This Part describes the current state of the fair use doctrine, paying special attention to the changes wrought by Sony and Harper & Row, argues that we should strive to improve the doctrine, and provides the raw material for the project of reconstruction begun in Part III. ³⁹

 [*1668]  A. "An Equitable Rule of Reason" ⁴⁰

The first thing one notices when reading the opinions in the two cases is the absence of any effort to prescribe a rule to govern future controversies. The consensus of the Court was that "fair use analysis must always be tailored to the individual case." ⁴¹ "The inquiry is necessarily a flexible one, and the endless variety of situations that may arise precludes the formulation of exact rules." ⁴²

What the Justices offered instead was a list of "factors" to be considered by lower courts when deciding whether particular uses are fair. Some of these criteria consisted of glosses on the four statutory considerations; ⁴³ others were creations of the Court. In neither case did the Court venture a precise or exhaustive enumeration of the factors. ⁴⁴ However, a close reading of the two majority opinions reveals a set of questions the Court deemed relevant. In approximate order of importance, they are: (1) Did the infringement have a material impact on the "potential market" for the copyrighted work? (2) Was the use "commercial" or "noncommercial"? (3) Had the copyrighted work been published at the time of the copying? (4) How much -- quantitatively and qualitatively -- of the putatively infringing work was drawn from the copyrighted work? (5) Was the unauthorized use consistent with customary standards of propriety?

In the course of explicating these factors, the Court identified four objectives that copyright law in general and the doctrine of fair use  [*1669]  in particular ought to serve: (a) advancing social utility by increasing the supply of intellectual products and facilitating their distribution; (b) enforcing an author's natural right to a reasonable portion of the fruits of his labor; (c) protecting an author's interest in controlling the way in which his creations are presented to the world; and (d) aligning the law with custom and popular conceptions of decent behavior. Attention to these underlying goals, the Court suggested, would facilitate both interpretation of the factors enumerated above and identification of other appropriate criteria in future cases.

Detailed studies of the components of this composite test follow.

B. The Factors

1. Impact on the Potential Market. -- Justice Stevens, writing for the Court in Sony, centered the fair use analysis on the fourth of the considerations mentioned in section 107: "the effect of the [putatively infringing] use upon the potential market for or value of the copyrighted work." ⁴⁵ "The purpose of copyright," he contended, "is to create incentives for creative effort." ⁴⁶ "[A] use that has no demonstrable effect upon the potential market for, or the value of, the copyrighted work need not be prohibited in order to protect the author's incentive to create" -- and therefore should be deemed "fair." ⁴⁷ Justice Stevens argued that application of this factor to the facts of the case suggested that time-shifting should be considered a fair use. Pointing to the district court's finding that the studios had failed to prove "that the practice has impaired the commercial value of their copyrights or has created any likelihood of future harm," ⁴⁸ he maintained that it would be senseless to forbid it.

Justice Blackmun disagreed. Relying on a case decided a few years earlier by the Second Circuit, ⁴⁹ Justice Blackmun argued that Justice  [*1670]  Stevens' analysis presupposed an indefensibly narrow conception of the "market" for the studios' works. "[P]ersons who find it impossible or inconvenient to watch . . . programs at the time they are broadcast, and who wish to watch them at other times," he contended, surely constitute in some sense a "potential market" for the programs; privileging the use of VCRs has the effect of denying to the copyright owners the opportunity to exact fees from the members of that market. ⁵⁰ Generalizing the point, Justice Blackmun argued that, to avoid an adverse finding on the market-impact factor:

the infringer must demonstrate that he had not impaired the copyright holder's ability to demand compensation from (or to deny access to) any group who would otherwise be willing to pay to see or hear the copyrighted work. . . . [T]he fact that a given market for a copyrighted work would not be available to the copyright holder were it not for the infringer's activities does not permit the infringer to exploit that market without compensating the copyright holder. ⁵¹

Justice Blackmun's argument is powerful. Only on the basis of a conception of a "market" more restrictive than a "group of persons who would . . . be willing to pay to see" the work can it be rebutted. ⁵² In view of Justice Stevens' failure to provide any such conception in  [*1671]  Sony, it is not surprising that Justice O'Connor, writing for the Court in Harper & Row, quietly adopted Justice Blackmun's definition of "potential market." ⁵³

Unfortunately, Justice O'Connor simultaneously adopted Justice Stevens' severe view of the significance of adverse impact on such a market. In Sony, Justice Stevens suggested that, to avoid a finding of fair use, a plaintiff need only "show[] by a preponderance of the evidence that some meaningful likelihood of future harm exists." ⁵⁴ In Harper & Row, Justice O'Connor took an equally rigid position: "Fair use, when properly applied, is limited to copying by others which does not materially impair the marketability of the work which is copied.'" ⁵⁵ What Justice O'Connor failed to recognize is that, in almost every case in which the fair use doctrine is invoked, there will be some material adverse impact on a "potential market" as Justice Blackmun -- and now the Court -- define that phrase. After all, in any such suit, the defendant is seeking to use the plaintiff's copyrighted work in a fashion ostensibly forbidden by section 106 of the Copyright Act. To permit the defendant to engage in the activity for free prevents the plaintiff from exacting a fee from the defendant. ⁵⁶ Thus, in all but the rare cases in which the defendant for some reason would be unwilling to pay the plaintiff anything, ⁵⁷ a finding that the defendant's conduct is "fair" will "impair the marketability of the work."

In short, fairly applied, the version of the market-impact factor adopted in Harper & Row will almost always tilt in favor of the plaintiff -- and is therefore nearly useless in differentiating between  [*1672]  fair and unfair uses of copyrighted materials. To be helpful, it must be modified. At least one change is essential: if Justice Blackmun's understanding of "potential market" is to be retained -- and no alternative definition is in the offing -- then a court confronted with a fair use defense must estimate the magnitude of the market impairment caused by privileging the defendant's conduct; merely ascertaining the existence of adverse impact will not suffice. ⁵⁸

2. "Commercial" and "Noncommercial" Uses. ⁵⁹ -- With the exception of market impact, the factor that looms largest in the two opinions is the one specified first in section 107: whether the putatively infringing use is "commercial" in nature. ⁶⁰ Two dangers attend use of this criterion to separate fair from unfair uses. The first is that of "double counting." The only plausible justification for the test -- as Justice Stevens implicitly recognized ⁶¹ -- is that it may indicate the  [*1673]  severity of the probable harm to the copyright owner of allowing the defendant to engage in his activity. Although one can imagine noncommercial activities that would seriously impair a potential market for a work, ⁶² it seems likely that commercial uses usually result in more severe injuries to copyright owners than noncommercial uses. If that is the significance of the commercial character of an activity, however, to tally it as a separate factor is misleading. At a minimum, its status as an adjunct to the market-impact test should be borne in mind.

The second danger derives from the ambiguity of the term "commercial" -- an ambiguity the Court has enhanced. In the various opinions in Sony and Harper & Row, the term was used in three markedly different ways. Justice Stevens equated it with moneymaking. ⁶³ Justice Blackmun, in his dissent, contrasted commercial activities with activities motivated by "humanitarian impulse[s]," implicitly defining commercial as selfish. ⁶⁴ Justice O'Connor eschewed Justice Blackmun's inquiry into the user's motivations and ventured a more objective definition: "The crux of the profit/nonprofit distinction is not whether the sole motive of the use is monetary gain but whether the user stands to profit from exploitation of the copyrighted material without paying the customary price." ⁶⁵ Each of these definitions  [*1674]  is problematic, ⁶⁶ but more serious than their individual defects is the fact that the Court has not indicated which is authoritative.

Again, therefore, if the commercial/noncommercial factor is to be retained as a component of the fair use doctrine, it should be modified. At a minimum, the Supreme Court ought to make clear which of the definitions now in circulation the lower courts should employ. ⁶⁷

3. Publication. -- A factor unmentioned in section 107 that the Court identified and emphasized in Harper & Row is whether, at the time of the copying, the copyrighted work had been published. Prior to 1976, unpublished works were protected not by the Copyright Act, but by common law copyright rules administered for the most part by state courts, which accorded little scope to the idea of "fair use." ⁶⁸ The 1976 reform of the Copyright Act extended its coverage to unpublished materials. ⁶⁹ In Harper & Row, Justice O'Connor argued that some of the concerns that underlay the common law doctrine forbidding most unauthorized uses of unpublished works should inform the interpretation of section 107 of the enlarged federal statute.

Specifically, three considerations supported the conclusion that "the unpublished nature of a work is '[a] key, though not necessarily determinative, factor' tending to negate a defense of fair use." ⁷⁰ First, Justice O'Connor contended, the unauthorized copying of unpublished materials is likely to have a large impact on the economic value of the copyright in those materials. ⁷¹ Second, a risk that his manuscript  [*1675]  will be released (in whole or in part) to the public by someone else will induce an author to publish prematurely and will therefore deprive readers of the benefits associated with well-polished works. ⁷² Third and finally, the unauthorized publication of excerpts of unpublished material creates a danger that readers will be shown a garbled or distorted version of the author's creation before they have access to the creation itself; selective prepublication copying thus threatens the author's "personal interest in creative control" over his work. ⁷³ The implication of this combination of considerations, Justice O'Connor argued, is that "the scope of fair use is narrower with respect to unpublished works." ⁷⁴

The publication factor adopted in Harper & Row stands up better to critical scrutiny than either the market-impact or commercial/noncommercial factors. To be sure, there is force to Justice Brennan's contention that not all cases of prepublication copying implicate the three concerns identified by the majority. ⁷⁵ But it is not obvious that we should therefore abandon the factor in favor of an open-ended analysis of the degree to which the concerns themselves are implicated by particular cases. That issue, along with the larger question of the relevance of the unpublished nature of a work to alternative general theories of copyright, is taken up in Part V. For now it suffices to observe that this factor does not seem seriously defective.

4. Amount of Copying. -- Since the inception of the fair use doctrine in this country, most courts have expressly considered the extent of a defendant's copying in determining whether his activity should give rise to liability. ⁷⁶ Unfortunately, to date they have failed to devise a reliable method of determining how much is too much. One situation has commonly been thought clear-cut: prior to the decision in Sony, the courts were almost unanimous in holding that copying an entire copyrighted work would certainly give rise to liability. ⁷⁷ But the question of how much usage short of appropriation  [*1676]  of an entire work would result in loss of the fair use shield has never been settled. ⁷⁸ Even the terms of the inquiry have remained unclear. Should a court consider the quality of the copied material or only the quantity? Should it attend to the ratio between the amount of material appropriated and the size of the defendant's work or only the ratio between the amount taken and the size of the copyrighted work? ⁷⁹ Section 107(3) of the 1976 Act incorporates without clarifying this amorphous set of concerns: in determining whether a given activity should be considered a fair use, courts are directed to consider "the amount and substantiality of the portion used in relation to the copyrighted work as a whole." ⁸⁰

Perhaps sensing the stubborn vagueness of the "amount of copying" factor ⁸¹ -- or perhaps merely aware of the degree to which it threatened his argument -- Justice Stevens discounted its importance in Sony. ⁸² In Harper & Row, Justice O'Connor accorded it much more  [*1677]  importance. Emphasizing "the expressive value of the excerpts [taken from President Ford's manuscript] and their key role in the infringing work," Justice O'Connor concluded that the extent of the copying was substantial and counted against The Nation's invocation of the fair use doctrine. ⁸³

As revived and explicated in Harper & Row, the "amount of copying" factor has two defects. The first is its notorious fuzziness. Indeed, the long-acknowledged difficulty of applying this factor to particular cases has been increased by the manner in which Justice O'Connor elaborated on it. At the outset of the pertinent paragraph, Justice O'Connor brushed aside The Nation's argument that it had copied only a tiny portion of Ford's manuscript. ⁸⁴ That the material copied was "[i]n absolute terms . . . insubstantial" she deemed irrelevant. ⁸⁵ What mattered was the importance to the book of the passages reproduced -- how "powerful" they were, their "expressive value." ⁸⁶ In effect, Justice O'Connor directed courts applying this factor in future cases to consider primarily "the qualitative nature of the taking." ⁸⁷ The difficulty of making a judgment of that sort will be exacerbated by another aspect of the test enunciated in Harper & Row. Although Justice O'Connor purported at one point to adopt Learned Hand's position that the ratio of the amount of copied material to the size of the defendant's work is irrelevant to the fair use analysis, ⁸⁸ she went on to treat that fraction as a good indicator of the importance the "plagiarist" ascribes to the copied material and therefore of its qualitative importance. ⁸⁹ Justice O'Connor thus perpetuated a crucial ambiguity in this factor -- whether "amount of copying" is to be assessed with reference to the copyrighted work or the infringing work.

The second defect of the factor is that, by according importance to the question of how a copyrighted work is defined, it creates a bizarre and inefficient system of incentives. Artists who wish to maximize their protection against unauthorized copying will devise ways  [*1678]  of subdividing or registering their works that make the copying of even short passages appear "substantial." Courts' efforts to distinguish legitimate from sham identifications of the boundaries of copyrighted works can be expected to produce confusion. ⁹⁰

Even if these two problems could be alleviated or tolerated, it would still be important, when applying this factor, to bear in mind its derivative character. Like the commercial/noncommercial factor, its principal function seems to be that of a proxy for the amount of injury sustained by the copyright owner. To avoid "double counting," that status must not be forgotten.

5. Propriety and Custom. -- In fair use cases decided prior to 1976 courts occasionally referred either to the decency of the defendant's behavior or to its consistency with customary practices. ⁹¹ Express  [*1679]  invocation of either morality or custom was, however, uncommon prior to 1976 reform of the Copyright Act. Not surprisingly, section 107, which was designed to codify the common law doctrine, ⁹² makes no mention of those considerations.

The majority opinion in Sony continued the trend established by the lower-court decisions. Although Justice Stevens might well have invoked the fact that by 1980 time-shifting was a widespread practice regarded by most people as legitimate, he did not. In Harper & Row, by contrast, Justice O'Connor treated a combination of fairness and custom as an important factor in the fair use analysis. ⁹³ "Also relevant to the 'character' of the [challenged] use," she held, "is 'the propriety of the defendant's conduct.'" ⁹⁴ Her opinion was salted with references to the improper way in which The Nation obtained and made use of the Ford manuscript. ⁹⁵ The overall message was that a defendant's conformity with customary standards of "good faith" and "fair dealing" should affect significantly his ability to avail himself of the fair use defense. ⁹⁶

The first thing to observe about this factor is that it requires a decisionmaker to look beyond the positive law for standards by which to evaluate the defendant's conduct. As Justice Brennan implicitly  [*1680]  recognized in his dissent in Harper & Row, ⁹⁷ if one defines propriety in terms of legality, the factor is circular. If "good faith and fair dealing" mean engaging only in activities not violative of the Copyright Act, then it is impossible to decide whether a given activity violates the Act by considering its propriety.

But if courts must look beyond positive law, how are they to identify or determine the ethical standards with which the defendant's conduct is assessed? One possible answer is that "industry practice" should be the gauge. Justice Brennan seemed to have such a theory in mind when he criticized the majority in Harper & Row for ignoring the fact that The Nation's behavior was consistent with "standard journalistic practice." ⁹⁸ This approach has much to recommend it if all parties to a copyright infringement suit are active participants in the industry or community whose practices are invoked. In many areas of property law, courts have been willing, when determining litigants' entitlements, to rely upon a custom prevailing in a group to which they both belong, ⁹⁹ and the considerations that justify that willingness ¹⁰⁰ are readily applicable to disputes over intellectual property. Unfortunately,  [*1681]  however, the parties to most fair use suits are not members of a community sharing a set of customary practices. For example, although both Harper & Row and The Nation are publishers, it is unlikely that Harper & Row belongs to the portion of the industry that engages in the activity, described by Justice Brennan, of publishing without permission stories on forthcoming books. ¹⁰¹ In short, in the majority of fair use cases, courts could not rely upon industry practice to make sense of and justify the "propriety and custom" factor.

What else might a court look to? Justice O'Connor seemed to assume that popular morality can and should be the yardstick. Her opinion appeared to take for granted that everyone knows the meaning of "fair dealing" and the difference between "'"a true scholar and a chiseler."'" ¹⁰² Fair use doctrine, she implied, simply incorporates that common understanding. The hazards of this sort of argument are well recognized in other fields. In constitutional law, for example, most scholars and judges for years have acknowledged the difficulty of deriving substantive standards from "society's widely shared values." In a culture as fractured as that of the United States today, as to few issues does there exist any "consensus to be discovered ¹⁰³ (and to the extent that one may seem to exist, [it] is likely to reflect only the domination of some groups by others)." ¹⁰⁴ Were there by chance a consensus on a relevant issue, it would not be "reliably discoverable, at least not by courts." ¹⁰⁵

 [*1682]  There remains a final possibility. Instead of attempting to derive an ethical standard from extant public opinion, a court could invoke or invent a moral philosophy -- a theory that would enable it to distinguish good behavior from bad. Whether development of such a theory would be feasible and helpful is considered at length in Part V. For the time being, it suffices to observe that, when Justice O'Connor urged lower courts to take into account "the propriety of the defendant's conduct" in deciding fair use cases, ¹⁰⁶ she undoubtedly had nothing of that sort in mind.

6. Vestigial Considerations. -- In addition to the five factors examined in detail above, traces of three other considerations can be found in the majority opinions in Sony and Harper & Row. Despite their relevance to the facts of the two cases, they receive only modest play, suggesting that their role in the fair use doctrine as it emerges from the two cases is marginal. The histories and present statuses of these three considerations are discussed briefly in this subsection to round out the analysis of the current state of the law. The question whether they ought to play more of a role than they do now is deferred to Parts IV and V.

(a) Fact and Fiction. -- Some of the lower federal courts responsible for the development of the fair use doctrine took the position that copyrighted works that are "primarily informational" are entitled to less protection than works that are primarily "creative." ¹⁰⁷ In an influential article, Professor Robert Gorman identified four concerns that, in combination, seem to have given rise to that bias. ¹⁰⁸ First, there is a strong "public interest in access to facts about ourselves, the world about us, and our history and future." Second, in many nonfiction works, the underlying facts are inseparable from the author's expression; facilitating the dissemination of the facts thus requires reducing copyright protection for the expression. Third, both the first amendment and the fair use doctrine require that room be given to "commentary concerning political, social, and historical phenomena." Fourth,  [*1683]  works produced mainly by "the sweat of the brow" appear somehow less deserving of copyright protection than the fruits of artistic genius. ¹⁰⁹

In his dissent in Sony, Justice Blackmun invoked this distinction, arguing that the fact that most of the material copied using VCRs consists of "entertainment shows" rather than "informational works" weakened the claim that time-shifting is a fair use. ¹¹⁰ Justice Stevens did not flatly reject the argument, but gave it short shrift, implicitly deemphasizing the factor on which it rested. ¹¹¹ Disagreement over the relevance of the fact/fiction distinction persisted in Harper & Row. Citing Justice Blackmun's dissent in Sony, Justice Brennan contended that "the scope of fair use is generally broader when the source of borrowed expression is a factual or historical work." ¹¹² Justice O'Connor, writing for the Court, admitted that there was some force to Justice Brennan's point, but then, in a subtle analytical maneuver, limited its impact upon both the case at hand and future fair use cases. She conceded that "[t]he law generally recognizes a greater need to disseminate factual works than works of fiction or fantasy," but argued that this principle permits direct quotation only when "necessary adequately to convey the facts." ¹¹³ Because direct quotation is seldom "necessary" to disclose facts ¹¹⁴ (which cannot be copyrighted  [*1684]  in any event), ¹¹⁵ the net effect of Justice O'Connor's interpretation of this factor will be to limit sharply the role it plays in the administration of fair use doctrine. ¹¹⁶

(b) Necessity. -- A question often asked by lower courts construing the fair use doctrine was: assuming the defendant's objective was laudable, could he have achieved it without copying the plaintiff's copyrighted material? If so, the courts held, the defendant's invocation of the fair use defense was less persuasive than it would have been otherwise. ¹¹⁷ An ambiguity latent in this factor from its inception was whether copying the plaintiff's material had to be truly essential to the achievement of the defendant's goal, or whether he could avail himself of the fair use doctrine as long as the copying somewhat enhanced his ability to realize his objective. ¹¹⁸ Perhaps wary of the dangers inherent in this ambiguity, the authors of the majority opinions in Sony and Harper & Row discounted the necessity factor. Justice Stevens made no explicit mention of it in Sony. ¹¹⁹ In Justice O'Connor's analysis, the idea of necessity figured only in her concession that copying essential to communicate facts may be privileged -- a circumstance that will rarely arise. ¹²⁰

(c) Productivity. -- The most dramatic change wroght by Sony and Harper & Row in the fair use doctrine was the subordination of the idea of productivity. The notion that, to qualify as a fair use, a putatively infringing activity had to advance the common good by  [*1685]  somehow adding to the collection of intellectual products available to the public was either emphasized or taken for granted by most of the lower courts that shaped the doctrine. In the typical fair use case, the defendant had somehow folded copyrighted material into an original work of his own such as a biography, a critical essay, a parody, or a comparative advertisement. In their opinions, courts often explicitly considered whether the social value of the defendant's original contribution was sufficient to entitle him to the defense. ¹²¹ That the defendant had to have made some contribution was usually assumed. ¹²² Leading commentators generally concurred with this view. ¹²³

In his dissent in Sony, Justice Blackmun argued strenuously that the productivity of a defendant's activity should be a crucial consideration in the fair use calculus. The primary purpose of the doctrine, he contended, was "to facilitate the creation of new works." ¹²⁴ Uses of copyrighted material that did not entail the creation of something new should therefore almost never be deemed fair. ¹²⁵ In contrast, Justice Stevens came close to declaring the idea of productivity irrelevant to a fair use determination. Justice Stevens' analysis of the permissibility of time-shifting made no mention of productivity. Only in the last footnote of his opinion did he address the issue, and he was distressingly evasive. He began with the apparent concession that "[t]he distinction between 'productive' and 'unproductive' uses may be helpful in calibrating the balance, but it cannot be wholly determinative." ¹²⁶ The thrust of the balance of the footnote, however, was that productivity is too vague or controversial a criterion to be of any analytical use. Teachers who copy educational programs for their own edification, voters who copy news programs in order to decide how to vote, hospital patients who copy entertainment programs  [*1686]  they would otherwise miss in order to improve their "psychological well-being" are all engaged in arguably productive uses. ¹²⁷ If those activities fit the bill, Justice Stevens implied, what does not?

In Harper & Row, Justice Brennan argued for a more limited version of the approach advocated by Justice Blackmun in Sony. Instead of proposing a general productivity requirement, Justice Brennan contended that uses of copyrighted material that contribute to "the spread of knowledge and information" should be treated especially leniently under the fair use doctrine. ¹²⁸ Surprisingly, Justice O'Connor accepted Justice Brennan's implicit equation of news reporting with productivity, but she went on to minimize its importance. She began by acknowledging the relevance of "[t]he fact that an article arguably is 'news' and therefore a productive use." ¹²⁹ However, the manner in which Justice O'Connor analyzed The Nation's invocation of this factor effectively reduced it to insignificance. She insisted that, although The Nation was free "to be the first to publish information," it was not entitled to "mak[e] a 'news event' out of [an] unauthorized first publication of a noted figure's copyrighted expression." ¹³⁰ As Justice O'Connor conceded, information is never protected by the copyright laws, ¹³¹ so a person who merely reports information would never need to invoke the fair use doctrine. Fair use only becomes an issue when the reporter has reproduced "'copyrightable expression.'" ¹³² Under those circumstances, she seemed to hold, the reporter stands on no better footing than any other plagiarist.

It would be an exaggeration to say that the decision in Sony and Harper & Row have expunged the concept of productivity from the fair use doctrine. ¹³³ In combination, however, the two decisions have sharply reduced the role played by this factor.

C. The Underlying Objectives

Resolution of a fair use case using the catalogue of factors just reviewed will rarely be a mechanical exercise. ¹³⁴ The individual criteria  [*1687]  will often be difficult to construe and apply. ¹³⁵ When they point in different directions, weighing the various factors will not be easy. ¹³⁶ And there always remains the possibility that a factor other than those discussed by the Court should be taken into account. ¹³⁷

Confronted with these various sources of uncertainty, a judge trying to decide a fair use case is likely to refer directly to the underlying purposes of the doctrine. In other words, to resolve the dispute before him, he will ask the questions that Professor Fuller contended must always be asked if a piece of positive law is to be interpreted responsibly: "What can this rule be for? What evil does it seek to avert? What good is it intended to promote?" ¹³⁸ For answers, he most likely will return to the two Supreme Court opinions. ¹³⁹ A close reading of them will unearth four objectives that the Court apparently believes are and should be served by copyright law in general and by the fair use doctrine in particular.

The first and most prominent of the goals is social utility. As Justice Stevens explained in Sony, the elaborate combination of grants and reservations that comprise the Copyright Act is designed to advance the public welfare by rewarding creative intellectual effort sufficiently to encourage talented people to engage in it, while at the same time making the fruits of their genius accessible to as many people as possible as quickly and as cheaply as possible. ¹⁴⁰ Congress cannot anticipate and provide a resolution for every situation in which these two policies stand in tension. The fair use doctrine enables the judiciary to permit unauthorized uses of copyrighted works in particular situations when doing so will result in wider dissemination of those works without seriously eroding the incentives for artistic and intellectual innovation. ¹⁴¹

 [*1688]  The utilitarian theory just outlined is undoubtedly the most venerable and oft-recited of the justifications for the American law of intellectual property. The constitutional provision upon which the pertinent statutes rest emphasizes the maintenance of incentives for creativity, ¹⁴² and the Supreme Court, when construing provisions of those acts, has frequently stressed that the public interest in generating and disseminating original works is the ultimate goal of the law. ¹⁴³ Most commentators ¹⁴⁴ and lower courts ¹⁴⁵ have taken the same position. It is therefore not surprising that this theme looms large in all of the opinions in Sony and Harper & Row. ¹⁴⁶

The second of the four goals is distilled in a single sentence: "The rights conferred by copyright are designed to assure contributors to the store of knowledge a fair return for their labors." ¹⁴⁷ This statement, though it appears in the midst of Justice O'Connor's discussion of the need to balance incentives for creativity against public access, describes an objective fundamentally different from social utility. The notion it embodies is that authors and inventors deserve a reward for  [*1689]  their labor and should be given it regardless of whether they would continue their work in the absence of such compensation.

This conception of authors' entitlements, though it has never dominated the Anglo-American law of intellectual property, has long had a place in it. For example, in an influential early statement of the function of the law, Lord Mansfield mingled the notions of desert and social planning:

[W]e must take care to guard against two extremes equally prejudicial; the one, that men of ability, who have employed their time for the service of the community, may not be deprived of their just merits, and the reward of their ingenuity and labour; the other, that the world may not be deprived of improvements, nor the progress of the arts be retarded. ¹⁴⁸

In an important modern opinion, Justice Stewart relied upon Lord Mansfield's statement in depicting as consistent the goals of "secur[ing] a fair return for an 'author's' creative labor" and "stimulat[ing] artistic creativity for the general public good." ¹⁴⁹ Lower court opinions crafting the fair use doctrine have also occasionally drawn upon the idea that creative labor ought to be rewarded. ¹⁵⁰ The imprimatur of the majority opinion in Harper & Row ¹⁵¹ will undoubtedly contribute to the currency and influence of the theory.

 [*1690]  The third of the objectives identified by the Court is the protection of what may loosely be described as "personal rights" of artists and authors. This idea surfaces in two contexts in the majority opinion in Harper & Row. First, in the course of her rejection of The Nation's first amendment defense, Justice O'Connor insists that the law must take into account, not only a copyist's right to speak, but also the copyright owner's "'right to refrain from speaking,'" ¹⁵² suggesting that the doctrine should be construed to protect the privacy rights of authors and artists. ¹⁵³ The other, related manifestation of this general theme is the Court's expressed concern to protect an author's "personal interest in creative control." ¹⁵⁴ One of the facts that counted against The Nation was that its article "was hastily patched together and contained 'a number of inaccuracies.'" ¹⁵⁵ The premise of the Court's invocation of this circumstance is that an artist's desire to shape the manner in which his creations are apprehended should be considered in deciding what uses are fair.

Like the desert theory, the notion that artists' personal rights deserve protection has long played a subordinate but significant role in Anglo-American intellectual property law. ¹⁵⁶ When it reformed the Copyright Act in 1976, Congress in several provisions implicitly approved this objective. ¹⁵⁷ Prior to the decision in Harper & Row, however, few judicial decisions construing the fair use doctrine took into account either privacy rights or rights of artistic integrity. By explicitly incorporating such concerns in her analysis, Justice O'Connor has significantly expanded the set of principles upon which the doctrine rests.

 [*1691]  The fourth and final of the underlying objectives has already been discussed at length in connection with the "propriety and custom" factor. ¹⁵⁸ As noted there, the notion that seems to have prompted Justice O'Connor to add this criterion to the fair use doctrine is that copyright law should track (or at least take into account) popular conceptions of decent behavior and the habits of communities to which disputants belong.

Unfortunately, having identified these objectives, the judge struggling with a difficult case is likely to discover that, for two reasons, they are no more helpful than the enumerated "factors." First, the generality of the goals makes them hard to apply. What exactly is a "fair return" for creative labor? How exactly should the policies of encouraging creativity and disseminating creations be balanced? More precise answers than the Court has yet provided would be necessary to give these considerations bite.

Second, the judge will probably find that the underlying goals, like the "factors," point in different directions. ¹⁵⁹ To a large extent, this difficulty derives from the fact that the four objectives are drawn from four separate and markedly different traditions in political or legal theory, and the Court has failed to combine them into an integrated approach. The tradition that underlies the first and primary theme is utilitarianism; the idea that the fair use doctrine should be adjusted to advance the public good by providing incentives both for creating original works and for facilitating their dissemination is just a particular application of the general proposition that the law should be consciously crafted to promote "general happiness." ¹⁶⁰ The tradition corresponding to the second theme is the version of the theory of natural rights popularized by John Locke; the seed of the idea that authors deserve a fair return for their creative labor is Locke's argument in chapter five of The Second Treatise that labor upon an unowned object gives rise to a natural property right in that object, and that the principal duty of the state is to protect through the  [*1692]  positive law natural property rights so acquired. ¹⁶¹ The foundation of the third objective is the notion, first developed by Hegel and his followers, that property rights should be allocated in the fashion that best enables persons to develop and exercise their faculties. ¹⁶² Finally, the fourth objective issues from a tradition emphasizing the limited power of the positive law and the degree to which it must and should track customs and popular understandings -- a tradition whose most insightful exponent was David Hume. ¹⁶³

Arguments that spring from such diverse outlooks do not mesh comfortably. They may not be altogether incompatible; the Justices might have been able, had they put their minds to it, to knit the four objectives into a coherent vision of the fair use doctrine -- or at least to provide some guidance as to which of the four goals should predominate in which contexts. But none of the opinions in Sony or Harper & Row ventures such an integration. In its absence, a judge who recurs to the Court's discussion in the hope of determining what resolution of a fair use case would advance the ultimate ends of the doctrine will likely receive several inconsistent signals.

D. Confusion and Its Costs

The preceding two sections identified a number of independent defects in the current fair use doctrine. As currently formulated, the  [*1693]  market-impact factor that the Court has pushed to the center of the field is unhelpful. ¹⁶⁴ Two of its satellite factors -- the commercial/noncommercial and amount-of-copying tests -- suffer from crucial ambiguities. ¹⁶⁵ The latter test, moreover, fosters wasteful registration and storage practices. ¹⁶⁶ In most cases, there will be no defensible referent for the newly minted "propriety and custom" factor. ¹⁶⁷ Last but not least, the normative foundation of the doctrine is fragmented. ¹⁶⁸

Of the many reasons to regret such a state of affairs, two bear emphasis. First, the disarray of the doctrine impairs the ability of the creators and users of intellectual products to ascertain their rights and to adjust their conduct accordingly. ¹⁶⁹ The most telling indication of the seriousness of this problem is the character of the advice currently being given the members of those groups by their lawyers. A few examples should make the point. A recent article in the Journal of College and University Law, the stated purpose of which is to assist lawyers representing universities in counselling their clients, concludes lamely:

The implications of Nation Enterprises for scholars who wish to use unpublished materials in scholarly publications are unclear but encourage caution. . . . College and university counsel advising scholars . . . need to create an awareness of the copyright implications in materials which many researchers may not be aware carries with it significant copyright concerns. In the absence of permission from the copyright owners of the unpublished material, careful consideration should be given to the copyright implications of using unpublished materials and alternatives suggested which do not intrude upon the rights granted by copyright. ¹⁷⁰

 [*1694]  Another article, directed at professional historians, also pertaining to the use of unpublished materials, ¹⁷¹ is rife with equally tentative judgments and suggestions. ¹⁷² Finally, as almost any college teacher can attest, the information presently being given faculty by university counsel regarding how much copyrighted material they may reproduce for classroom use is distinctly unhelpful. ¹⁷³ That persons affected by the fair use doctrine do not know -- and cannot find out -- their entitlements ¹⁷⁴ may not be the worst imaginable situation, but it is not a happy one.

Second and more fundamentally, the incoherence of the doctrine exacerbates the stagnation and inconclusiveness of contemporary moral and political discourse. It is routine nowadays for both popular and academic arguments over how a person or government ought to behave to devolve into shouting matches, in which the participants invoke competing "standards of justice or generosity or duty," ¹⁷⁵ each torn from the historical or philosophical context in which it evolved, without advancing our understanding of either the competing standards or the matter at issue. ¹⁷⁶ If we wish to escape such conceptual  [*1695]  ruts -- and especially if we also wish to foster a culture less riven and more nourishing than the one in which we live ¹⁷⁷ -- we must develop a new way of talking. Legal thought and practice afford some of us opportunities to contribute to that project. The current fair-use doctrine, far from aiding in the effort, helps perpetuate the problem, by reinforcing the impression that, when confronted with a question of public policy, we can do no better than "balance" inconsistent claims derived from conventional, incommensurable premises. ¹⁷⁸

Part III: What is To Be Done?



FOOTNOTES:

n39 Because the principal objective of the Article is to show that the fair use doctrine can and should be improved, the analysis of this Part concentrates on the ways in which the Court defined and applied the doctrine in Sony and Harper & Row and does not address the question whether the cases were rightly decided. The reader should not infer from the paucity of overt criticism of the Court's judgments that Sony and Harper & Row were "easy cases" -- that, no matter how the doctrine was construed, they should have and would have been decided as they were. One indication of their difficulty is the level of disagreement among the judges who passed upon them; in each case, the district court was reversed by the court of appeals, the court of appeals was reversed by the Supreme Court, and the Supreme Court was closely divided. The penultimate section of the Article returns to the cases and considers how they might have been resolved had the Court employed the "reconstructed" doctrine outlined in Part V.

n40 Sony, 464 U.S. at 448; Harper & Row, 471 U.S. at 560; id. at 588 (Brennan, J., dissenting). This phrase appears in the House Report on section 107, see H.R. REP. NO. 1476, 94th Cong., 2d Sess. 65 (1976), reprinted in 1976 U.S. CODE CONG. & ADMIN. NEWS 5659, 5679, which in turn derived the language from several lower-court opinions developing the fair use doctrine.

n41 Harper & Row, 471 U.S. at 552.

n42 Sony, 464 U.S. at 479-80 (Blackmun, J., dissenting). The reference in the text and in Justice Blackmun's opinion to "rules" calls to mind the conventional distinction between a rule and a standard and suggests that the doctrine crafted by the Court is simply an example of the latter. Cf. Kennedy, Form and Substance in Private Law Adjudication, 89 HARV. L. REV. 1685, 1687-89 (1975) (describing a "rule" as a legal norm that provides that simultaneous occurrence of a set of "easily distinguishable factual aspects of a situation" shall have a determinate consequence and a "standard" as a norm that requires adjudicators to assess the facts of each situation in light of a designated substantive objective or social value). The plausibility of that characterization of the Court's rulings is undercut, however, by the importance to the doctrine of a set of "factors," see infra section II.B, and by the incoherence of the values the doctrine ostensibly serves, see infra section II.C. Under such conditions, use of the traditional nomenclature would be more distracting than helpful.

n43 The list of statutory factors is set forth in note 4 above.

n44 Indeed, the Court left open the possibility that lower courts in future cases would identify other factors relevant to the problems presented to them. See Harper & Row, 471 U.S. at 560 ("The factors enumerated in [section 107] are not meant to be exclusive. . . ."); id. at 588 (Brennan, J., dissenting) ("The [four] statutory factors . . . provide substantial guidance to courts undertaking the proper fact-specific inquiry" but "are not necessarily the exclusive determinants of the fair use inquiry.").

n45 17 U.S.C. § 107(4) (1982).

n46 464 U.S. at 450. Fuller discussion of the utilitarian premises of Justice Stevens' opinion can be found in section II.C below.

n47 Id. at 450-51. In the course of his analysis, Justice Stevens discussed several other considerations, but all of them either depended on or were subordinated to the market-impact factor.

n48 Id. at 421.

n49 Iowa State Univ. Research Found. v. American Broadcasting Cos., 621 F.2d 57 (2d Cir. 1980). The case provides a nice illustration of Justice Blackmun's argument. Iowa State owned the copyright on a short film about the life of a college wrestler who subsequently qualified for the United States Olympic team. ABC obtained a copy of the film and, without permission, used some of the footage in a brief biography of the wrestler broadcast during its coverage of the Olympics. See id. at 59. In response to the infringement action filed by Iowa State, ABC contended that, because it enjoyed the exclusive right to televise the Olympics, the unauthorized copying did not "harm" Iowa State in any way, because, without ABC's permission, Iowa State could not have obtained access to the television viewers. In rejecting ABC's contentions, the court of appeals reasoned:

ABC did foreclose a significant potential market to Iowa -- sale of its film for use on television in connection with the Olympics. . . . Iowa had no right to insist that the network use its film, but its copyright entitled it to attempt to exploit the commercial market controlled by ABC, and, if it could not, to withhold permission to use the film in that market.

Id. at 62.

n50 464 U.S. at 485 (Blackmun, J., dissenting); see also id. at 497-98.

n51 Id. at 485.

n52 If Justice Blackmun's definition of "potential market" is accepted, the only questionable aspect of his reasoning is his contention that finding time-shifting to be a fair use would prevent the studios from exploiting the market constituted by the time-shifters. The owners of copyrighted programs ordinarily derive their compensation not through direct payments from viewers, but from license fees paid to them by the networks. Those fees are determined in substantial part by the number of persons who see the advertisements embedded in the programs when broadcast. The operation of this scheme might not appear to be affected by whether the viewers see the programs at the time they are broadcast or at a later time. Of the many responses to this point, the following is sufficient: a substantial proportion of the persons who watch taped programs use the "fast-forward" buttons on their VCRs to minimize the amount of time they must spend watching advertisements. See id. at 483 n.35 (noting that surveys submitted by both parties indicated that time-shifters "avoided [commercials] at least 25% of the time"). The advertisers, who are aware of this practice, are understandably reluctant to pay for the portion of the audience who engage in it. See Advertising: If Viewers Tune out TV Ads, N.Y. Times, Aug. 12, 1982, at D17, col. 3; cf. Mandese, Tale of the Tape, ADWEEK (Eastern ed.), Aug. 11, 1986, at 1 (documenting the advertisers' continued unhappiness with the situation). Thus, at a minimum, the copyright holders have been denied the opportunity to derive a profit from the market composed of the subset of time-shifters who delete the advertisements.

n53 The crucial passages in her opinion are: "To negate fair use one need only show that if the challenged use 'should become widespread, it would adversely affect the potential market for the copyrighted work,'" 471 U.S. at 568 (quoting the majority opinion in Sony but adding emphasis to the word "potential" and citing the portions of Justice Blackmun's dissent in which he defines the phrase expansively); "[t]his inquiry must take account not only of harm to the original but also of harm to the market for derivative works," id. (citing Iowa State).

n54 464 U.S. at 451 (emphasis in original); see also id. at 456 (using similar language). Indeed, Justice Stevens argued that the likelihood of such harm could be presumed in cases involving commercial uses. See id. at 451; infra note 60.

n55 471 U.S. at 566-67 (quoting 1 M. NIMMER, NIMMER ON COPYRIGHT § 1.10[D], at 1-87 (1984)); see also id. at 568 ("'If the defendant's work adversely affects the value of any of the rights in the copyrighted work . . . the use is not fair.'") (quoting 3 M. NIMMER, NIMMER ON COPYRIGHT § 13.05[B], at 13-77-13-78 (1984)).

n56 All of the rights conferred on the copyright owner by section 106 are alienable and, in practice, are commonly alienated.

n57 It might be objected that such cases would not be so rare -- because many potential users if denied the right to copy modest portions of a copyrighted work would refuse "on principle" to purchase permission to do so. Perhaps so, but surely it would be odd to let the availability of the fair use defense turn on whether the defendant was convinced that he ought to be able to use the plaintiff's work for free. Moreover, defendants' incentive to misrepresent their attitudes on this score would make it hazardous to try to take their convictions into account when administering the doctrine.

n58 How the magnitude of injury ought to figure in the fair use calculus remains to be seen -- and will be considered in some detail in Parts IV and V. But that it should be the magnitude that matters seems undeniable.

n59 This factor has a complex pedigree. In several early fair use cases, lower courts emphasized the "commercial character" of the challenged activity in refusing to excuse it. See, e.g., Henry Holt & Co. v. Liggett & Myers Tobacco Co., 23 F. Supp. 302, 304 (E.D. Pa. 1938). In 1966, an influential decision by the Second Circuit repudiated this factor; in Rosemont Enters, v. Random House, Inc., 366 F.2d 303 (2d Cir. 1966), the court reasoned that, "whether an author or publisher has a commercial motive . . . is irrelevant to a determination of whether a particular use . . . constitutes a fair use." Id. at 307. Other courts were not willing to go so far, asserting instead that commercial motives, though not decisive, were relevant in deciding whether to privilege a particular use of a copyrighted work. See, e.g., Triangle Publications v. Knight-Ridder Newspapers, 626 F.2d 1171, 1175 (5th Cir. 1980) (holding that "any commercial use tends to cut against a fair use defense"). Eventually even the Second Circuit retreated from the stand it had taken in Rosemont, concluding that, "[f]or a determination whether the fair use defense is applicable . . . it is relevant whether the [copyrighted work was] used . . . predominantly for commercial exploitation." Meeropol v. Nizer, 560 F.2d 1061, 1069 (2d Cir. 1977); see also, e.g., Consumers Union of United States v. General Signal Corp., 724 F.2d 1044, 1049 (2d Cir. 1983) (stating that the commercial nature of the use is relevant but not decisive). For a thorough review of this line of cases, see W. PATRY, cited above in note 6, at 72-91. Section 107(1) of the 1976 Act reflects what had become more-or-less orthodox doctrine by the time of its enactment; in deciding whether a particular use is fair, a court is directed to consider "the purpose and character of the use, including whether such use is of a commercial nature or is for nonprofit educational purposes." 17 U.S.C. § 107(1) (1982).

n60 In his majority opinion in Sony, Justice Stevens highlighted this issue: "Although not conclusive, the first factor [mentioned in section 107] requires that 'the commercial or nonprofit character of an activity' be weighed in any fair use decision." 464 U.S. at 448-89 (quoting H.R. REP. NO. 1476, 94th Cong., 2d Sess. 66 (1976)). More specifically, Justice Stevens used this factor to allocate the burden of proof on the question of market impact. To prevail in a suit challenging "a noncommercial use of a copyrighted work," the copyright holder must prove "by a preponderance of the evidence that some meaningful likelihood of future harm exists"; by contrast, "[i]f the intended use is for commercial gain, that likelihood may be presumed." Id. at 451. In Harper && Row, Justice O'Connor took a similar line, expressly approving Justice Stevens' use of this factor to establish a presumption of injury. See 471 U.S. at 562. At least one lower court has since taken this presumption seriously. See Financial Information, Inc. v. Moody's Investors Serv., 751 F.2d 501, 508-09 (2d Cir. 1984).

n61 See supra note 60.

n62 For example, suppose that 25 law students pooled their resources and bought a home photocopier, then borrowed from the library a set of the casebooks assigned in their courses and made 25 copies of each volume. Arguably, this activity would not fit the colloquial definition of "commercial," but it would materially reduce the demand for the casebooks. An activity with greater economic importance is the common practice of taping phonorecords or compact discs on casette recorders. Were this practice forbidden, the profits available to record and "CD" manufacturers -- and thus to performing artists -- would undoubtedly be substantially higher. (It is not altogether clear that the activity is in fact lawful; for discussion of the issue, see Leete, Betamax and Sound Recordings: Is Copyright in Trouble?, 23 AM. BUS. L.J. 551 (1986).) Finally, consider the longstanding practice of the Natioanl Institutes of Health of copying articles in medical journals and distributing the copies free to medical researchers and libraries. The activity could fairly be described as "noncommercial," but it nevertheless adversely affects the market for the journals in which the articles are published. See Williams & Wilkins Co. v. United States, 487 F.2d 1345 (Ct. Cl. 1973) (holding the use fair), aff'd per curiam by an equally divided Court, 420 U.S. 376 (1975).

n63 Thus, he repeatedly used the term as the opposite of "nonprofit." If the monetary benefits to the user exceed the costs, he implied, the use is "commercial." See, e.g., 464 U.S. at 449. A similar understanding of the term seems to have guided the district court in Sony. See 480 F. Supp. at 449-50, 453-54. Justice Stevens' express reliance on the district court's finding that time-shifting is "a noncommercial, nonprofit activity," 464 U.S. at 449, suggests that he found the lower court's definition congenial.

n64 464 U.S. at 496 (Blackmun, J., dissenting).

n65 471 U.S. at 562. Applying this test, Justice O'Connor reasoned that, because The Nation had profited from its use of excerpts of President Ford's memoirs and did not pay the customary price for serialization rights, its conduct must be deemed commercial. Justice Brennan in dissent argued that the inclusion of "news reporting" in the introductory sentence of section 107 required that the presumption against commercial uses established by Sony not be applied to news businesses. See id. at 592 & n.16 (Brennan, J., dissenting). However, Justice Brennan did not challenge either Justice O'Connor's general definition of "the profit/nonprofit distinction" or the applicability of the Sony presumption to activities not enumerated in section 107 itself.

n66 In particular, Justice O'Connor's definition, insofar as it differs from Justice Stevens', is circular: there will be a "customary price" to be paid for the right to engage in a particular use if and only if unauthorized engagement in that use is held to violate the Copyright Act; the existence of a customary price, therefore, cannot determine whether the use violates the Act.

n67 For examples of recent cases in which courts have struggled to determine whether the use in question was "commercial," see Hustler Magazine v. Moral Majority, 796 F.2d 1148, 1152-53 (9th Cir. 1986) (rejecting the district court's conclusion that copying an obscene parody for the purpose of raising money to finance a suit against the parodist was not a commercial use), and Fisher v. Dees, 794 F.2d 432, 437 (9th Cir. 1986) (suggesting that "many parodies . . . may be 'more in the nature of an editorial or social commentary than . . . an attempt to capitalize financially on the plaintiff's original work'" (quoting Pillsbury Co. v. Milky Way Prods., 215 U.S.P.Q. 124, 131 (BNA) (N.D. Ga. 1981))).

n68 See, e.g., Stanley v. Columbia Broadcasting Sys., 35 Cal. 2d 653, 661, 221 P.2d 73, 78 (Cal. 1950) (en banc); W. PATRY, supra note 6, at 439-41; cf. 3 M. NIMMER, NIMMER ON COPYRIGHT § 1305, at 13-62 & n.2 (1987) (observing that, although under common law copyright the fact that a work was unpublished was "a factor tending to negate the defense of fair use," it did not bar invocation of the doctrine).

n69 See 17 U.S.C. §§ 102(a), 104(a) (1982).

n70 471 U.S. at 554 (quoting S. REP. NO. 473, 94th Cong., 1st Sess. 64 (1975)).

n71 Justice O'Connor argued that "the potential damage to the author from judicially enforced 'sharing' of the first publication right with unauthorized users of his manuscript is substantial," id. at 553, in part because this practice would threaten the author's valuable "property interest in exploitation of prepublication rights," id. at 555.

n72 See id. at 555.

n73 Id. at 555, 564.

n74 Id. at 564; see also id. at 555 ("Under ordinary circumstances, the author's right to control the first public appearance of his undisseminated expression will outweigh a claim of fair use.").

n75 See id. at 595-96 (Brennan, J., dissenting).

n76 For example, in Folsom v. Marsh, 9 F. Cas. 342 (C.C.D. Mass. 1841) (No. 4901), perhaps the most famous of the early cases, Justice Story, riding circuit, was called upon to determine whether reproducing several copyrighted letters as part of a biography of their author constituted a "justifiable use" of those letters. Among the factors Justice Story identified as critical to the analysis was the "quantity and value of the materials used." Id. at 348.

n77 See, e.g., Leon v. Pacific Tel. & Tel. Co., 91 F.2d 484, 486 (9th Cir. 1937) (denying that "wholesale copying and publication of copyrighted material can ever be fair use"). But see Williams & Wilkins Co. v. United States, 487 F.2d 1345, 1353 & n.12 (Ct. Cl. 1973) (calling Leon's assertion an "overbroad generalization"), aff'd per curiam by an equally divided Court, 420 U.S. 376 (1975).

n78 Thus, in his opinion in Folsom, Justice Story quoted approvingly Lord Cottendam's comment:

When it comes to quantity, it must be very vague. One writer might take all the vital part of another's book, though it might be but a small proportion of the book in quantity. It is not only quantity, but value, that is always looked to. It is useless to refer to any particular cases, as to quantity.

9 F. Cas. at 348 (quoting Bramwell v. Halcomb, 40 Eng. Rep. 1110 (1836)). For a similar admission, see, for example, Chicago Record-Herald Co. v. Tribune Assoc., 275 F. 797, 799 (7th Cir. 1921) (stating that substantiality "cannot be determined alone by lines or inches").

n79 In the Folsom case itself, Justice Story shifted confusingly from discussion of the portion (quantitative and qualitative) of the allegedly infringing biography that consisted of copied material to discussion of the importance of the copied material to the copyrighted volume from which it was taken. See 9 F. Cas. at 349. Similar uncertainty as to how the amount of copying is to be measured infects most other opinions addressing the issue. See, e.g., Roy Export Co. v. CBS, Inc., 503 F. Supp. 1137, 1145 (S.D.N.Y. 1980) (discussing both qualitative and quantitative substantiality), aff'd on other grounds, 672 F.2d 1095 (2d Cir. 1982). An exception is Sheldon v. Metro-Goldwyn Pictures Corp., 81 F.2d 49 (2d Cir.), cert. denied, 298 U.S. 669 (1936), where Judge Learned Hand tried (unsuccessfully) to lay one of the questions to rest: "[N]o plagiarist can excuse the wrong by showing how much of his work he did not pirate," id. at 56-57.

n80 17 U.S.C. § 107(3) (1982).

n81 Although Justice Stevens' opinion in Sony makes no explicit reference to the ambiguity of this factor, his sensitivity to the difficulties created by the precisely analogous aspect of current "takings" doctrine, see Keystone Bituminous Coal Ass'n v. DeBenedictis, 107 S. Ct. 1232, 1248 (1987) (Stevens, J.), suggests that concerns other than a desire to let Sony escape liability might have led him to try to reduce the role the "amount of copying" factor plays in the fair use calculus.

n82 As the plaintiff studios pointed out, and as Justice Blackmun insisted in his dissent, time-shifting ordinarily entails the verbatim copying of an entire copyrighted television program -- a fact that would seem to incline strongly toward a finding of liability. See 464 U.S. at 497 (Blackmun, J., dissenting). Justice Stevens rejected this inference in a single sentence:

[W]hen one considers the nature of a televised copyrighted audiovisual work, and that time-shifting merely enables a viewer to see such a work which he had been invited to witness in its entirety free of charge, the fact that the entire work is reproduced does not have its ordinary effect of militating against a finding of fair use.

464 U.S. at 449-50.

n83 471 U.S. at 566.

n84 See id. at 564-65. Justice O'Connor assumed, arguendo, that only the direct quotation of 300 words from the book violated section 106 of the Act. See supra p. 1667. The book itself was approximately 200,000 words long. See 471 U.S. at 598 (Brennan, J., dissenting). Thus, The Nation reproduced only 0.15% of the manuscript.

n85 See 471 U.S. at 564-65.

n86 Id. at 565, 566.

n87 Id. at 565.

n88 See id. (quoting Sheldon v. Metro-Goldwyn Pictures Corp., 81 F.2d 49, 56-57 (2d Cir.), cert. denied, 298 U.S. 669 (1936) (discussed in note 79 above)).

n89 See id. On this basis, Justice O'Connor emphasized the fact that, although the ratio of copied material to the size of Ford's manuscript was tiny, see supra note 84, the ratio of the copied material to the size of The Nation's article was 13% -- which she deemed a sizeable figure. See 471 U.S. at 565-66.

n90 The problem is illustrated by a pair of cases. In 1977, the publisher of the Miami Herald reproduced and published copyrighted covers of TV Guide magazine to promote by comparative advertising its own competing listing of TV programs. One of the factors relied upon by the Fifth Circuit in concluding that the Herald's behavior constituted a fair use was the small proportion of an issue of TV Guide represented by its cover. The court dismissed as inconsistent with "logic and common sense" the plaintiff's contention that the cover of a magazine "is separately copyrighted and that therefore [the defendant had] reproduced an entire copyrighted work." Triangle Publications v. Knight-Ridder Newspapers, 626 F.2d 1171, 1177 & n.15 (5th Cir. 1980) (emphasis in original). In 1984, the Eleventh Circuit was presented with a similar problem. A commercial "newsclipping organization" recorded a brief "news feature" aired by a local television station as part of a half-hour news program and then sold the recording to the college that was the subject of the story. In ruling that the copyist could not invoke the fair use doctrine, the court found that the news story "stands alone as a copyrighted work in this case" and that therefore the defendant had "copied an entire work." The court rested its finding on the facts that:

[t]he Floyd Junior College story stands alone as a coherent narrative, and WXIA saves it as a distinct unit for future reference apart from the rest of the March 11 broadcast. [Moreover, t]he Register of Copyrights issued a certificate of copyright for the Floyd Junior College segment [as well as] for the entire broadcast.

Pacific & Southern Co. v. Duncan, 744 F.2d 1490, 1497 (11th Cir. 1984). The considerations invoked by the Eleventh Circuit to distinguish the case before it from Triangle Publications have little to recommend them. The manner in which a copyright owner stores his materials and the way in which he frames his copyright applications would seem to have little bearing on the legitimacy of copying portions of those materials. The establishment of such criteria is likely merely to induce sophisticated authors and publishers to submit multiple applications and to subdivide their files. And the question whether a segment of a larger work "stands alone as a coherent narrative" is nearly useless as a test. Would the cover of a magazine, often produced by an independent artist, qualify? What of the chapter of a scholarly book that could be published as an article?

n91 For example, in Time v. Bernard Geis Assocs., 293 F. Supp. 130 (S.D.N.Y. 1968), the district court argued that "[f]air use presupposes 'good faith and fair dealing'" and based its conclusion that the defendants' use of copyrighted photographs of President Kennedy's assassination did not give rise to liability partly on its finding that the defendants' conduct met that standard. Id. at 146 (quoting Schulman, Fair Use and the Revision of the Copyright Act, 53 IOWA L. REV. 832 (1968)). Taking a slightly different tack in Rosemont Enters. v. Random House, Inc., 366 F.2d 303 (2d Cir. 1966), the Second Circuit emphasized that, when writing a biography, "it is both reasonable and customary . . . to refer to and utilize earlier works." Id. at 307. On that basis, the Court held that the inclusion in a biography of Howard Hughes of paraphrases of portions of two copyrighted magazine articles about Hughes did not give rise to liability. See id.; see also Holdridge v. Knight Publishing Corp., 214 F. Supp. 921, 924 (S.D. Cal. 1963) (holding that appropriation so extensive as to be "neither reasonable nor customary" cannot constitute a fair use).

n92 See supra note 9.

n93 The importance of this factor to Justice O'Connor's approach is suggested by the way in which she frames her discussion. At one point she suggests that the fair use analysis might be reduced to the following inquiry: "'[W]ould the reasonable copyright owner have consented to the use [in question]?'" 471 U.S. at 550 (quoting A. LATMAN, FAIR USE OF COPYRIGHTED WORKS 15 (1958)). The "reasonable person" who figures in this formula cannot be the rational economic actor familiar to modern lawyers, because an actor whose sole interest is maximizing his own welfare would demand a fee from every person who wished to use his work -- unless the use in question would somehow indirectly redound to the actor's benefit. Thus, the "reasonable person" envisioned by Justice O'Connor must be someone whose reasonableness consists of something like neighborliness -- either decency itself or a willingness to defer to popular conventions of decent behavior.

n94 Id. at 562 (quoting 3 M. NIMMER, NIMMER ON COPYRIGHT §13.05[A], at 13-72 (1984)).

n95 See, e.g., id. at 563 ("The Nation knowingly exploited a purloined manuscript. . . . Fair use 'distinguishes between a "true scholar and a chiseler who infringes a work for personal profit."'" (quoting Wainwright Sec. v. Wall Street Transcript Corp., 558 F.2d 91, 94 (2d Cir. 1977), cert. denied, 434 U.S. 1014 (1978) (quoting Hearings on Bills for the General Revision of the Copyright Law Before the House Comm. on the Judiciary, 89th Cong., 1st Sess. 1706 (1966)))); id. at 542 ("Working directly from the purloined manuscript. . . ."); id. at 556 (disdaining "the piracy of verbatim quotations for the purpose of 'scooping' the authorized first serialization").

n96 For indications that the lower courts have gotten the message, see Fisher v. Dees, 794 F.2d 432, 437 (9th Cir. 1986); Haberman v. Hustler Magazine, 626 F. Supp. 201, 211 (D. Mass. 1986); and Radji v. Khakbaz, 607 F. Supp. 1296, 1300-01 (D.D.C. 1986).

n97 See 471 U.S. at 593-94 (Brennan, J., dissenting).

n98 Id. at 593. In a footnote, Justice Brennan cited five articles by the New York Times reporting on the contents of forthcoming books, arguing that the custom evidenced by those articles suggested that The Nation's similar conduct constituted a fair use. See id. at 591 n.14. Justice O'Connor dismissed this argument on the ground that the articles cited by Justice Brennan had not been properly introduced into evidence and were "not a proper subject for this Court's judicial notice." 471 U.S. at 562 n.7. Another example of invocation of industry practice to resolve a fair use case is provided by Triangle Publications v. Knight-Ridder Newspapers, 626 F.2d 1171 (5th Cir. 1980). See supra note 90. In justifying its decision, the court of appeals emphasized that the Miami Herald's confrontational advertising campaign was "done in a manner which is generally accepted in the advertising industry." 626 F.2d at 1176 & n.13.

n99 An example familiar to most first-year law students involves the whaling industry in the nineteenth century. When disputes between two whalers claiming ownership of the same whale reached the courts, judges often deferred to the customs prevailing in the sector of the industry in which the disputants were operating. See, e.g., Swift v. Gifford, 23 F. Cas. 558 (D.C.D. Mass. 1872) (No. 13,696); cf. H. MELVILLE, MOBY-DICK 1216-19 (1st ed. 1851; Library of America ed. 1983) (describing different customs prevailing in different parts of the whale fishery).

n100 In combination, three such arguments have considerable force. First, in cases of this sort, both parties usually have participated in the industry or community in question for some time. The party disadvantaged in the case at bar by the customary standard probably benefitted from it in the past and ought not question it now. Second, many customary practices arise because the members of the community come to recognize that they are "efficient" -- in the sense that they will maximize the total profits available to industry participants and that each participant is more likely to gain by them than lose by them. See Ghen v. Rich, 8 F. 159 (D.C.D. Mass. 1881) (recognizing the reasonableness and efficiency of a whaling industry custom); Ellickson, Of Coase and Cattle: Dispute Resolution Among Neighbors in Shasta County, 38 STAN. L. REV. 623, 685-87 (1986). Third, most participants in a community or industry are aware of prevailing customs, but relatively few are aware of the content of relevant positive law. If the two systems of norms diverge, the result will be wasteful conflict between the people who assume custom governs and those who know and stand upon their legal rights.

n101 Certainly no evidence was introduced by The Nation that Harper & Row engaged in such activities. The chances are somewhat more likely that The Reader's Digest, the other plaintiff, was a participant in the relevant sector of the industry.

n102 471 U.S. at 563 (quoting Wainwright Sec. v. Wall Street Transcript Corp., 558 F.2d 91, 94 (2d Cir. 1977), cert. denied, 434 U.S. 1014 (1978) (quoting Hearings on Bills for the General Revision of the Copyright Law Before the House Comm. on the Judiciary, supra note 95, at 1706)).

n103 Cf. B. ACKERMAN, PRIVATE PROPERTY AND THE CONSTITUTION 94-100 (1977) (discussing the difficulties of ascertaining the "dominant" popular view regarding the legitimate scope of governmental authority to interfere with property rights). That the Supreme Court in Harper & Row was wrong to suppose such a consensus existed with regard to the legitimacy of The Nation's behavior is suggested by the diversity of public reactions first to the court of appeals' decision and then to the Supreme Court's ruling. See, e.g., Serrill, When a Scoop Is Piracy, TIME, June 3, 1985, at 64; Kaplan, No Unity Over Nation, Nat'l L.J., Dec. 5, 1983, at 3.

n104 J. ELY, DEMOCRACY AND DISTRUST 63 (1980) (criticizing constitutional theories founded on the proposition that "'constitutional law must now be understood as expressing contemporary norms'" (quoting Sandalow, Judicial Protection of Minorities, 75 MICH. L. REV. 1162, 1193 (1977))). For general discussion both of the processes by which a society's widely shared values come to reflect the interests of its dominant groups and of the limits on those processes, see P. RICOEUR, LECTURES ON IDEOLOGY AND UTOPIA 261-66 (1986), and Lears, The Concept of Cultural Hegemony: Problems and Possibilities, 90 AM. HIST. REV. 567 (1985).

n105 J. ELY, supra note 104, at 64. For discussion of the general problems entailed by invocation of "shared values" to stabilize adjudication, see R. UNGER, KNOWLEGE AND POLITICS 100-03 (1975).

n106 471 U.S. at 562.

n107 See, e.g., Eckes v. Card Prices Update, 736 F.2d 859, 862 (2d Cir. 1984); Consumers Union of United States, Inc. v. General Signal Corp., 724 F.2d 1044, 1049 (2d Cir. 1983).

n108 See Gorman, Fact or Fancy? The Implications for Copyright, 29 J. COPYRIGHT SOC. 560, 562 (1982). Gorman actually identified five concerns, but the one not mentioned in the text -- that "because many fact works do not permit of varieties of expression by later artists, similarity of appearance between their works and their predecessors'" is often excused, id. -- is best seen as a premise of the second factor discussed below. Gorman's analysis built upon his own prior treatment of the topic, Copyright Protection for the Collection and Representation of Facts, 76 HARV. L. REV. 1569 (1963), as well as a recent study by Professor Denicola, Copyright in Collections of Facts: A Theory for the Protection of Nonfiction Literary Works, 81 COLUM. L. REV. 516 (1981).

n109 Gorman, Fact or Fancy, supra note 108, at 562. Gorman acknowledges that, in some circumstances, these concerns may not be compelling. In particular, the first of the four reasons (the public interest in access to facts) may sometimes suggest that the copyright protection for nonfiction works be strengthened (in order to increase incentives for the discovery and compilation of such facts) rather than weakened. See id. at 562, 586. As applied to "works of history and biography," the second of the concerns has little force. See id. at 586. And the inability of a nonfiction author to claim copyright protection for her "story line" arguably makes it even more critical that the language in which she expresses her argument be protected from piracy. See id. at 586. For further discussion of these and related matters, see sections IV.D.5 and V.C. I.b below.

n110 See 464 U.S. 417, 496-97 (1984) (Blackmun, J., dissenting).

n111 Justice Stevens' discussion of the issue is confined to a few sentences in his final footnote. See id. at 455 n.40. In keeping with his view that economic impact ought to be the central consideration in a fair use analysis, Justice Stevens conceded that some types of copyrighted works, because their value inheres substantially in "secondary markets," may be more vulnerable to devaluation through copying than other types. On that basis, he acknowledged that "[c]opying a news broadcast may have a stronger claim to fair use than copying a motion picture." Id. But the placement of this discussion in a concluding footnote, divorced from his analysis of time-shifting, suggests that he considered it relatively unimportant.

n112 471 U.S. at 594 (Brennan, J., dissenting).

n113 471 U.S. at 563.

n114 The example that Justice O'Connor selected to indicate how her theory would apply to the facts of Harper & Row suggests how rare will be the occasions in which it could be invoked successfully: "Some of the briefer quotes from the memoirs are arguably necessary adequately to convey the facts; for example, Mr. Ford's characterization of the White House tapes as the 'smoking gun' is perhaps so integral to the idea expressed as to be inseparable from it." Id. at 563.

n115 See, e.g., Miller v. Universal City Studios, 650 F.2d 1365, 1368 (5th Cir. 1981).

n116 This reading of Justice O'Connor's opinion is further supported by her assertion that the primary objective of copyright law -- rewarding authors in order to benefit the public -- "applies equally to works of fiction and nonfiction." 471 U.S. at 546.

n117 For example, in Rosemont Enters. v. Random House, Inc. 366 F.2d 303 (2d Cir. 1966), the Second Circuit held that the question of fair use "turns initially" on whether the author "requires some use of prior materials dealing with the same subject." Id. at 307 (emphasis added); see also Meeropol v. Nizer, 560 F.2d 1061, 1070-71 (2d Cir. 1977) (implying that the fair use doctrine applies to verbatim quotations from copyrighted letters only if they are needed for historical accuracy). In Consumers Union of United States v. General Signal Corp., 724 F.2d 1044 (2d Cir. 1983), the same court was asked whether a manufacturer had a right to include in advertisements for its vacuum cleaner excerpts from a favorable copyrighted review of the product. The court's affirmative answer was based in part of its judgment that direct quotation may be "the only valid way precisely to report" such an evaluation. Id. at 1049.

n118 Compare Time v. Bernard Geis Assocs., 293 F. Supp. 130, 146 (S.D.N.Y. 1968) (finding the fair use privilege undiminished even though use of copyrighted photographs merely made the copyist's work "easier to understand") with W. PATRY, supra note 6, at 98 (arguing that a use is not fair merely because it makes an unauthorized user's task easier).

n119 Some sensitivity to this consideration may perhaps be inferred from Justice Stevens' recitation of the district court's factual finding that time-shifting enables many persons to see programs they would otherwise be unable to view because of "'the basic need to work'" or "'the competitive practice of counterprogramming,'" 464 U.S. at 425 n.8 (quoting Universal City Studios v. Sony Corp., 480 F. Supp. 429, 454 (C.D. Cal. 1979)), but Justice Stevens made no reference to necessity in any of his criteria for finding a fair use.

n120 See supra p. 1683.

n121 See, e.g., Rosemont Enters. v. Random House, Inc., 366 F.2d 303, 308 (2d Cir. 1966) (arguing that biographies advance the public interest because "their subject matter is human nature and they reflect the social, economic and political forces of the particular era involved").

n122 See, e.g., Universal City Studios v. Sony Corp., 659 F.2d 963, 969-70 (9th Cir. 1981), rev'd, 464 U.S. 417 (1984); Rubin v. Boston Magazine, 645 F.2d 80, 83 (1st Cir. 1981). The one major exception to this pattern was Williams & Wilkins Co. v. United States, 487 F.2d 1345 (Ct. Cl. 1973), aff'd per curiam by an equally divided Court, 420 U.S. 376 (1975), in which the Court of Claims held that the fair use doctrine shielded the photocopying of copyrighted articles in medical journals. See supra note 62.

n123 See, e.g., L. SELTZER, EXEMPTIONS AND FAIR USE IN COPYRIGHT 23-24, 27-28 (1978); 3 M. NIMMER, NIMMER ON COPYRIGHT § 13.05[A][I] (1987).

n124 464 U.S. at 495 (Blackmun, J., dissenting).

n125 At several points in his opinion, Justice Blackmun implied that productivity should be a prerequisite to qualification as a fair use. See, e.g., id. at 473. However, when it came time to state his view on this point, he softened his stance somewhat, contending merely that, "when a user reproduces an entire work and uses it for its original purpose, with no added benefit to the public, the doctrine of fair use usually does not apply." Id. at 480 (emphasis added).

n126 464 U.S. at 455 n.40.

n127 See id.

n128 471 U.S. at 590-91 (Brennan, J., dissenting). More specifically, relying on the introductory clauses of section 107, Justice Brennan contended that copying for the purpose of "news reporting" should receive favored treatment. See id; supra note 65.

n129 471 U.S. at 561.

n130 Id.

n131 See supra p. 1684.

n132 471 U.S. at 561 (quoting W. PATRY, supra note 6, at 119) (emphasis omitted).

n133 At least two lower courts, presumably attempting to follow the Supreme Court's lead, have recently taken into account the degree to which a copyist's use was "creative." See Financial Information, Inc. v. Moody's Investors Serv., 751 F.2d 501, 509 (2d Cir. 1984); Pacific & Southern Co. v. Duncan, 744 F.2d 1490, 1496 (11th Cir. 1984). Both cases were decided after Sony but before Harper & Row.

n134 See Harper & Row, 471 U.S. at 588 (Brennan, J., dissenting) (insisting that the four factors set forth in section 107 "do not mechanistically resolve fair use issues").

n135 See supra pp. 1673-74, 1675-78 (discussing the fuzziness of the "commercial/noncommercial" and "amount of copying" factors); cf. 3 M. NIMMER, NIMMER ON COPYRIGHT § 13.05[A][5] (1987) ("Because the protean factors enumerated in Section 107, standing by themselves, lack the concreteness to provide definite answers to difficult cases, another test must be invoked . . . to determine whether a given use . . . constitutes a fair use. . . .").

n136 The Supreme Court provided the lower courts virtually no guidance regarding how such a balancing process might proceed. In both Sony and Harper & Row, the Court contended that all of the relevant factors pointed toward a single result. How a court should handle a less clear-cut situation was never considered.

n137 See supra note 44. At least one lower court has accepted the Supreme Court's invitation to develop new factors. See DC Comics, Inc. v. Unlimited Monkey Business, 598 F. Supp. 110, 119 (N.D. Ga. 1984).

n138 Fuller, Positivism and Fidelity to Law -- A Reply to Professor Hart, 71 HARV. L. REV. 630, 665 (1958). On the likelihood that a puzzled judge would look for guidance to the purposes of doctrine, see R. UNGER, cited above in note 105, at 94-99.

n139 But cf. R. DWORKIN, TAKING RIGHTS SERIOUSLY 118 (1977) (arguing that a judge, when relying upon prior judicial decisions for guidance in resolving a "hard case," should "assign . . . only an initial or prima facie place in his scheme of justification" to the principles that the authors of those opinions asserted as the bases of their rulings).

n140 See 464 U.S. at 431-32.

n141 See id. at 450-56; supra p. 1669.

n142 See supra p. 1662.

n143 See, e.g., Mazer v. Stein, 347 U.S. 201, 219 (1954); Fox Film Corp. v. Doyal, 286 U.S. 123, 127-28 (1932); Kendall v. Winsor, 62 U.S. (21 How.) 322, 327-28 (1858); Grant v. Raymond, 31 U.S. (6 Pet.) 218, 241-42 (1832) (patent statute).

n144 See, e.g., B. KAPLAN, AN UNHURRIED VIEW OF COPYRIGHT 78 (1966); Breyer, The Uneasy Case for Copyright: A Study of Copyright in Books, Photocopies, and Computer Programs, 84 HARV. L. RE. 281, 291 (1970).

n145 See, e.g., Hustler Magazine v. Moral Majority, 796 F.2d 1148, 1151 (9th Cir. 1986); Consumers Union of United States v. General Signal Corp., 724 F.2d 1044, 1048 (2d Cir. 1983); Iowa State Univ. Research Found. v. American Broadcasting Cos., 621 F.2d 57, 60 (2d Cir. 1980); Meeropol v. Nizer, 560 F.2d 1061, 1068 (2d Cir. 1977); Rosemont Enters. v. Random House, Inc., 366 F.2d 303, 307 (2d Cir. 1966).

n146 Justice Blackmun, dissenting in Sony, contended that the majority was overly interested in fostering dissemination of original works and insufficiently concerned with sustaining the incentives for creating them. Unlike Justice Stevens, see 464 U.S. at 454, Justice Blackmun saw no "public benefit" in "increase[d] public access to" television programs, id. at 480 (Blackmun, J., dissenting), and fretted that the Court's application of the fair use doctrine "risks eroding the very basis of copyright law, by depriving authors of control over their works and consequently of their incentives to create," id. at 481. However, Justice Blackmun agreed with Justice Stevens that the doctrine was centrally concerned with balancing those two goals. See id. at 477. In Harper & Row, Justice O'Connor quoted the passages from Sony explicating the instrumental theory, see 471 U.S. at 546, and several of the arguments she advanced in the course of her analysis of The Nation's conduct manifest her desire both to maintain incentives for creativity and to facilitate the dissemination of creations, see, e.g., id. at 557 (arguing that liberalization of the fair use doctrine when the memoirs of public figures are copied would sharply reduce the "incentive to create or profit in financing such memoirs, and the public would be denied an important source of significant historical information"). Justice Brennan, like Justice Stevens, was less concerned with providing incentives for creative effort than with fostering the dissemination of intellectual work, but he agreed with Justice O'Connor that balancing those two goals is the principal purpose of copyright law. See id. at 589-90 (Brennan, J., dissenting).

n147 471 U.S. at 546.

n148 Sayre v. Moore, quoted in Cary v. Longman, 1 East 358, 361 n.b, 102 Eng. Rep. 138, 140 n.b (1801). Early American commentary on copyright law also frequently invoked labor-desert theories. See, e.g., Report of Henry Clay, submitted with S. 223, 24th Cong., 2d Sess. 148 (1837), reprinted in Tyerman, The Economic Rationale for Copyright Protection for Published Books: A Reply to Professor Breyer, 18 UCLA L. REV. 1100, 1100 n.3 (1971) ("That authors and inventors have, according to the practice among civilized nations, a property in the respective productions of their genius is incontestable; and that this property should be protected as effectually as any other property is, by law, follows as a legitimate consequence.").

n149 Twentieth Century Music Corp. v. Aiken, 422 U.S. 151, 156 (1975). Justice O'Connor cited Justice Stewart's opinion to support her allusion to an author's right to a "fair return" for his labor. See Harper & Row, 471 U.S. at 546.

n150 See, e.g., Wainwright Sec. v. Wall Street Transcript Corp., 558 F.2d 91, 96 (2d Cir. 1977) (criticizing the defendant for appropriating the plaintiff's financial analyses, "which represent a substantial investment of time, money and labor"); Triangle Publications v. Knight-Ridder Newspapers, 626 F.2d 1171, 1174 (5th Cir. 1980) (describing the fair use doctrine as a "'"rule of reason" fashioned by Judges to balance the author's right to compensation . . . against the public's interest in the widest possible dissemination of ideas and information'" (quoting Sobel, Copyright and the First Amendment: A Gathering Storm?, 19 COPYRIGHT L. SYMP. (ASCAP) 43, 51 (1971))).

n151 Of the four opinions in Sony and Harper & Row, Justice O'Connor's most clearly invokes the labor-desert theory. Justices Stevens and Brennan, in their respective opinions, purport to reject it outright; both quote approvingly a 1909 report of the House Judiciary Committee to the effect that "[t]he enactment of copyright legislation . . . is not based upon any natural right that the author has in his writings, . . . but upon the ground that the welfare of the public will be served and progress of science and useful arts will be promoted" by it. See Sony, 464 U.S. at 429 n.10; Harper & Row, 471 U.S. at 580 (Brennan, J., dissenting). But some phrases and arguments in even their opinions suggest a sensitivity to concerns other than pure instrumentalism. For example, Justice Stevens refers at one point to "the interests of authors and inventors in the control and exploitation of their writings and discoveries," 464 U.S. at 429, and occasional passages in his opinion bespeak resentment of what he sees as the studios' greed in demanding the right to exploit a market developed by someone else -- in other words, the attitude that the plaintiffs were asking for more than they were, in some moral sense, entitled to. In any event, because Justice O'Connor's analysis is the most recent opinion of the Court construing the fair use doctrine, its invocation of the idea of desert is more important for the purposes of this article than the skepticism of Justices Stevens and Brennan.

n152 471 U.S. at 559 (quoting Wooley v. Maynard, 430 U.S. 705, 714 (1977)).

n153 See id. at 550-52; see also id. at 564 (emphasizing "the author's right to control the first public appearance of his expression").

n154 Id. at 555; see also supra p. 1675.

n155 471 U.S. at 564 (quoting Appendix to Petition for Cert. at 300b-300c, Harper & Row, No. 83-1632 (testimony of Victor Navasky)).

n156 For a description of the protection afforded such interests prior to 1976 under state law, see Note, An Author's Artistic Reputation Under the Copyright Act of 1976, 92 HARV. L. REV. 1490, 1496-1500 (1979). For discussion of the more important role this notion plays in most European copyright systems, see DaSilva, Droit Moral and the Amoral Copyright: A Comparison of Artists' Rights in France and the United States, 28 BULL. COPYRIGHT SOC. 1 (1980). The question whether such concerns ought to receive greater attention in the administration of copyright law is considered briefly in note 494 below.

n157 See Note, Artistic Reputation, supra note 156, at 1501-05.

n158 See supra section II.B.5.

n159 Consider, for example, the question of what uses of a work of historical scholarship should be deemed fair. As Justice Brennan conceded in his Harper & Row dissent, an answer designed to optimize the balance between encouraging the writing of history and fostering the dissemination of historical information is likely to have the effect of denying "historians a return commensurate with the full value of their labors." 471 U.S. at 589 (Brennan, J., dissenting). Justice Brennan regarded this effect as acceptable, primarily because "Congress made the affirmative choice that the copyright laws should apply in this way." Id. A judge who, like Justice O'Connor and unlike Justice Brennan, concludes that the fair use doctrine seeks to reconcile social utility and individual desert would find Justice Brennan's concession paralyzing.

n160 See J. BENTHAM, FRAGMENT ON GOVERNMENT 393 (1776); J. BENTHAM, AN INTRODUCTION TO THE PRINCIPLES OF MORALS AND LEGISLATION 11-12 (1789). See generally H. SIDGWICK, METHODS OF ETHICS 411-509 (7th ed. 1907). Section V.A.3 below briefly returns to the teachings of utilitarianism.

n161 See J. LOCKE, THE SECOND TREATISE OF GOVERNMENT, in TWO TREATISES OF GOVERNMENT 303-20 (1789) (P. Laslett ed. 1970). Justice O'Connor's statement of this objective in Harper & Row does embody two substantial modifications of the Lockean labor-desert theory. First, the idea that by devising a unique way of expressing an idea, one deserves to own that form of expression constitutes a substantial extension of the proposition that, by mixing one's labor with a physical object, one acquires a property right in that object. Second, the adjective "fair" in Justice O'Connor's formulation is the precipitate of the gradually acquired insight that, to make normative sense, Locke's theory must be supplemented with an idea of proportionality between the labor exerted and the right to which it gives rise. See L. BECKER, PROPERTY RIGHTS 32-56 (1977); A. RYAN, PROPERTY AND POLITICAL THEORY 32-35 (1984); Brody, Redistribution Without Egalitarianism, 1 SOC. PHIL. & POLICY 71, 74-75 (1983). But the heart of Justice O'Connor's contention remains the Lockean notion that natural entitlements arise out of the activity of labor, and that positive law should respect and enforce those entitlements.

n162 See G. HEGEL, PHILOSOPHY OF RIGHT 33-66 (T. Knox trans. 1942); T. H. GREEN, LECTURES ON THE PRINCIPLES OF POLITICAL OBLIGATION 163-78 (1927). For good secondary studies, see A. RYAN, cited above in note 161, at 118-41; Radin, Property and Personhood, 34 STAN. L. REV. 957, 971-78 (1982); and Stillman, Property, Freedom and Individuality in Hegel's and Marx's Political Thought, 22 NOMOS 130 (1980).

n163 See D. HUME, A TREATISE OF HUMAN NATURE 484-513 (1739-1740) (L.A. Selby-Bigge ed. 1975). Further provocative reflection on these matters can be found in the writings of Justice Holmes. See, e.g., Holmes, The Path of the Law, 10 HARV. L. REV. 457 (1897), reprinted in COLLECTED LEGAL PAPERS 199-200 (1920); cf. M. Horwitz, The Place of Justice Holmes in American Legal Thought 30-33, 38-40, 46-69 (unpublished manuscript) (discussing the development of Holmes' ideas concerning the proper relationship between custom and positive law).

n164 See supra pp. 1671-72.

n165 See supra pp. 1673-74, 1677.

n166 See supra p. 1678.

n167 See supra p. 1681.

n168 See supra pp. 1691-92.

n169 Each of the defects just summarized contributes to this uncertainty, but the most deep-seated of causes is the fragmentation of the normative base of the doctrine. Cf. R. DWORKIN, LAW'S EMPIRE 188-89(1986) (arguing that coherence in a system of rules fosters "efficiency" by enabling persons to predict better courts' decisions); Clark, The Morphogenesis of Subchapter C: An Essay in Statutory Evolution and Reform, 87 YALE L.J. 90 (1977) (showing how tensions between the seven "animating principles" that underlie the law of corporate taxation and the absence of an overarching theory knitting the principles together give rise to uncertainty and frequent litigation).

n170 Olson, Copyright and Fair Use: Implications of Nation Enterprises for Higher Education, 12 J.C. & U.L. 489, 508 (1986). A previous article in the same journal reached similar conclusions. See Cardozo, To Copy or Not to Copy for Teaching and Scholarship: What Shall I Tell My Client?, 4 J.C. & U.L. 59 (1976-77) (emphasizing the confusion and uncertainty generated by the fair use doctrine and by other ambiguous provisions of the Copyright Act).

n171 Note that the Supreme Court explicitly addressed this topic in Harper & Row. See supra section II.B.3. If uncertainty persists on this question, the confusion regarding the applicability of the fair use doctrine to other problems is probably even greater.

n172 See Benedict, Historians and the Continuing Controversy over Fair Use of Unpublished Manuscript Materials, 91 AM. HIST. REV. 859 (1986). An example of the hesitancy with which the author expresses his conclusions: "As of 1986, then, it is very possible, perhaps probable, that the law of fair use of unpublished materials may return to something very close to what it was before Congress passed the Copyright Revision Act of 1976." Id. at 875.

n173 See, e.g., Multiple Copying and the Copyright Laws: A Guide for Harvard Faculty (prepared by the Office of the General Counsel, Harvard University, Dec. 1986). The guide concludes:

In short, if you copy a very brief portion of a non-fiction work, for classroom use, in such a way that the value of the work will not be adversely affected, you have a relatively high degree of protection. If you copy a substantial portion of a work, for profit, harming the work's value, you have little or no protection. Between these extremes, your protection rises or falls depending on the relationship of the four [statutory] factors discussed [above].

Id.

n174 For other manifestations of artists' and users' uncertainty, see Adamo, Problems Connected with Acquisition, Licensing and Enforcement of Intellectual Property, 50 ALB. L. REV. 475, 493 (1986) (advising practitioners that "[t]he concept of fair use" is "something you've got to worry about"); and Blau, High Court Refuses to Review Salinger Book Ruling, N.Y. Times, Oct. 6, 1987, at C17 (recounting practitioners' predictions of "a lot of confusion in publishing houses and among biographers and historians" resulting from the unsettled state of the law concerning if and when a biographer may reproduce portions of an author's copyrighted letters).

n175 A. MACINTYRE, AFTER VIRTUE 9 (2d ed. 1984).

n176 The most insightful account of this condition is A. MACINTYRE, cited above in note 175, at 6-11. See also Cornell, Toward a Modern/Postmodern Reconstruction of Ethics, 133 U. PA. L. REV. 291, 292-93, 300-02, 307-08 (1985). For a provocative piece of legal scholarship -- partly descriptive, partly symptomatic of the disease -- see Left, Unspeakable Ethics, Unnatural Law, 1979 DUKE L.J. 1229.

n177 This aspiration is important to the argument of Part V of the Article. For theorists who share it, see note 366 below.

n178 See A. MACINTYRE, supra note 175, at 8. This is not to suggest, of course, that a well-built fair use doctrine would solve our quandary, but every incoherent field of law represents both a part of the problem and a neglected opportunity to begin to solve it.