid rigid application of the copyright statute when, on occasion, it would stifle the very creativity that law is designed to foster").

FN18. Compare Whelan Assocs., 797 F.2d at 1222 (extending copyright protection to sequence structure and organization of computer programs) with Plains Cotton Coop. Ass'n v. Goodpasture Computer Serv., Inc., 807 F.2d 1256 (5th Cir.), cert. denied, 484 U.S. 821 (1987) (declining to adopt Whelan).

FN19. Compare Roth Greeting Cards v. United Card Co., 429 F.2d 1106 (9th Cir. 1970) (basing infringement on general appearance and style of greeting cards) with Aliotti v. R. Dakin & Co., 831 F.2d 898 (9th Cir. 1987) (declining to find copyright infringement despite general similarity in appearance between stuffed dinosaur toys). For further discussion of the doctrinal and constitutional issues raised by these cases and copyright's general ambiguity, see Yen, A First Amendment Perspective on the Idea/Expression Dichotomy and Copyright in a Work's "Total Concept and Feel," 38 EMORY L.J. 393 (1989).

FN20. Sony Corp. of Am. v. Universal City Studios, Inc., 464 U.S. 417, 454­ 56 (1984) (5­4 decision declaring the off air videorecording of television programs for timeshifting purposes a fair use).

FN21. For examples, see supra note 5.

FN22. See Posner, Utilitarianism, Economics, and Legal Theory, 8 J. LEGAL STUD. 103, 119 (1979) (measuring welfare as "the value in dollars or dollar equivalents . . . of everything in society"). As a theoretical matter, economists could use a measure other than money to measure welfare. However, the practical necessity of measuring and comparing the value of disparate interests makes money a particularly attractive measure of welfare. See also infra note 96 and accompanying text.

FN23. Landes & Posner, supra note 5, at 326. See infra notes 98­102 and accompanying text for discussion of how this process of maximization might be accomplished.

FN24. Cf. Cooter, supra note 4, at 1264­65 (referring to deductive mathematical proof); Landes & Posner, supra note 5, at 341­42 (providing mathematical expression of copyright regime which maximizes social welfare). For an extended mathematical treatment of welfare maximization, see H. VARIAN, MICROECONOMIC ANALYSIS 152­55 (1978).

FN25. No copyright analyst has demonstrated that efficiency actually is copyright's primary concept. However, some analysts use efficiency as if this were the case. See Landes & Posner, supra note 5 (describing copyright's major doctrines as the promotion of economic efficiency). See also A.M. POLINSKY, AN INTRODUCTION TO LAW AND ECONOMICS 119­27 (2d ed. 1989) (arguing that efficiency considerations should be the primary basis for choice of legal rules); R. POSNER, ECONOMIC ANALYSIS OF LAW 24 (3d ed. 1986) ("[t]he economic theory of law is the most promising positive theory of law extant"). For additional articles in which efficiency is the primary tool used to analyze copyright, see supra note 5.

FN26. See Gordon, Book Review, Toward a Jurisprudence of Benefits: The Norms of Copyright and the Problem of Private Censorship, 57 U. CHI. L. REV. 1009, 1042­43 (1990).

FN27. This danger is eloquently expressed by Professor Frank Michelman in his article Norms and Normativity in the Economic Theory of Law, 62 MINN. L. REV. 1015, 1028 (1978):

The economic fashion in legal criticism has, I believe, a vigorous weedy propensity. In various fields of law we are being presented with studies­­ many of them fine ones, as far as they go­­that not only analyze the efficiency properties of legal doctrines but also contribute­­perhaps carelessly or heedlessly, but also perhaps blindly or bemusedly­­to a general sense that the law is good or bad, right or wrong, true or false, mainly insofar as it is or is not efficient.

FN28. It is also possible that efficiency should be adopted in copyright because efficiency is a superior ethical basis on which to base all legal rules. See R. POSNER, THE ECONOMICS OF JUSTICE (1981). However, this position has been successfully discredited by numerous critics. See infra note 102. This article therefore proceeds without revisiting the question of whether efficiency is the true ethical basis of law.

FN29. Gordon, supra note 5, at 1633 (explaining fair use doctrine and application to parody in terms of efficiency); P. GOLDSTEIN, supra note 5, at 187­212 (taking efficiency approach to both fair use doctrine and parody); Landes and Posner, supra note 5, at 357­61 (applying efficiency model to fair use and parody). Although each of these authors suggests that parody's fair use treatment promotes efficiency, none actually undertakes a detailed examination of whether this is in fact the case. As this article later shows, it is difficult, if not impossible, to defend parody's fair use treatment as efficient. See infra notes 120­138 and accompanying text.

FN30. See supra note 2. For more detail, see infra notes 35­75 and accompanying text.

FN31. See infra notes 35­75 and accompanying text.

FN32. See, e.g., Fisher v. Dees, 794 F.2d 432 (9th Cir. 1986); Elsmere Music, Inc. v. National Broad­casting Co., 482 F. Supp. 741 (S.D.N.Y.), aff'd, 623 F.2d 252 (2d Cir. 1980); Berlin v. E.C. Publications, Inc., 329 F.2d 541, 545 (2d Cir.) ("[f]or, as a general proposition, we believe that parody and satire are deserving of substantial freedom") (emphasis in original), cert. denied, 379 U.S. 822 (1964).

FN33. H.R. REP. NO. 1476, 94th Cong., 2d Sess. 65, reprinted in 1976 U.S. CODE CONG. & ADMIN. NEWS 5659, 5978 (listing parody as an area in which courts could appropriately use the fair use doctrine to limit the exclusive rights of authors).

FN34. It might be contended that parody's fair use treatment is not a good test of efficiency's applicability in copyright because parody is a unique way of exploiting copyrightable subject matter. In my opinion, such a contention is mistaken, for efficiency theorists themselves have cited parody's fair use treatment as an example of copyright's supposed efficiency. See supra note 29. For articles which support parody's fair use treatment, see Light, Parody, Burlesque, and the Economic Rationale for Copyright, 11 CONN. L. REV. 615 (1979); Dorsen, Satiric Appropriation and the Law of Libel, Trademark, and Copyright: Remedies Without Wrongs, 65 B.U.L. REV. 923 (1985); Goetsch, supra note 1; Note, The Parody Defense to Copyright Infringement: Productive Fair Use after Betamax, 97 HARV. L. REV. 1395 (1984); M. NIMMER & D. NIMMER, supra note 17, at ß 13.05[D] ("[o]nly by the recognition of a fair use defense is society likely to reap the benefit of this socially useful literary genre [of parody]").

FN35. Normally the author is the copyright holder, for copyright generally vests copyright in the author. 17 U.S.C. ß 201 (1988). Accordingly, this article will use copyright holder and author interchangeably.

FN36. 17 U.S.C. ß 106(1), (3), (4), (5).

FN37. 17 U.S.C. ß 106(2).

FN38. 17 U.S.C. ß 501(b).

FN39. 17 U.S.C. ß 506(a). Criminal prosecutions apply only to willful infringement for commercial advantage or private financial gain. Id.

FN40. 17 U.S.C. ß 501(a).

FN41. The copyright code defines a "derivative work" as

a work based upon one or more preexisting works, such as a translation, musical arrangement, dramatization, fictionalization, motion picture version, sound recording, art reproduction, abridgment, condensation, or any other form in which a work may be recast, transformed, or adapted. A work consisting of editorial revisions, annotations, elaborations, or other modifications which, as a whole, represent an original work of authorship, is a "derivative work."

17 U.S.C. ß 101 (1988). See also Goldstein, supra note 2, at 234 ("[p] arodies and satires represent derivative uses no less than dramatizations, abridgments or other arrangements of the underlying work").

FN42. See, e.g., Benny v. Loew's Inc., 239 F.2d 532 (9th Cir. 1956), aff'd by an equally divided court sub nom. Columbia Broadcasting Sys. v. Loew's, Inc., 356 U.S. 43 (1958) (parody appeared on commercial television). Fisher v. Dees, 794 F.2d 432 (9th Cir. 1986) (parody distributed on record); Elsmere Music, Inc. v. National Broadcasting Co., 483 F.Supp. 741 (S.D.N.Y.), aff'd, 623 F.2d 252 (2d Cir. 1980) (parody part of Saturday Night Live television show); Berlin v. E.C. Publications, Inc., 329 F.2d 541 (2d Cir. 1964), cert. denied, 379 U.S. 822 (1964) (parody distributed in Mad Magazine).

FN43. For a discussion of this reasoning in an efficiency context, see infra notes 104­105 and accompanying text.

FN44. Although the parodist's special privilege to borrow from copyrighted works is presently well established, its somewhat confused development is worth repeating. The relevant history begins with the May 6, 1955 decision in Loew's Inc. v. Columbia Broadcasting Sys., 131 F. Supp. 165 (S.D. Cal. 1955), aff'd sub nom. Benny v. Loew's, Inc., 239 F.2d 532 (9th Cir. 1956), aff'd by an equally divided court sub nom. Columbia Broadcasting Sys. v. Loew's, Inc., 356 U.S. 43 (1958). In Loew's, the plaintiffs owned the copyright in the movie Gaslight, which starred Charles Boyer and Ingrid Bergman. Id. at 168. The defendant, Jack Benny, created Autolight, a television burlesque of Gaslight. Id. at 169­70. The plaintiffs sued for copyright infringement. Benny defended on grounds of fair use. Id. at 172­ 73. In ruling for the plaintiffs, Judge Carter rejected Benny's contention that parodists should be given special permission to borrow from copyrighted originals. The court specifically held that "a parodized or burlesqued taking is treated no differently from any other appropriation." Id. at 177.

Interestingly, Judge Carter changed his mind a mere seven months later. In Columbia Pictures Corp. v. National Broadcasting Co., 137 F. Supp. 348 (S.D. Cal. 1955), he considered a copyright suit against the creators of From Here to Obscurity, a parody of the movie From Here to Eternity. Despite his ruling in Loew's, Judge Carter wrote:

When the alleged infringing work is of the same character as the copyrighted work, viz., a serious work with a taking from another serious copyrighted work, then the line [between infringement and permissible borrowing] is drawn more strictly than when a farce or comedy or burlesque takes from a serious copyrighted work or vice versa.

Id. at 350. Under normal circumstances, Judge Carter's decision would have established the parodist's special right to borrow from copyrighted originals. In the meantime, however, the losing defendants in Loew's had appealed to the Ninth Circuit. On December 26, 1956, that court affirmed Judge Carter's earlier decision in Loew's. Benny v. Loew's, Inc., 239 F.2d 532 (9th Cir. 1956), aff'd by an equally divided court sub nom. Columbia Broadcasting Sys. v. Loew's, Inc., 356 U.S. 43 (1958). Thus, despite Judge Carter's new ruling in Columbia, the controlling law in the Ninth Circuit was that parodies were to be treated in the same way as other forms of copyright infringement.

This state of affairs persisted until 1964 when the Second Circuit expressly held that parodies are "deserving of substantial freedom" in Berlin v. E.C. Publications, 329 F.2d 541, 545 (2d Cir. 1964). The resulting conflict between the Second and Ninth Circuits was not resolved until 1978, when the Ninth Circuit repudiated its statement in Benny. See Walt Disney Prods. v. Air Pirates, 581 F.2d 751, 757 n.13 (9th Cir. 1978) ("In so construing Benny, we necessarily disagree with its dictum that a parody is treated no differently than any other taking."), cert. denied, 439 U.S. 1132 (1979).

FN45. See, e.g., Fisher v. Dees, 794 F.2d 432, 436 (9th Cir. 1986); MCA, Inc. v. Wilson, 677 F.2d 180, 185 (2d Cir. 1981); Walt Disney Prods. v. Air Pirates, 581 F.2d 751, 758 n.15 (9th Cir. 1978), cert. denied, 439 U.S. 1132 (1979). See also Note, 55 TEMPLE L. Q. 1100 (1982). But see Elsmere Music, Inc. v. National Broadcasting Co., 482 F. Supp. 741, 746 (S.D.N.Y.), aff'd, 623 F.2d 252 (2d Cir. 1980) ("the issue to be resolved by a court is whether the use in question is a valid satire or parody, and not whether it is a parody of the copied song itself").

FN46. Fisher v. Dees, 794 F.2d 432, 437­38 (9th Cir. 1986); New Line Cinema Corp. v. Bertlesman Music Group, Inc., 693 F. Supp. 1517, 1528 (S.D.N.Y. 1980); Metro­Goldwyn­Mayer, Inc. v. Showcase Atlanta Coop. Prods., Inc., 479 F. Supp. 351, 361 (N.D. Ga. 1979). See also Note, supra note 34, at 1409­11; M. NIMMER & D. NIMMER, supra note 17, at ß 13.05[B].

FN47. Berlin v. E.C. Publication, Inc., 329 F.2d 541, 545 (2d Cir. 1964); Fisher v. Dees, 794 F.2d 432, 438­39 (9th Cir. 1986); Elsmere Music, Inc. v. National Broadcasting Co., 482 F. Supp. 741, 747 (S.D.N.Y.), aff'd, 623 F.2d 252 (2d Cir. 1980).

For some time there was a conflict between the Second and Ninth Circuits over the amount a parodist could borrow under the fair use doctrine. The Ninth Circuit took the position that the parodist could borrow only as much as was absolutely necessary to conjure up the original work. See Walt Disney Prods. v. Air Pirates, 581 F.2d 751, 758 (9th Cir. 1978), cert. denied, 439 U.S. 1132 (1979):

[The parodist's] desire to make the "best parody" is balanced against the rights of the copyright owner in his original expressions. That balance has been struck at giving the parodist what is necessary to conjure up the original. . . .

(Emphasis added.) By contrast, the Second Circuit allowed the author the right to borrow substantially more: A parody is entitled at least to "conjure up" the original. Even more extensive use would still be fair use, provided the parody builds upon the original, using the original as a known element of modern culture and contributing something new for humorous effect or commentary.

Elsmere Music, Inc. v. National Broadcasting Co., 623 F.2d 252, 253 n.1 (2d Cir. 1980) (emphasis added). The Ninth Circuit in Fisher v. Dees, 794 F.2d 432 (9th Cir. 1986) eventually adopted a position similar to that of the Second Circuit. In citing Elsmere with approval, the Fisher court expressly disagreed with the "rigid view" of the Air Pirates court. Id. at 438­39. See also M. NIMMER & D. NIMMER, supra note 17, at ß 13.05[C].

FN48. 17 U.S.C. ß 107 (1988).

FN49. Although presently codified, the fair use doctrine originated as a judge­ made exception to an author's rights under the Copyright Code. Congress explicitly recognized fair use for the first time when it revised the Copyright Code in 1976. H.R. REP. NO. 1476, supra note 33, at 65. In doing so, Congress clearly stated its intention to simply restate existing judicial practice. Congress specifically encouraged continued development of the fair use doctrine through the common law process. Id. at 66 ("Beyond a very broad statutory explanation of what fair use is and some of the criteria applicable to it, the courts must be free to adapt the doctrine to particular situations on a case­ by­case basis. Section 107 is intended to restate the present judicial doctrine of fair use, not to change, narrow, or enlarge it in any way.").

FN50. 17 U.S.C. ß 107.

FN51. A complete discussion of the fair use doctrine is beyond the scope of this article. Suffice it to say that an accurate and succinct explanation for the doctrine eludes the best legal minds. For example, Justice Blackmun has labeled fair use "the most troublesome in the whole law of copyright." Sony Corp. of Am. v. Universal City Studios, Inc., 464 U.S. 417, 475 (1984) (Blackmun, J., dissenting) (quoting Dellar v. Samuel Goldwyn, Inc., 104 F.2d 661, 662 (2d Cir. 1939)). A telling anecdote about the difficulty of fair use is the cartoon by Bion Smalley which Professor Melville Nimmer included in his textbook. In the cartoon, a traveler seeking wisdom climbs a mountain to ask a wise man, "What is fair use?" M. NIMMER, COPYRIGHT 67 (3d ed. 1985). For three recent extended discussions of fair use, see Gordon, Fair Use as Market Failure, supra note 5; Fisher, supra note 5; Dratler, Distilling the Witches' Brew of Fair Use in Copyright Law, 43 U. MIAMI L. REV. 233 (1988).

FN52. See Rosemont Enters., Inc. v. Random House, Inc., 366 F.2d 303, 307 (2d Cir. 1966), cert. denied, 385 U.S. 1009 (1967) ("[t]he fundamental justification for the [fair use] privilege lies in the constitutional purpose in granting copyright protection in the first instance, to wit, 'To Promote the Progress of Science and the Useful Arts"'); M. NIMMER & D. NIMMER, supra note 17, at ß 13.05. See also Dratler, supra note 51; Leval, Toward a Fair Use Standard, 103 HARV. L. REV. 1105 (1990). Reference to copyright's general purpose of encouraging creativity is by far the most common touchstone for fair use analysis. However, other perspectives have been suggested. See H.R. REP. NO. 1476, supra note 33, at 65 (characterizing fair use as an equitable rule of reason); Weinreb, Fair's Fair: A Comment on the Fair Use Doctrine, 103 HARV. L. REV. 1137 (1990) (explaining fair use in terms of fairness).

FN53. Although courts do not consistently relate parody's fair use treatment to the four factors outlined in the copyright statute, it is worth noting how this would be done. The first threshold requirement (that the parody contain some criticism of the underlying original) is analogous to the first fair use factor (the purpose and character of the use). The second threshold requirement (that the parody not function as a substitute for the original) corresponds to the fourth fair use factor (the effect on the market for the original work). The parodist's ability to borrow enough so as to "conjure up" the original work relates to the third fair use factor (the amount of the copyrighted work used). Thus, if a court finds that a given parody is a fair use, three of the four statutory fair use factors will have been resolved in the defendant's favor. Only the second fair use factor (the nature of the copyrighted work) would go in the plaintiff's favor. See Fisher v. Dees, 794 F.2d 432 (9th Cir. 1986) (applying fair use defense to parody in light of statutory factors).

FN54. See Elsmere Music, Inc. v. National Broadcasting Co., 623 F.2d 252, 253 (2d Cir. 1980) ("in today's world of often unrelieved solemnity, copyright law should be hospitable to the humor of parody"); Netterville, supra note 1, at 227 (describing benefits of laughter).

FN55. See Berlin v. E.C. Publications, Inc. 329 F.2d 541 (2d Cir. 1964) ("we believe that parody and satire are deserving of substantial freedom­­both as entertainment and as a form of social and literary criticism"), cert. denied, 379 U.S. 822 (1964); Dorsen, supra note 34, at 924 ("satire is a potent form of social commentary"); Goetsch, supra note 1, at 41 (describing value of parody as arising from both entertainment and criticism).

FN56. See New Line Cinema Corp. v. Bertlesman Music Group, Inc., 693 F. Supp. 1517, 1528 (S.D.N.Y. 1988). A fourth consequence has been identified and consistently ignored by courts and commentators alike. As the reader might well imagine, parodies sometimes harm an author's emotions or reputation. Authors understandably resent being ridiculed, especially through the twisting of their own work. See Dorsen, supra note 34, at 925 ("satire often causes hurt feelings or embarrassment"). The reasons for ignoring this consequence seem intuitive and have never been made clear. B. KAPLAN, AN UNHURRIED VIEW OF COPYRIGHT 69 (1967) ("we must accept the harsh truth that parody may quite legitimately aim at garroting the original"); Dorsen, supra note 34, at 964; Light, supra note 34, at 634; Fisher v. Dees, 794 F.2d 432, 437 (9th Cir. 1986) (citing Kaplan and excluding critical impact of parody from cost­benefit analysis). Three plausible reasons for ignoring this harm come to mind.

First, since copyright generally works through the provision of financial incentives it would seem that protecting the author's emotional loss bears no relationship to encouraging creative activity. See Gordon, supra note 26, at 1040­41. Second, the first amendment may well require that copyright not stifle the expression of a critical opinion. See B. KAPLAN, supra, at 69 ("I will not conceal my view that it was wrong­­and possibly unconstitutional­­to hold Jack Benny for his television parody of the movie Gaslight"); Goetsch, supra note 1. Finally, one might contend that any harm suffered by authors at the hands of parodists is a fair consequence of their fame. See Note, supra note 34, at 1409 (distinguishing fair from unfair incentives in the context of parody); MCA, Inc v. Wilson, 677 F.2d 180, 191 (2d Cir. 1981) ("permissible parody, whether or not in good taste, is the price an artist pays for success, just as a public figure must tolerate more personal attack than the average private citizen") (Mansfield, J., dissenting).

FN57. See supra note 42.

FN58. For a discussion of this in an efficiency context, see infra note 91 and accompanying text.

FN59. See Dorsen, supra note 34, at 925.

FN60. Gordon, supra note 26, at 1043; Dorsen, supra note 34, at 960. See also Fisher v. Dees, 794 F.2d 432, 437 (9th Cir. 1986), where the court stated:

Parodists will seldom get permission from those whose works are parodied. Self­esteem is seldom strong enough to permit the granting of permission even in exchange for a reasonable fee. The parody defense to copyright infringement exists precisely to make possible a use that generally cannot be bought. (citations omitted).

FN61. Light, supra note 34, at 633. As the attorney for the copyright holder in Benny v. Loew's stated, "[W]e are not against laughter­­at least so long as the laugh is not at our expense." Schooner, Obscene Parody, J. ARTS MGMT. & L., Fall 1984, at 91 n.158.

FN62. See M. NIMMER & D. NIMMER, supra note 17, at ß 13.05[C] ("[o]nly by the recognition of a fair use defense is society likely to reap the benefit of [parody]"); Netterville, supra note 1, at 237 (noting that availability of parodies on television decreased dramatically after Benny v. Loew's (discussed supra at note 44)).

FN63. See supra note 54.

FN64. This reasoning is perhaps best reflected in the statement of Judge Carter:

The defense, "I only burlesqued" the copyrighted material is not per se a defense. To hold otherwise would seriously jeopardize rights of property in copyrights and investments in such works, and would ultimately seriously damage the prices to be paid to authors for their literary works . . . . Unlimited and unrestrained taking by burlesque could destroy the Copyright Act, undermine the motion picture industry, the legitimate stage, and reduce the author to his status of 300 years ago,­­dependent on the largess of the Prince or Patron.

Columbia Pictures Corp. v. National Broadcasting Co., 137 F. Supp. 348, 350­ 51 (S.D. Cal. 1955) (citation omitted). The Second Circuit more recently took a similar position in MCA, Inc. v. Wilson, 677 F.2d 180, 185 (2d Cir. 1981):

We are not prepared to hold that a commercial composer can plagiarize a competitor's copyrighted song, substitute dirty lyrics of his own, perform it for commercial gain, and then escape liability by calling the end result a parody or satire on the mores of society. Such a holding would be an open­ ended invitation to musical plagiarism.

See also M. NIMMER & D. NIMMER, supra note 17, at ß 13.05[C] (criticizing possibility that fair use treatment for parody might allow wholesale appropriation of copyrighted material as long as some element of humor was added).

FN65. See P. GOLDSTEIN, supra note 5, at 211­12 (arguing that sufficient public domain material exists for creation of parodies which do not criticize the underlying original).

FN66. See Chagares, Parody or Piracy: The Protective Scope of the Fair Use Defense to Copyright Infringement Actions Regarding Parodies, 12 COLUM. J.L. & ARTS 229, 230­31 (1988):

Artistic as well as legal scholars universally agree that parody fulfills an important function in society. Aside from its entertainment value, parody is an important vehicle for both artistic and societal criticism, accomplishing its purpose by "exposing the mediocre and the pretentious." Consequently, this art form influences the development of popular culture and, generally, the development of society.

(Footnotes omitted). See also Dorsen, supra note 34, at 924 (describing social value of parody); Goetsch, supra note 1, at 41­42.

FN67. See, e.g., MCA, Inc. v. Wilson, 677 F.2d 180 (2d Cir. 1981) (holding that the parody "Cunnilingus Champion of Company C" was not a fair use of "Boogie Woogie Bugle Boy of Company B," in part because the defendant's work did not criticize the underlying original); New Line Cinema Corp. v. Bertlesman Music Group, Inc., 693 F. Supp. 1517 (S.D.N.Y. 1988) (refusing to apply fair use defense while expressing doubt about critical value of defendants' parody); Metro­Goldwyn­Mayer, Inc. v. Showcase Atlanta Coop. Prod., Inc., 479 F. Supp. 351 (N.D. Ga. 1979) (noting that defendant's parody was not critical commentary on the underlying original work and refusing to apply fair use defense). See also infra note 72.

FN68. See supra note 56 and accompanying text.

FN69. Cf. Gordon, supra note 5, at 1632 ("[t]he usual economic assumption is that the owner of a resource will either exploit that resource himself, or will sell it to someone else who will").

FN70. See supra notes 8­10 and accompanying text.

FN71. See supra notes 59­61 and accompanying text.

FN72. See New Line Cinema Corp. v. Bertlesman Music Group, Inc., 693 F. Supp. 1517 (S.D.N.Y. 1988). In New Line, the plaintiff copyright holders of the movie A Nightmare on Elm Street decided to make a rap music video based on the movie and its lead character, Freddy Krueger. However, before the plaintiffs' video could be finished, the defendants produced their own competing rap video. The plaintiffs sued, alleging copyright violation.

In response to the plaintiffs' motion for preliminary injunction, the defendants argued that their video was a parody of the plaintiffs' work and that fair use should therefore apply. Id. at 1524. The court correctly rejected this contention. Id. at 1531. First, the court expressed serious doubt about whether the defendants' video made any critical statements about the underlying original work. Id. at 1525. More importantly, the court noted that the defendants' video was a direct substitute for the plaintiffs' video. Id. at 1528. The resulting harm to the plaintiffs' financial copyright interests played a major role in the court's final decision.

New Line provides an excellent illustration of the relationship between criticism, a copyright holder's willingness to exploit his work through parody, and the intuitive cost­benefit analysis courts apply to parodies. Even though their use of the underlying original was humorous, the defendants lost because the court found that their work lacked social value and threatened the plaintiffs' copyright incentives. The reasoning behind this conclusion seems clear. Since the defendants' parody did not criticize the copyrighted movie and characters, its social value was marginal. At the same time, this lack of criticism meant that the plaintiffs therefore became willing to exploit this use of the underlying original themselves, and the defendants' work clearly harmed the plaintiffs' financial copyright incentives.

FN73. Light, supra note 34, at 633.

FN74. See supra notes 63­67 and accompanying text.

FN75. See Note, supra note 34, at 1410 ("[t]he standard [for establishing the potential substitution of a parody for the original work] should include prospective competition with uses that the copyright holder is reasonably likely to attempt in the future").

FN76. See supra note 45 and accompanying text.

FN77. See supra note 46 and accompanying text.

FN78. Remember, successful parody won't exist at all without fair use treatment. See supra note 62 and accompanying text.

FN79. See supra note 47 and accompanying text.

FN80. This concept is named after Vilfredo Pareto, the Italian economist who developed it. See M. LUTZ & K. LUX, THE CHALLENGE OF HUMANISTIC ECONOMICS 92 (1979).

FN81. J. COLEMAN, supra note 3, at 97­98.

FN82. Id. at 98­100. Put somewhat differently, a situation A is Kaldor­Hicks superior to situation B if A is potentially Pareto superior to B. It therefore follows that situation A is Kaldor­Hicks efficient if no other potentially Pareto superior situations exist.

FN83. Among other things, the rational person has preferences which are complete and transitive. Completeness guarantees that any two alternatives can be compared. Thus, the rational person either prefers X to Y, prefers Y to X, or is indifferent between X and Y. Transitivity guarantees that if A is preferred to B and B is preferred to C, than A is preferred to C. Economists also assume that these preferences never change. See Georgescu­Roegen, Choice, Expectations and Measurability, 68 Q.J. ECON. 503, 505­10 (1954) (defining the traits of rational economic actor); H. VARIAN, supra note 24, 80­84 (presenting basic assumptions about consumer preferences).

FN84. See R. COOTER & T. ULEN, LAW AND ECONOMICS 23 (1988) (the preferences of a consumer are "purely subjective"; they are "his or her preferences, to be discovered by finding out what he or she likes, not by telling him or her what to like"); M. LUTZ & K. LUX, supra note 80, at 93 ("[a]ccording to [the theory of the rational person], whatever he or she does in the market is deemed the rational and preferable thing to do"). The sovereignty of subjective preferences is of paramount importance to any efficiency based theory of law, for it forces the economist to take account of all human desires, no matter how foolish or immoral. Interestingly, it is precisely the economist's acceptance of subjective preferences which ultimately frustrates the efficiency explanation for parody's fair use treatment. See infra notes 127­138 and accompanying text.

FN85. R. COOTER & T. ULEN, supra note 84, at 44­45.

FN86. The perfect market is characterized by the presence of perfect information and the absence of transaction costs, monopolies, externalities, and public goods. Id. at 45. The failure of these conditions destroys the market's presumptive advancement of the social welfare. See infra notes 92­95 and accompanying text.

For our purposes, the assumed nonexistence of externalities is of particular interest. See infra notes 109­115 and accompanying text. Externalities exist when a person fails to enjoy (or bear) the full economic consequences of her actions. When externalities occur, the market's presumptive social optimality disappears because voluntary exchange no longer accounts for the full social costs and benefits of a given transaction. To take a common example, consider the actions of a steel mill that pollutes the surrounding air. In deciding how much steel to produce, the mill weighs the cost of materials, energy, and labor against the price for which it can sell the steel. Under perfect conditions, the price consumers are willing to pay for the steel reflects its social value. Thus, the steel mill produces the socially optimal amount of steel when it meets the demands of its customers. However, the existence of pollution creates an externality which destroys this result. Although the price of steel reflects its value to the steel mill's customers, it does not reflect the harm caused by pollution to the steel mill's neighbors. Thus, the mill believes that its steel is more valuable to society than it actually is, and the mill produces more than the socially optimal level of steel. See RUFFIN & GREGORY, PRINCIPLES OF MICROECONOMICS 225 (4th ed. 1990) (defining externality and its effect on the market); see also A.M. POLINSKY, supra note 25, at 15­25 (discussing efficiency criteria in the determinations of nuisance law remedies).

FN87. R. COOTER & T. ULEN, supra note 84, at 44­47.

FN88. $500,000 revenue ­­ $440,000 cost = $60,000.

FN89. $500,000 revenue ­­ $450,000 cost = $50,000.

FN90. See supra note 87 and accompanying text.

FN91. Indeed, judicial attempts to interfere with the status quo are likely to be defeated by the process of voluntary exchange, so there is generally no point in trying to do anything about the status quo anyway. This conclusion is the result of the so­called Coasean theorem, which states that under perfect conditions society will allocate resources in a Pareto efficient manner regardless of how the law assigns those resources. See Coase, The Problem of Social Cost, 3 J.L. & ECON. 1 (1960). See also A.M. POLINSKY, supra note 25, at 11­13; R. COOTER & T. ULEN, supra note 84, at 4­6.

FN92. This is a failure of the no transaction costs assumption. The costs might represent legal fees, telephone bills, office supplies, or time spent negotiating.

FN93. The studio is willing to spend $59,999. However, $10,000 must be spent to cover the transaction costs, thereby reducing the amount left for actual purchase to $49,999. The author will not sell for this price.

FN94. Remember, the cartoonist can still realize a $50,000 profit on his own.

FN95. The author's $50,000 profit is less than the $50,001 or more the studio would have paid him if there had been no transaction costs.

FN96. See A.M. POLINSKY, supra note 25, at 10 (a standard assumption of economic analysis that is made in analyzing the efficiency of legal rules provides that "all benefits and costs can be measured in terms of a common denominator­­dollars") This customary reduction is critically discussed in Georgescu­Roegen, supra note 83, at 10­522. See also M. LUTZ & K. LUX, supra note 80, at 59­75, 317­22.

FN97. A.M. POLINSKY, supra note 25, at 10.

FN98. See R. POSNER, supra note 28, at 60 ("[t]he most important thing to bear in mind about the concept of value is that it is based on what people are willing to pay for something rather than on the happiness they would derive from having it"). See also infra note 100.

FN99. It is important to remember that the proposed monetary compensation must be completely satisfactory to the "losers" if the Kaldor­Hicks criterion is to be correctly applied. As noted previously, Pareto efficiency corresponds to a social optimum because it is the result of voluntary individual transactions. See supra notes 87­91 and accompanying text. Since the Kaldor­Hicks criterion identifies potential Pareto improvements, it follows that the "winners" in any Kaldor­Hicks change must be able to pay the "losers" an amount which converts the change to a Pareto improvement. This can happen only if the payment is one that the "losers" would accept voluntarily. See supra notes 82, 84 (equating Kaldor­Hicks improvement with potential Pareto improvement and noting importance of subjective preferences). See also infra note 100.

FN100. This entire analytical process has been clearly expressed by economist E.J. Mishan:

Since we are committed to the concept of a potential Pareto improvement, we must allow ourselves to be guided by its implications. The notion of an economic event, or reorganization, that can make everyone better off requires that we use the CV concept only. [Professor Mishan defines the CV concept as the sum that maintains welfare at the original level when an economic event occurs].

To be more explicit, all those affected by the economic event can be divided into gainers and losers. Irrespective of which law is operative, the CV of each of the gainers is the maximum sum he is willing to pay for the event. The CV of each of the losers, on the other hand, is the minimum sum he can be made to accept to put up with the event in question. If, then, with respect to some specific economic event, the maximum sums the gainers are prepared to pay (treated as a positive magnitude) exceed the minimum sums acceptable to the losers (treated as a negative magnitude), the algebraic sum of the CVs will be positive, and, by definition, a potential Pareto improvement will have been realized by the event.

E.J. MISHAN, COST­BENEFIT ANALYSIS 137 (1976) (emphasis in original).

FN101. See supra note 97 and accompanying text.

FN102. Alternatively, the economist may use mathematical techniques to maximize society's wealth. Since all values are expressed in terms of dollars, the economist constructs an individual utility function for each member of society. This function expresses the individual's subjective dollar valuation of his or her unique bundle of assets. Once this has been accomplished, society's welfare may be expressed as a function (often the sum) of the individual utility functions. This leads to the expression of social welfare in terms of each individual's distribution of resources. A series of calculations then leads to the mathematical expression of the precise allocation of resources that maximizes society's wealth. See H. VARIAN, supra note 24, at 149­54; Tribe, Policy Science: Analysis or Ideology?, 2 PHIL. & PUB. AFF. 66, 68­73 (1972). This is normally done by solving for the allocation at which the marginal rates of substitution between each pair of goods is equal. H. VARIAN, supra at 152; Tribe, supra at 69 n.7.

Conceptually, these techniques are the same. If no more moves can be made such that the benefits outweigh the costs, then social welfare cannot be improved. This article uses the compensation technique because it is more widely used by legal analysts and because application of mathematical techniques necessary is practically impossible.

It should also be noted that the presentation of efficiency theory here and elsewhere in this article should not be perceived as an unqualified endorsement of efficiency analysis in copyright or law more generally. Instead, efficiency is presented here in order to test its ability to explain parody's fair use treatment. See supra notes 5, 21­34 and accompanying text.

It certainly is possible to construct arguments which support the efficient production of wealth as the sole criterion for evaluating law. See R. POSNER, supra note 28. However, these arguments are subject to convincing criticism. First, although wealth is certainly valued by society, it is not the only thing that society desires. Thus, arranging social affairs by sole reference to the efficient production of wealth rests on the dubious proposition that such efforts necessarily lead to all other goods that society desires. See Dworkin, Is Wealth a Value?, 9 J. LEGAL STUD. 191 (1980); J. COLEMAN, supra note 3, at 112 ("If the pursuit of wealth is a good, it must be because pursuing wealth promotes other things of value."). This ethical criticism is echoed by economists who argue against the ordering of society solely upon efficiency considerations. See E.J. MISHAN, THE ECONOMIC GROWTH DEBATE 36 (1977) ("[W]ho doubts that the wealthier and economically more efficient society can also be the less healthy, the less honest, the less secure and the less contented?"); Rizzo, The Mirage of Efficiency, 8 HOFSTRA L. REV. 641, 646 (1980) (pointing out that markets poorly reflect the value of moralisms).

Second, even if one accepts the notion that the blind pursuit of wealth is an ethically acceptable way to order society, economic theory itself offers numerous reasons to take recommendations based solely on efficiency with a grain of salt. As an initial matter, the efficiency analyst seldom, if ever, obtains all the information necessary to unambiguously and completely measure society's wealth. In these situations a rule which appears to make one sector of the economy efficient may well reduce social welfare overall. See Rizzo, supra at 652­53 (describing the problem of second best), E.J. MISHAN, supra note 100, at 98­101. Moreover, the economist sometimes becomes trapped in a never ending cycle of conflicting recommendations. See Scitovszky, A Note on Welfare Propositions in Economics, 9 REV. ECON. STUD. 77 (1941) (outlining the so­called "Scitovskzy paradox"); E.J. MISHAN, supra note 100, at 140­41, 398­ 401 (discussing wealth effects). These conflicts show that society could not completely order itself on economics even if it wanted to.

A detailed examination of these issues is beyond the scope of this article. However, it is worth noting that these criticisms cast doubt upon whether economics can ever provide a complete theory of copyright law. See Yen, supra note 5, at 544­46.

FN103. See Gordon, Fair Use as Market Failure, supra note 5.

FN104. See supra note 44.

FN105. See supra note 43 and accompanying text.

FN106. See supra notes 44, 47 and accompanying text.

FN107. Berlin v. E.C. Publications, Inc., 329 F.2d 541, 545 (2d Cir. 1964), cert. denied, 379 U.S. 822 (1964).

FN108. See supra note 66 and accompanying text.

FN109. See supra notes 88­91 and accompanying text.

FN110. Id.

FN111. See supra note 66 and accompanying text.

FN112. Economically sophisticated readers will recognize this as a failure of the no­externality condition. In this situation an external benefit exists because parodists are unable to capture the entire social benefit of their work in the price they realize upon its sale. See supra note 86. Some commentators have explained the market failure associated with parody as one of imperfect information. Under this view, parody is seen as a form of book review, thereby helping consumers decide which works to buy. See Gordon, Fair Use as Market Failure, supra note 5, at 1634. Although this is plausible, I do not find it convincing because consumers of parodies find them useful after viewing the original, not before. Thus, while parody does affect consumer taste, the notion that it plays a significant role in the commercial success or failure of a given work seems a bit far­fetched. Rather, as stated in the text, its value is in a broader humorous perspective and commentary.

FN113. See supra note 42 (listing cases in which parody is widely disseminated as entertainment).

FN114. See Gordon, supra note 26, at 1042 (noting that critics rarely capture the social value of their work in the price they receive).

FN115. See supra notes 92­95 and accompanying text. The foregoing also explains why an economist would support judicial refusals to apply the parody defense to works whose value lies strictly in its humor. For example, in Metro­Goldwyn­ Mayer, Inc. v. Showcase Atlanta Coop. Prods., Inc., 479 F. Supp. 351 (N.D. Ga. 1979), the Northern District of Georgia considered a copyright suit by the owners of Gone with the Wind against the producers of Scarlett Fever, a three hour play which humorously followed the Gone with the Wind plot. Id. at 354­ 55. Although the defendants claimed that Scarlett Fever was a parody of Gone with the Wind, the court imposed a preliminary injunction in the plaintiffs' favor. In so ruling, the court noted that "the work as a whole is not a critical commentary on either the film or the novel Gone with the Wind." Id. at 357. From an efficiency perspective, this decision was correct. Since Scarlett Fever contained no criticism of the underlying original, it was socially valuable primarily as entertainment. Thus, its market price probably reflected accurately its contribution to public welfare. Forcing the defendants to purchase derivative works rights from the owners of Gone with the Wind would therefore encourage the efficient allocation of resources.

FN116. See supra note 45.

FN117. See supra note 82 and accompanying text.

FN118. Normally, the purchase price of a work reflects its social value. In this case, however, one must also take into account the value of parody that is not reflected in its price because of market failure. See supra note 86 and accompanying text.

FN119. As noted previously, parody's fair use treatment inflicts two possible harms on authors. First, authors lose their ability to exploit their works through parody themselves. Second, they are directly harmed by the ridicule of parody. See supra note 56 and accompanying text.

FN120. See supra notes 57­67 and accompanying text.

FN121. See supra notes 68­75 and accompanying text.

FN122. See supra notes 76­79 and accompanying text.

FN123. See supra note 84 and accompanying text.

FN124. See supra note 82 and accompanying text.

FN125. See supra notes 78­79 and accompanying text.

FN126. See supra notes 82, 99­101 and accompanying text.

FN127. If sale of parodies did not generate any external benefits, then parodists would be able to charge society the entire amount society is willing to pay for access to parody.

FN128. Since society's gains are large enough to generate compensation for parodists, the court orders the fair use treatment of parody under the Kaldor­ Hicks criterion. Although this change inflicts a loss upon the author, the general assumption that compensation will be paid turns a Kaldor­Hicks improvement into a Pareto improvement. See supra notes 99­101 and accompanying text.

FN129. This sort of preference structure is an example of lexicographic, or lexical, ordering. Lexical ordering occurs when an individual values good A in such a way that she will not accept good B as a substitute, no matter how much of good B she can have instead of good A. Harrison, Egoism, Altruism, and Market Illusions: The Limits of Law and Economics, 33 UCLA L. REV. 1309, 1328 (1986); Tribe, supra note 102, at 90; Georgescu­Roegen, supra note 83, at 510­ 18; M. LUTZ & K. LUX, supra note 80, at 68­75, 317­26.

The possibility of lexical ordering is important because efficiency theorists have always assumed that people will not value different goods in this way. By assuming that all value can be expressed in terms of money, economists construct a world where individuals will accept money as a substitute for any loss they may suffer. Brief reflection shows that the nonexistence of lexical ordering is the very linchpin of efficiency analysis.

For example, the Kaldor­Hicks criterion rests on the notion that the payment of some amount of money compensates for any loss imposed on individuals in the name of efficiency. As long as individuals in fact consider money an adequate substitute for their losses, any Kaldor­Hicks superior move can be transformed into a Pareto superior move by the payment of money. See supra notes 99­101. However, if individuals value certain goods as lexically prior to money, the likelihood of promoting Pareto efficiency through the Kaldor­Hicks criterion diminishes tremendously.

Consider a situation in which the efficiency analyst is trying to decide whether to order construction of a dam which will flood A's property. Under a Kaldor­Hicks analysis, the dam should be built if the dollar value of the dam's benefits yields enough money to induce A to sell his property voluntarily. If A in fact values his property in money terms, the Kaldor­Hicks analysis provides a plausible yardstick for measuring the relative social value of the dam and A's property. However, if A refuses to sell at any price, (perhaps because the land is the traditional home of his family), any money oriented cost­benefit analysis which supports construction of the dam cannot promote Pareto efficiency. Since A will not accept money as compensation for his property, any forced flooding of A's land inflicts an uncompensated loss on A which cannot be squared with Pareto criterion's underpinnings in consent.

The effect of lexical ordering on the mathematical maximization of social welfare is even more striking. If individuals express their preferences in a lexically ordered fashion, a fundamental assumption which makes mathematical calculation of the welfare maximum possible is no longer true. In particular, economists assume that individual preferences may be expressed as continuous functions. This assumption is necessary to ensure that calculation of marginal rates of substitution proceeds smoothly. H. VARIAN, supra note 24, at 81; Tribe, supra note 102, at 69. However, if lexical ordering appears, the requirement of continuity is violated. H. VARIAN, supra, at 83­4. Once this occurs, the entire process of calculating a social welfare maximum fails. Tribe, supra, at 93 ("as long as both individual and societal preference orderings display this sort of 'lumpiness,' the entire scheme of assigning rights and liabilities so as to maximize efficiency, while using transfers of income to smooth out the distribution of wealth conceived as a homogeneous good, is doomed to failure"). As we shall see, these very problems interfere with any attempt to defend parody's fair use treatment as efficient. See infra notes 131­138 and accompanying text.

FN130. See supra notes 80­81 and accompanying text.

FN131. See Gordon, supra note 26, at 1043 ("[w]hen goods as important and irreplaceable as life or reputation are on the table persons are unlikely to sell what they own at any price"). See also Fisher v. Dees, 794 F.2d 432, 437 (9th Cir. 1986) ("[s]elf­esteem is seldom strong enough to permit the granting of permission [to a parodist] even in exchange for a reasonable fee"); Post, The Social Foundations of Defamation Law: Reputation and the Constitution, 74 CALIF. L. REV. 691, 707 (1986) (basing reputation on the "essential dignity and worth of every human being"); R. BEZANSON, J. SOLOSKI, & G. CRANBERG, LIBEL LAW AND THE PRESS 1 (1987) ("[a] person's good name is priceless") (emphasis in original). This intuition is further confirmed by the observation that parodies became rather scarce in the wake of Judge Carter's original decision in Loew's. Netterville, supra note 1, at 237 ("[n]o one can deny that since the %ADLoew's%BD decision, parody and burlesque involving copyrighted works have all but disappeared from the major medium for communication in the United States today­­the television screen").

FN132. R. BEZANSON, J. SOLOSKI, & G. CRANBERG, supra note 131.

FN133. Id. at 7.

FN134. Id. at 19­28.

FN135. Id. at 21­22. It is worth remarking that survey evidence cannot be considered conclusive, for the things people say may well be different from what they actually think. Nevertheless, the survey evidence presented here is certainly evidence of whether people consider money an adequate substitute for harm to their reputation and emotional tranquility.

FN136. Id. at 23­28.

FN137. Id. at 26.

FN138. The authors' apparent refusal to accept money as a substitute for reputation and emotional tranquility has devastating effects on the mathematical calculation of social welfare as well. See supra note 129.

FN139. See supra notes 96­98.

FN140. See supra note 56.

FN141. See supra notes 127­128 and accompanying text.

FN142. See supra notes 129­138 and accompanying text.

FN143. See New Era Publications, Int'l, ApS v. Carol Publishing Group, 904 F.2d 152 (2d Cir. 1990) (copyright suit against author who had written a critical biography of L. Ron Hubbard and the Church of Scientology).

FN144. See Salinger v. Random House, Inc., 811 F.2d 90 (2d Cir.), cert. denied, 484 U.S. 890 (1987); Craft v. Kobler, 667 F. Supp. 120 (S.D.N.Y. 1987).

FN145. Cf. Gilliam v. American Broadcasting Co., 538 F.2d 14 (2d Cir. 1976) (suit under Copyright and Lanham Acts to enjoin defendant's broadcasting an edited version of Monty Python's Flying Circus television program); Gordon, supra note 26, at 1033 ("[t]he author of a new work is unlikely to obtain permission from a prior author if he wishes to criticize the work or use the prior author's material in a way that rejects or undercuts the meaning the predecessor meant to invest in her materials or symbols").

FN146. See Dratler, supra note 51 (examining and clarifying factors relevant to fair use analysis); Leval, supra note 52 (advocating the resolution of fair use ambiguity by further attention to copyright's basic principles).

FN147. Indeed, courts and commentators have suggested non­utilitarian theories for why the intuitive cost­benefit explanation for parody's fair use treatment ignores any harm to an authors' emotions or reputation. From a constitutional perspective, some analysts have suggested that the parodist's right of free expression is simply greater than the author's right to be free of ridicule. See Goetsch, supra note 1; B. KAPLAN, supra note 56, at 69. This constitutional position may also reflect society's belief that ridicule is simply a fair and just consequence of an author's fame. See MCA, Inc. v. Wilson, 677 F.2d 180, 191 (2d Cir. 1980) ("permissible parody, whether or not in good taste, is the price an artist pays for success, just as a public figure must tolerate more personal attack than the average private citizen") (Mansfield, J., dissenting).