It is hard, nowadays, to find a piece of futurology that doesn't say we are entering an information age. Yet there have been few systematic critical studies of the rhetorical, ideological, political, and, ultimately, legal structures through which we deal with information.⁽²⁾ At the same time, the area of legal doctrine that acknowledges that its main concern is information-conventionally defined -- intellectual property-- is marked by severe disagreements over the most fundamental propositions. For example, within copyright there are continuing disputes over a wide range of basic issues-the classification of something as idea or expression, the definition of "fact works," and the dilemma of how to treat computer programs being some of the more obvious.⁽³⁾ Areas outside intellectual property, although superficially commonsensical, are also beset by basic conceptual puzzles. Blackmail and insider trading both represent proscribed forms of trading in information that, when defined in abstract terms, seem to conform exactly to moral and legal norms accepted elsewhere in society. In the former case, why should private individuals be forbidden to make private agreements regarding the dissemination of privately held information? Insider trading is even more puzzling. Markets are built on structured inequality. Why should we not let people trade from a position of advantage in information, just as we let them trade from a position of advantage in wealth? Finally, if we look at the legal treatment of genetic information-which, with electronic information, may define the international economy of the immediate future⁽⁴⁾--we find a morass of formalism, circular arguments, and incoherent theories of property.

Taking these puzzles as its raw material, this Article attempts to develop a theory of law and information. Its aim is not to produce a formula for resolving all problems, but rather to show the themes repeated in, and the possibilities suppressed by, our current treatment of information issues. In particular, the Article argues that questions of information regulation, commodification, and access are shaped by the intersection of two neglected processes of interpretive construction. First, such questions are often decided by an uncritical process of pigeonholing into a number of stereotypes of "public" or "private" information. These stereotypes have their roots in relatively basic assumptions about property, society, and privacy in a liberal state. Second, the tensions among and between these stereotypes are often lessened by an appeal to a particular (and particularly romantic) image of information production-the author who produces originality ex nihilo and thus should be rewarded with honor and property rights. The Article argues that the contradictions, assumptions, and lacunae of these two processes are an important part of any study of the current law of information, or the future social arrangements of the information society.

A note on method may be in order. The first half of this Article-with its concentration on the doctrinal and argumentative patterns generated by the apparent contradiction of "information property"-bears a certain resemblance to the structuralist legal theories of the early 1980s. Like the authors of those articles, I claim that understanding a particular set of tensions allows us to see patterns in otherwise diverse doctrinal developments, to relate legal issues to basic conflicts of political vision, to anticipate the structure of the rhetoric on both sides of any issue, and to see the moral and political inconsistencies in the way that we deal with information. This may sound suspiciously like overblown claims to have discovered the "fundamental contradiction" or even the driving wheel of history. My goal is rather different.

Structuralist and determinist accounts of social forces generally argue that one area of social life-the means and relations of production, the mediation of the fundamental tension between self and others, or the patterns of childrearing-is the real determinant of history. All else is chimera and superstructure. The claims I make here are much more modest. At most, I am trying to lay out the normative topography, the geography of assumptions within which issues are framed, possibilities foreclosed, and so on. This geography matters because it excludes some options from consideration (excludes them even from being seen, perhaps) or prompts a hasty leap to judgment, or because it is one of the many forces shaping the terrain of political struggle. But the ideas and ideologies I am describing are neither so deeply rooted in the culture that they can never be criticized, nor so determinate that they dictate only one solution: the process is neither a giant conspiracy nor a deterministic and inevitable deep structure of thought.

The second theme of this Article is its critique of romantic authorship. I claim that a series of concepts associated with the romantic notion of the "author" is offered as a device to mediate, defer, or apparently resolve the internal tensions of regulating information. Most of the time, courts accept the author gambit, even if the "author" is plainly recombining elements mined free of charge from the public sphere (or even the private body). My examples of this pattern come from copyright doctrine and the law of property in genetic information. Occasionally, however, for reasons apparently related to the pattern of tensions between public and private spheres, the author gambit is declined, and the putative "authors" are instead told they cannot profit from their use of information. Blackmail and insider trading are the two examples given here: both of these would seem like classic market transactions, were it not for the fact that they concern information and thus conjure up all the contradictions and tensions which beset the treatment of information in liberal social theory. The doctrine of blackmail defends the image of the private sphere by protecting an individual's subjectively defined "private" information from commodification in the marketplace. The prohibition of insider trading protects the image of the public sphere by imposing the ideas of formal equality on market transactions precisely because they concern information, rather than some other form of power or wealth. My thesis is strengthened by the fact that insider trading seems from scholarly reaction to lie particularly close to the line. Indeed, the acknowledged pioneer-critic of insider trading law relied heavily on the rhetoric of authorial, entrepreneurial genius to support decriminalization.⁽⁵⁾ Perhaps insider traders can be authors after all.

This aspect of the Article, and particularly its discussion of the social construction of authorship, is reminiscent of poststructuralist criticism. My conclusion-with its reliance on both economic analysis and literary criticism, and with its attempt to imagine the effects of a transposition of the traditions of nineteenth century liberalism to the social relations of an information society-could fashionably be styled as postmodernist. Thus, for those who are interested, this Article presents a kind of layer cake of the intellectual methods of the last fifteen years. There was a reason behind this methodological eclecticism. My hope is that this piece will be seen as an example of constructive engagement between the main schools of legal scholarship, prompting economic analysts to consider the interpretive construction that grounds their theories, devotees of law and literature to look beyond the text to the structure of incentives through which the text was created, and critical legal scholars to take seriously their professed commitment to decoding the social construction of reality in a way that produces practical suggestions for change. Nevertheless, I have tried as far as possible to make sure that familiarity with these academic movements is not necessary to understand the central arguments presented here.

The structure of the Article replicates that of this introduction. In Part I, I set up the four subjects mentioned in the title of the Article. My aim is to treat them as academic conundrums worthy of solution in their own right and as case studies that help us to lay the groundwork for a theory of law and information.

In Parts II and III, I argue that the doctrinal disputes and confusions in all of these areas can be usefully understood as reflections of a set of basic tensions-in particular the tensions between public and private, speech and property, formal equality and restitutio in integrum. These tensions are especially painful because speech, debate, and the exchange of information are so important to the liberal vision of politics and economics and to the construction of the public and private realms.

In Part IV, I turn to the area of information economics. The uncertainties and tensions in conventional legal and political discourse about information, combined with the obvious analytical importance of concepts such as incentives to production and barriers to exchange, make information issues seem particularly susceptible to economic analysis. Legal scholars now routinely turn to microeconomic concepts, whether of "the public goods problems of intellectual property" or the attributes of an "efficient capital market," to resolve problems in the production, generation, commodification, or use of information. Is this reliance on economic analysis warranted? Drawing on the recent theoretical literature and on the structure of tensions described in this Article, I argue that information economics is paradoxical or at least aporetic.⁽⁶⁾ Consequently, it provides no real surcease from the difficulties of liberal state theory or from the indeterminate topography of "public" and "private."

Sometimes economists treat information as just another product, something that must be commodified if producers are to be given an incentive to produce it. At other times, they treat it as part of the structure of the market: infinite, perfect, and circulating without let or hindrance. I argue that contemporary microeconomics offers only the most intuitive basis for evaluating which "aspect" of information is fundamental to the analysis. Yet it also offers no convincing way of reconciling the competing "aspects" into a uniquely correct ordering of preferences or structure of tradeoffs. Any attempted reconciliation ends in paradox, a fact noted by at least one prominent pair of economists.⁽⁷⁾ In fact, information presents almost the same kind of difficulties for neoclassical economics that rent did for classical economics: it is the problem case in which the basic assumptions of the discipline come into conflict with each other.

Although some economic analyses acknowledge the indeterminacies involved in choosing which aspect of information to highlight, most are silent on the issue. In fact, such analyses are frequently "grounded" on a subtextual, and probably unconscious, appeal to a particular vision of information production- the stereotype of the romantic author which I referred to earlier. Consequently, I argue that although economics frames some of the issues of information regulation well, the perception that it is capable of routinely settling those issues is a false one, a false one with important consequences.

In Part V I argue that, within the legal system, the concept of property traditionally bears the burden of resolving the tensions between public and private, but that it does so in ways that merely restate those tensions. Thus, when we turn to a field that can best be described as property in information, we have a set of contradictions that is literally redoubled.

In Part VI, I test the above hypothesis by turning to the first of my puzzles or case studies, the structure of copyright doctrine. I trace the parallel evolution of the romantic idea of authorship-a notion that would have been alien to Shakespeare-and the development of the concept of intellectual property. My argument is that copyright offers a vitally important mediating strategy, a constellation of rhetorical, conceptual, and thematic ideas that seems to resolve the tensions I have described so far. That constellation is formed by the figure of the romantic author, the theme of "original creation," and the distinction between "idea" and "expression." It apparently provides a basis for partial and limited property rights without raising the specter of property as merely a utilitarian, revocable state grant, it justifies the giving of entitlements according to a labor theory of property without spreading those entitlements too widely, and it resolves the tension between public and private by giving the idea to the public world and the property right in expression to the writer.⁽⁸⁾ By appealing to a particular notion of progress and of creation, the idea of "originality" downplays debts to language, culture, and genre and thus justifies the granting of limited monopolies, even monopolies centered around information.

The next two Parts explore the way in which the commodification of information raises different problems depending on whether the situation is typed as a "private" one or a "public" one. In the case of blackmail, discussed in Part VII, the ability to control information-alone out of all other social resources-seems to be central to our conception of personality. In the case of insider trading, discussed in Part VIII, we see information advantages, unlike other advantages in wealth or power, as subject to the norm of equality. In both cases, I argue that if we look at information holistically rather than within its narrower doctrinal boxes, these two longtime subjects of scholarly dispute can be explained if not actually resolved. In both, a refusal to commodify information is justified in part by an intuitive act of "typing" into realms of publicness or privateness, each with its own ideas about justice and its own theories of entitlement.

The mediating role of the author is also relevant to at least one of these two subjects. Blackmail offers us no convincing author figure on whom to pin a property right. But in insider trading, an area where decriminalization has been seriously suggested and defended, things are rather different. At the end of Part VIII I show that the germinal argument for the decriminalization of insider trading actually relies on an eerily perfect invocation of the rhetoric of the romantic author or, in this case, the romantic entrepreneur. This argument defines the entrepreneur in terms of originality, in terms of the creative destruction of settled arrangements. By this view, insider trading is a limited monopoly in information that rewards the entrepreneur for his or her efforts and thus ensures further entrepreneurial acts. Just as with copyright, society is purportedly compensated by the advances and contributions that the great genius throws off. In the face of such an essential contribution to human progress, what's a little monopoly here and there?

More generally, I claim that examination of insider trading scholarship and case law gives evidence of both of the processes this Article describes: the typing of information issues into "public" and "private" and the attractiveness of the author concept as a device to reconcile, or at least obscure, our conflicting notions about information.

The final puzzle, discussed in Part IX, comes from Moore v. Regents of the University of California,⁽⁹⁾ a case that concerns ownership and control of genetic information. The Moore case presents, in miniature, each of the conflicts referred to in this Article: the clash between a vision of property rights as absolute or relative, between the public and private characterization of information, between the market and "privacy," and between natural rights and the rights of creative labor. My analysis attempts to show the consequences, for "sources" such as Mr. Moore, of an entitlement regime built around the notion of romantic authorship.

In the Conclusion, I make three arguments. First, the Article demonstrates a particular method of "resolving" questions of information regulation by typing the issue into one of a series of contradictory visions of information. Second, despite Michel Foucault, postmodernism, and the death of the novel, the romantic vision of authorship is more important today than it was to Fichte and Krause, Pope and Macaulay. The facility with which the romantic vision appears to resolve the tensions I describe in this Article is one of the keys to its success or failure. Sometimes the romantic vision may be "right." Yet its ahistorical claims of universality and its overblown claims about the incentives necessary to production, to say nothing of its overweening romanticism, make it a dangerous template for the regulation of information. Using a string of examples, ranging from software development to the manufacture of pharmaceuticals, I argue that the romantic vision blinds us to the pragmatic, moral, and distributive claims of both "sources" and audience.

The final Section of the Conclusion is more speculative. Above and beyond my attempt to reorient long-running academic debates over copyright, blackmail, insider trading, and genetic information, this Article has a broader goal. It seems reasonable to assume that we are moving into a world that will increasingly be structured around the collection, manipulation, and use of information. At least at first, that world will be discussed, debated, and regulated using ideas about information that come from a society in which information had a rather different set of relationships to economic power and the rhetoric of entitlement. Consequently, the final Section of the Conclusion reflects on what the "information world" could, and should, look like.

One of our greatest cultural legacies is a collection of intellectual tools that help us to understand both the appeal and the limitations of the rhetoric of entitlement produced by a society like ours. Thus, we have a rich tradition of intellectual commentary on the production and distribution of tangible objects and on the consequences of the division between capital and labor, between ownership and control, between public and private. This commentary has produced conceptual tools such as the labor theory of value, the Pigouvian analysis of externalities, and the critique of the public/private split. Such tools have profound limitations and analytical shortcomings, but they do help us to uncover that which is suppressed or rendered invisible by the overwhelming familiarity of our social arrangements. The final Section of the Article attempts to imagine some of the equivalent tools for the information society.

One final note about method may be in order. My subject in this Article is "information." Elsewhere I have explained my rejection of the essentialist vision of language⁽¹⁰⁾ and the disciplinary ideas associated with it.⁽¹¹⁾ I see this Article as a continuation of that work, rather than as a renunciation of it. If I am not claiming that there is some preexisting, essential component that ties together every "information" issue "properly so-called,"⁽¹²⁾ then what holds the examples used in this study together? My attitude towards definition in this Article has been pragmatic, in the looser sense of that term. With Felix Cohen, I believe that a "definition ... is useful or useless.... [It] is useful if it insures against risks of confusion more serious than any that the definition itself contains."⁽¹³⁾

In this Article, I have focused on the communication and manipulation and trade of knowledge-whether found in RNA, typeface, programmer's code, or blackmailer's photographs-and on the parties who are seen as the rightful owners, controllers, or audiences of that knowledge. This has led me into cognate issues presented by copyright law, even though Pierre Menard: True Author of Don Quixote and The Name of the Rose are not normally seen as "information."⁽¹⁴⁾ Bringing these disparate areas together does have its advantages. Each of these areas involves situations in which tangible control of a single res is not enough. The same compromising photograph in a blackmailer's hands, piece of software, best-selling novel, or genetic program plugged into a vat of gene-spliced E. coli can be used to produce an infinite number of copies. In each, it is the message rather than the medium that is central to the analysis. In each, one is forced to confront the division and disaggregation of the property concept, and so on.

To some, this approach may seem doubtful. Professional economists have concentrated on "innovation" rather than "information"-particularly in coverage of intellectual property, but occasionally even in their coverage of such issues as insider trading. I chose quite deliberately to move away from this literature and to use a broad and open definition of information as my organizing concept. In part, this choice stems from the belief that there are interesting things to be said about both levels of generality, and that information was the conceptual box into which my particular theoretical interests fit most neatly. My choice also stems from the fact that the studies that concentrate on "innovation" are destined to repeat the paradigm of the original transformative genius, rather than subjecting it to critical assessment in each of the new contexts in which it is deployed.⁽¹⁵⁾ By themselves, of course, these explanatory remarks could not justify my choices. In the end, it will be the work created from the materials thus assembled that either justifies or discredits the criteria of inclusion. For epistemological reasons, if for no others, the proof of this pudding is in the eating.

Nothing is more familiar to the student of intellectual property than the claim that it is an area of unusual conceptual difficulties. We even have judicial authority for the proposition that copyright approaches "nearer than any other class of cases belonging to forensic discussions, to what may be called the metaphysics of the law, where the distinctions are, or at least may be, very subtle and refined, and, sometimes, almost evanescent."⁽¹⁶⁾ At first sight the current state of copyright doctrine would appear to justify that conclusion. The newcomer to the field is quickly dazzled by the apparently basic questions that surface even in relatively mundane cases. What is the distinction between idea and expression?⁽¹⁷⁾ Are the page numbers in the West Law Reports or the alphabetical compilations of names in a telephone directory actually copyrightable?⁽¹⁸⁾ What are the criteria for deciding such cases: The originality of the work? The amount of labor that has gone into it?⁽¹⁹⁾ The potential loss to the original compiler, or potential profit to the copying party, or to society?⁽²⁰⁾ Can anyone own "facts"?⁽²¹⁾ Does a computer program such as "Windows" infringe the copyright of the Apple operating system if it has a similar "look" or "feel," regardless of whether that look or feel is produced by lines of computer code that in no way resemble the original work?⁽²²⁾ What is the extent of the "fair use" exception to copyright?⁽²³⁾ How far does the "public domain" extend and how are we to conceive of it?⁽²⁴⁾ The list of questions could be extended almost infinitely. Indeed, in a comment on the state of "fact works" doctrine that could well be applied to the whole field of copyright, one scholar put the point succinctly: "[T]here is ample precedent deciding almost every copyright issue in almost every conceivable direction."⁽²⁵⁾ Is there something particular about intellectual property that explains this apparent doctrinal chaos?

Why is blackmail illegal? The crime is defined as an "[u]nlawful demand of money or property under threat to do bodily harm, to injure property, to accuse of crime, or to expose disgraceful defects."⁽²⁶⁾ It is easy to understand that the demand of money is unlawful when the accompanying threats are of bodily harm of property damage. After all, these threats themselves are illegal on other grounds. It is also possible to explain why you cannot demand money as the price of refraining from accusing someone of a crime. But what about the case where a private individual asks another private individual for money as the price of not revealing legally obtained information about activities perfectly legal in themselves: "If you do not pay me $100, I will reveal to your boyfriend the fact that I saw you coming out of another man's house at two o'clock in the morning." After all, in this case it would be perfectly legal to carry through with the threat. Clearly the person being blackmailed does not wish to pay, but then many of us do not wish to pay when others ask money from us, telling us that if we do not comply with their demands they will carry out some legal and unpleasant course of action. How is blackmail different from a baseball team "demanding" concessions from a city and local residents (in the form of tax reductions and parking spaces, for example) under threat of moving the team to another city? It begs the question to say that the baseball team has a right to do so, whereas the blackmailer does not. That, after all, is exactly the point we are supposed to be explaining.

Scholars have been drawn to blackmail like wasps to a picnic. Landes and Posner, Goodhart, Nozick, Coase, and Epstein have all suggested explanations⁽²⁷⁾-none terribly convincing. The commodification of this kind of information is generally reviled and legally prohibited, yet no one has explained why.⁽²⁸⁾ Is there a reason?

Securities law sometimes prohibits certain individuals from trading in securities on the basis of material nonpublic information. All scholars seem to agree that, despite widespread popular support for sanctions against insider trading, the reasons for such sanctions are hard to identify. In fact, a recent article supporting the prohibition of insider trading started with the following startling admission: "American jurisprudence abhors insider trading with a fervor reserved for those who scoff at motherhood, apple pie, and baseball. The commonly stated reasons for this reaction to insider trading are many and unpersuasive. The case law barely suggests why insider trading is harmful."⁽²⁹⁾ Needless to say, those arguing against the criminalization of insider trading are even less charitable towards the reasons offered. In the ubiquitously cited work on the subject, Henry Manne argued that prior to 1910 no one had ever publicly questioned the morality of insider trading.⁽³⁰⁾ Further, he claimed that since that time no one has explained why insider trading is morally or economically wrong.⁽³¹⁾

At first blush, it seems that Professor Manne has a point. Our society distributes wealth through a market system built on inequality of economic power and normally exalts an individual who is able to convert some temporary advantage in knowledge or economic power into a position of market advantage. Why not here? And why is it that, just as with blackmail, so many people share the sense that insider trading is wrong but find it hard to explain the reason?

So far, I have given three brief and general descriptions of areas of doctrine. My last example is longer and comes from a single case,⁽³²⁾ the rhetoric and reasoning of which is so extraordinarily revealing that it deserves extended consideration.⁽³³⁾ In 1976 John Moore started treatment for hairy-cell leukemia at the UCLA Medical Center. His doctors quickly became aware that some of his blood products and components were potentially of great commercial value. They performed many tests without ever telling him of their commercial interest, and took samples of every conceivable bodily fluid, including sperm, blood, and bone marrow aspirate. Eventually, they also removed Mr. Moore's spleen, a procedure for which there was an independent medical reason, but only after having first made arrangements to have sections of the spleen taken to a research unit. In 1981 a cell line established from Mr. Moore's T-lymphocytes was patented by the University of California, with Mr. Moore's doctors listed as the inventors. At no time during this process was Mr. Moore told anything about the commercial exploitation of his genetic material. The likely commercial value of the cell line is impossible to predict exactly, but there were "reports in biotechnology industry periodicals predicting a potential market of approximately $3.01 Billion Dollars by the year 1990 for a whole range of [such lymphokines] ...."⁽³⁴⁾ There were no estimates in the case for the markets after 1990.

This case hinges on issues of information-on at least two levels. On the most obvious level, Mr. Moore was not told about his doctors' financial interest in exploiting his genetic material, an interest that might well have conflicted with the demands of responsible medical care. Dealing with this issue, the Supreme Court of California had no difficulty in ruling that Mr. Moore had stated a cause of action for breach of fiduciary duty or lack of informed consent. After all, he had been denied information in which he had a legitimate interest and which the doctors had a corresponding duty to provide.

On a slightly more abstract level, this case concerned the ownership and control of another kind of information, genetic information. T-lymphocytes are white blood cells that have, coded into their genetic material, "blueprints" or "programs" for the production of lymphokines, proteins that regulate the immune system. If these genetic "programs" from the T-lymphocytes can be isolated, they can then be used to manufacture large amounts of the valuable lymphokine through a variety of recombinant DNA processes. For example, whole vats full of bacteria can be "told" to manufacture the particular lymphokine just as a computer word processing file can issue the same commands to any compatible printer. The key issue in the case was whether Mr. Moore owned the genetic information coded into his cells, or indeed, whether he owned the cells from which that information had been extracted. The court held that he had no property rights in either.

A fascinating array of reasons is offered for this decision. First, the court appears to believe that Mr. Moore had "abandoned" his cells when he consented to their removal. This argument is hard to square with the rest of the decision, however, where-while ruling on the issue of whether he had stated a cause of action in tort-the court did everything but hold as a matter of law that Mr. Moore had not been given sufficient information to consent to the removal.⁽³⁵⁾ Second, the court argued that the cells were not property anyway because California's genetic material statute "[b]y restricting how excised cells may be used and requiring their eventual destruction ... eliminates so many of the rights ordinarily attached to property that one cannot simply assume that what is left amounts to 'property' or 'ownership' for purposes of conversion law."⁽³⁶⁾ By implication, one cannot assume that property rights exist in exotic and highly regulated substances, such as plutonium, which are subject to exactly the same types of regulation. In fact, since almost every kind of property is regulated, what can the court mean?

The court also said that Mr. Moore could not sue under conversion for violation of rights to his cells based on the rights of publicity or privacy. These, the court suggests, are not really matters of property law, so Mr. Moore has no remedy in conversion. Thus, though Johnny Carson has an enforceable interest in the phrase "Here's Johnny"⁽³⁷⁾ (a phrase uttered by someone else), Mr. Moore does not have one in his own DNA. In any event, the court argues that since everyone's genetic material contains information for the manufacture of lymphokines, the particular genetic material is "no more unique to Moore than the number of vertebrae in the spine or the chemical formula of hemoglobin."⁽³⁸⁾

Finally, in perhaps its most interesting twist, the opinion concluded that Mr. Moore could not be given a property right in his genetic material because to do so might hinder research. To back up this argument the court painted a vivid picture of a vigorous, thriving public realm. Communally organized, altruistically motivated, and unhampered by nasty property claims, the world of research is moving dynamically towards new discoveries: "At present, human cell lines are routinely copied and distributed to other researchers for experimental purposes, usually free of charge. This exchange of scientific materials, which still is relatively free and efficient, will surely be compromised if each cell sample becomes the potential subject matter of a lawsuit."⁽³⁹⁾ This argument is a convincing one: property rights in the world of research would only slow down discovery. Convincing, that is, until one reads in the very next column that "the theory of liability that Moore urges us to endorse threatens to destroy the economic incentive to conduct important medical research."⁽⁴⁰⁾ Property rights given to those whose bodies can be mined for valuable genetic information will hamstring research because property is inimical to the free exchange of information. On the other hand, property rights must be given to those who do the mining, because property is an essential incentive to research. Do these assertions contradict one another? Do they tell us anything about the doctrinal chaos of copyright or the anomalies of blackmail and insider trading? Is there a reason that the court is willing to give Mr. Moore an entitlement to "decisional," but not to genetic, information? Finally, does the decision give us any logical or ideological hints about the future legal regime covering biotechnology? I would say that the answer to each question is yes.

I have presented four puzzles. My claim is that each one is best understood as a conflict over the use of information and that the conflict is structured by a recurring pattern of contradictions. It is to that pattern I now turn.

[T]he state as a state abolishes private property (i.e. man decrees by political means the abolition of private property) when it abolishes the property qualification for electors and representatives, as has been done in many of the North American States.... The property qualification is the last political form in which private property is recognized.

But the political suppression of private property not only does not abolish private property; it actually presupposes its existence. The state abolishes, after its fashion, the distinctions established by birth, social rank, education, occupation, when it decrees that birth, social rank, education, occupation are non-political distinctions; when it proclaims, without regard to these distinctions, that every member of society is an equal partner in popular sovereignty .... But the state, none the less, allows private property, education, occupation ... to manifest their particular nature. Far from abolishing these effective differences, it only exists so far as they are presupposed; it is conscious of being a political state and it manifests its universality only in opposition to these elements.⁽⁴¹⁾

The real dilemma of liberal theory is that it must exalt the virtues of egalitarianism, of each person's voice counting equally and, at the same time, confine that egalitarianism to the public sphere. Our vision of society must be a vision of two separate spheres, with two different governing principles, two different theories of justice, and even two different personae to go with them.

Where the political state has attained to its full development, man leads, not only in thought, in consciousness, but in reality, in life, a double existence-celestial and terrestrial. He lives in the political community, where he regards himself as a communal being, and in civil society where he acts simply as a private individual, treats other men as means ... and becomes the plaything of alien powers.⁽⁴³⁾

The law is implicated in every stage of this process.⁽⁴⁴⁾ First of all, the law draws, and in a more complex way depends on, the line between public and private. The central fear of the liberal political vision is that unrestrained state power will invade the private sphere. And yet the only force available to police the state is the state. The rule of law appears to be the answer to this dilemma. By policing the lines between public and private and between citizens and other citizens, the law offers us the hope of a world that is neither the totalitarian state nor the state of nature. In this sense, both the role of law and the rule of law depend on the public/private division.

On a more mundane level, both lawyers and citizens perceive issues through the lens of the public/private distinction. Controversial political and moral issues often resolve themselves as being questions of placement in either the public or private realms.⁽⁴⁵⁾ Access to medical professionals, for example, is in the private sphere. My access depends on my resources; there is no constitutional guarantee to equal health care, or even to minimal health care. Access to legal professionals, however, is at least partly in the public sphere. When I am tried for a crime that may carry a substantial jail term, I have the right to an attorney whether or not my private resources will let me pay for one.⁽⁴⁶⁾

This example suggests one last important point about the public/private distinction in law: our conception of justice differs depending on whether we are dealing with public law or private law. A driver negligently knocks over a pedestrian. If the victim is poor, homeless, and out of work, the law is likely to put him back in exactly that position. Tort damages, after all, are compensatory. If the victim is a $200,000 a year investment banker, the injurer is likely to find himself paying out a lot more in lost wages, among other things. Yet when we turn from private law to public law, the picture changes completely. Should the law punish an assault against an investment banker more seriously than an offense against a homeless person? Our sensibilities are outraged at the thought (even if we suspect this frequently may be the reality). In the private sphere our ideal of justice is, broadly speaking, restitutio in integrum. In public law we aim for equality.⁽⁴⁷⁾

One of the claims of this Article is that every dispute about property rights in information resolves itself into a dispute about whether the issue "is" in the public or the private realm. This rhetoric of geographic placement suggests that we are engaged in a factual inquiry about the location of a preexisting entity within a well-charted and settled terrain. Nothing could be further from the truth. In fact, the process is one of contentious moral and political decisionmaking about the distribution of wealth, power, and information. The supposedly settled landscape is in fact an ever-changing scene that folds back onto itself like a Mobius strip. The market, for example, is on the public side of the map when we are talking about commercial exploitation of private information about families. As citizens we feel the need to keep the impersonal, public world of commerce out of our private realm of home and hearth. Yet the market is simultaneously on the private side of the line as against the state. When the FDA requires a drug company to reveal details of its internal testing practices, the company lobbyists will probably stress the importance of defending private enterprise against public meddling. If a geography metaphor is appropriate at all, the most likely cartographers would be Dali, Magritte, and Escher.

My own views on information issues are strongly influenced by two goals. The first is egalitarian-having to do with the relative powerlessness of the group seeking access to or protection of information. The second is the familiar radical republican goal of creating and reinforcing a vigorous public sphere of democracy and debate.⁽⁴⁸⁾ Yet, for all of the usual postmodern, pragmatist reasons, I would not suggest that we erect these two criteria into a new grand theory. All of the problems of value-choice, contradiction, and indeterminacy of meaning would reappear in the new theory. Thus, I have no criteria that would "replace" the language of public and private, at least no criteria that would algorithmically resolve the questions I put forward here. Why, then, do I spend so much time taking apart that language? My point is that because there is in fact no intelligible geography of public and private, the attempt to resolve issues through a process of line-drawing gives us only an empty exchange of stereotypes, "illusions about which one has forgotten that this is what they are; metaphors which are worn out and without sensuous power; coins which have lost their pictures and now matter only as metal, no longer as coins."⁽⁴⁹⁾ Those illusions, however, have considerable motive power and this Article aims to show where they take us. Even without a grand theory, it may be somewhere we do not want to go.⁽⁵⁰⁾

Now, having introduced the public/private split, let us turn to the second part of this conceptual background, the particular role of information.

Information plays a central, if not defining, role in both the public and the private worlds of the liberal political vision. When we talk about the private world of the family and the home, we define these institutions partly by the idea of an entitlement to withhold information-privacy.⁽⁵¹⁾ The right to withhold information is also one of the main forms of protection given to private citizens facing an accusing state. Fourth and Fifth Amendment protections are the classic cases, but the attorney-client privilege is also a good example.

In the private world of the market, information is again a defining feature. The analytical structure of microeconomics includes "perfect information"- meaning free, complete, instantaneous, and universally available-as one of the defining features of the perfect market. At the same time, the actual market structure of contemporary society depends on information itself being a commodity-costly, partial, and deliberately restricted in its availability. When I discuss information economics,⁽⁵²⁾ this paradox will be of central importance.

Finally, in the public world of politics-which is defined in the liberal vision by the information-centered ideas of debate, exchange, and decision-the free flow of information is a prerequisite for atomistic citizens to form and then communicate their preferences in the great marketplace of ideas. At the same time, a citizen's access to information is thought to be as important a check on governmental activity as that provided by the rule of law. This point was made most famously by James Madison: "A popular Government, without popular information, or the means of acquiring it, is but a Prologue to a Farce or a Tragedy; or, perhaps both. Knowledge will forever govern ignorance; And a people who mean to be their own Governors, must arm themselves with the power which knowledge gives."⁽⁵³⁾

So far I have argued that information, loosely defined, is central to our conception of the family, the market, and the democracy. I claimed that there are tensions "between spheres" in the roles we expect information to play. For example, the public interest in a sphere of vigorous debate and discussion often clashes with the demands of personal privacy. In contrast, claims to own certain information in the market mix uneasily with the values of the First Amendment.⁽⁵⁴⁾ I also have claimed that, within spheres, information is often conceived of in apparently conflicting ways. Looking at the market through the lens of microeconomics, we find that information is both an analytical prerequisite for the model and a commodity to be traded under the model. In First Amendment theory, analysts sometimes talk as if information exchange had its own inevitable tilt towards democratic values and the good life ("the cure for bad speech is more speech"). At other times they present the First Amendment as the jewel in the crown of liberalism, drawing its nobility precisely from the fact that it is value neutral as to content ("I loathe what you say but would die for your right to say it").⁽⁵⁵⁾

Some might ask for explanations: Why do we think such different things about information? This is a vital question, and I am not sure that I can answer it satisfactorily. Part of the answer seems to be that the matrix of conflicts between the theories of justice that we apply to the family, the market, and democracy is overlaid by another set of conflicts caused by the fact that information is conceived of as both finite and infinite, as both product and process. As an infinite good, information seems to be that magical thing-a gift that can be given without making the giver any poorer. I explain Pythagoras' theorem to you, or teach you how to work out the area of a circle. Afterwards, I seem no poorer in the sense that we both have the knowledge. This is the positive side of the public goods dilemma. The same unit of the good apparently satisfies the needs of an infinite number of consumers. Perhaps this is one of the reasons that in moments of high moral or ideological conflict, we often reach for a solution that involves giving the parties more information. If we are thinking of information as a resource that is infinite in this sense, then the distribution of wealth does not seem to have been changed. Yet there are occasions when courts and scholars switch perspectives. Far from being an infinite resource, a good that may be given infinitely without impoverishing the giver, information is reconceived as a finite good, whose production and distribution are subject to the same economic laws as any other. Suddenly it becomes necessary to give producers an incentive to produce more information. Mandatory information transfer is suddenly viewed as an inefficient, forced exchange that undermines the incentives necessary to produce more information.⁽⁵⁶⁾
 
 

So far, I have described information's various roles separately and in a rather static and synchronic way. But the historical importance of the connection between information, the market, and liberal democracy should not be underestimated. Perhaps the most familiar version of the relationship comes originally from the philosophers of the Scottish Enlightenment: commerce was desirable largely because it would force people from widely separated areas to talk to each other, to obtain information about the beliefs and practices of others, and inexorably to question the basis for their own. Thus, the invisible hand would subject social practices and traditions to the test of reason. Doux commerce would be the crucible in which superstition and myth were burnt away and the rationalism of the Englightenment brought to the provinces. This message was extraordinarily influential on the political theorists of the early American republic.⁽⁵⁷⁾ In later years, both in Scotland and in the United States, the message changed. Increasingly, both the Scots and the American republicans began to worry that commerce would produce enormous disparities in wealth and power (including power over information) and that these disparities would be subversive of the republican form of government.⁽⁵⁸⁾ Sadly, this change of emphasis never received the same attention as the original optimistic message.⁽⁵⁹⁾

If the concept of information has potentially conflicting roles to play in family, market, and state, and if information itself is sometimes conceived of as infinite and sometimes as finite, how are social problems involving information decided? Much of the time the answer is, "by drawing lines." We "type" certain situations or conflicts as "public" or "private" and then act as if we have solved the problem. Unfortunately, we have merely restated it. As I pointed out earlier, the notion of "private" is defined in one common understanding largely by the idea of the justified ability to withhold information. Yet the same word, with its connotations of "that-with- which-we-cannot-interfere," conjures up the freedom of the market from state intervention. The fact that we think of the private sphere as encompassing both the market (vis-á-vis the state) and the family (vis-á-vis the market and the state) produces a Laocoon of ideological and rhetorical contradictions.⁽⁶⁰⁾

For example, many consumers do not wish biographical details, provided to a retailer for another purpose, to be traded in the flourishing direct marketing industry. They might argue that this information is "private" and that the state should step in to prevent the companies involved from passing it on, compiling it into larger databases, or whatever. Others might want the state to protect the private sphere of home and family from information coming in from the outside. The telemarketing phone call interrupting the family dinner is the most frequently used example. In both cases, the classification as "private" is supposed to trigger, or at least justify, state protection. Yet the owners of the databases would protest the unfairness of the public world of the state interfering in a private disposition of private property-in this case involving mailing lists or databases of consumer information. The telemarketers might say the same thing, but they would probably also claim that-since information, rather than some other form of property is involved-the issue is one which should be settled by appeal to the constitutional norms that govern the public realm.⁽⁶¹⁾ In other words, they might argue both that the government should not interfere because this is a private activity in the market and that the government should not interfere because this is a public issue of free speech-and equal protection, for that matter.⁽⁶²⁾

If there really were an intelligible geography of public and private, and a unitary concept of information, then we might hold up the hope that one set of claims could be proved to be "true" and the other "false." But since the era of the legal realists, that hope has seemed chimerical.

The story cannot end here, however. One of the themes of this Article is that the implicit frameworks within which the regulation of information is discussed are contradictory-or at least aporetic-and indeterminate in application. As far as the rhetoric of public and private goes, that seems an unexceptionable conclusion.⁽⁶³⁾ Since that rhetoric dominates popular discussion of information, I explored at some length the multiple ways in which liberalism portrays information as central to both public and private realms. It is hard to read public debates on any issue involving information without coming to the conclusion that a great deal of it is an exercise in line-drawing or typing, increasingly isolated from the moral and political ideals the lines are supposed to represent. Perhaps this is the best we can do. But then again, perhaps not.

So much for public debate. Is scholarship any different? Increasingly, scholarly discussions of information are turning away from liberal constitutionalism and rights theory and towards the language of microeconomics.⁽⁶⁴⁾ Whether the issue is copyright,⁽⁶⁵⁾ patent,⁽⁶⁶⁾ insider trading,⁽⁶⁷⁾ blackmail,⁽⁶⁸⁾ or simply "valuable information,"⁽⁶⁹⁾ some of the most ambitious recent scholarship employs some kind of economic approach. Yet, microeconomics provides no surcease from the paradoxes of information. As I will try to show in the next Part, those paradoxes are just as central to the discipline of economics as they are to liberal political theory.

As usual, information has not one role, but many. The analytical structure of microeconomics includes the concept of "perfect" information-meaning free, complete, instantaneous, and universally available-as one of the defining features of the perfect market.⁽⁷⁰⁾ At the same time, the actual market structure of contemporary society depends on information being a commodity- costly, partial, and deliberately restricted in its availability. To put it briefly, perfect information is a defining conceptual element of the analytical structure used to analyze markets driven by the absence of information, in which imperfect information is actually a commodity.⁽⁷¹⁾ If an analogy is needed, imagine a theology that postulates ubiquitous God-given manna-food from heaven-in its vision of the heavenly city, but otherwise assumes that virtue and hard work are both maximized under conditions of scarcity. Now use that theology as the basis for a practical discussion of the ethics of food shortages.⁽⁷²⁾
 
 

This basic theoretical aporia⁽⁷³⁾ explains the weakness of much economic analysis of information regimes. My point is not that the reality is more complicated than the abstraction. That would be a critique of all abstractions, and abstractions are necessary to life. My claim is that information is a problem case for that specific set of abstractions we call economic theory, that it can and must be represented within the theory in two conflicting ways, and that certain concrete problems follow as a result. Some economists believe that these problems can be solved by changing the level of analysis- from perfect to imperfect markets, from imperfect information as commodity to imperfect information as transaction cost to perfect information as component of the analytical structure-just as Russell and Whitehead believed that they could banish paradoxes from mathematics by segregating the component parts of the paradox on different levels of analysis. Godel's Theorem convinced mathematicians of the impossibility of getting rid of this pattern of circularity, recursive definition, and self-swallowing analysis. Sadly, with some notable exceptions that I shall discuss in a moment, economists seem to have avoided any comparable moment of professional modesty.⁽⁷⁴⁾

The first manifestation of this paradox in information economics is the fact that the requirements of "motivation" and those of "efficiency" seem contradictory.⁽⁷⁵⁾ For example, if markets are to be efficient, the prices must perfectly reflect available information. Yet information is costly to obtain. If prices perfectly reflect available information, with no part of the price going to the producer of the information, then there is no incentive to produce more information. To postulate efficiency in the production of information we must assume away the incentive necessary to produce. To postulate the incentive is to make efficiency impossible. It looks like a classic paradox. This is not an observation confined to those skeptical of information economics, some of the most sophisticated economists writing in the area have acknowledged this problem: "There is a fundamental conflict between the efficiency with which markets spread information and the incentives to acquire information."⁽⁷⁶⁾ Are property rights in information a transaction cost that impedes the full and efficient circulation of information? It might seem obvious that they are. After all, perfect information is one of the elements of the perfect market. If information can be commodified, then a host of transaction costs are introduced into information flow and a limited monopoly is granted in the midst of a system supposedly based on competition.⁽⁷⁷⁾
 

Yet the picture changes when information is viewed not as an element within the theoretical structure of the economic system, but rather as a commodity produced and distributed according to the rules of that system. In fact, most economic analysis of information takes this "commodity perspective."⁽⁷⁸⁾ From that perspective, the goal of the analysis is to discover the level of property rights that will produce the optimal level of production. Take the classic case of International News Service v. Associated Press.⁽⁷⁹⁾ Associated Press (AP) operated a news-gathering service. An international network of correspondents and wire services provided news which was printed in the AP papers. Unfortunately for AP, International News Service (INS), which operated a far less expansive news-gathering apparatus, made a practice of free riding on AP's efforts. INS employees would gather news from AP's noticeboards and from early editions of its East Coast newspapers and would then reprint the news, often taking advantage of the time differential between the east and west coasts.

The INS case raises a difficult question for economic analysis. One line of rhetoric and analysis indicates that we should secure to producers the fruits of their labors, and thus induce them to produce more. Without some legally protected interest in the news it gathers, AP will presumably be under a competitive disadvantage. News will become a "public goods problem." Unable to exclude its competitors from the fruits of its efforts, AP will be driven to cut back its news-gathering activities-as will all the other newspapers. Thus, though consumers might be willing to pay for a higher level of news gathering, it will be impractical for any individual newspaper to provide such a service. Put this way, the majority's creation of a legally protected interest in freedom from unfair competition in news gathering is the perfect solution to the public goods problem. By allowing commodification, it ensures continued production and avoids the prisoner's dilemmas set up by the alternative regime.

Yet one cannot solve the problems of economic analysis merely by adopting the commodity perspective-leaving perfect competition and information efficiency concerns aside. The problem of the free flow of information reappears within the new model. For example, should we approach the question of "fair use" in copyright through the lens of the commodity perspective? If we do so, will we only tolerate limitations on intellectual property rights when those limitations are necessary to minimize transaction costs or accomplish well-defined public goals? The most sophisticated scholarly analysis takes this approach.⁽⁸⁰⁾ Consequently, it tries to preserve incentives for creators, even establishing a typology of "fair uses"-assigning or denying property rights in part according to whether those uses would tend to reduce the reward available to the author.⁽⁸¹⁾ Yet the analysis largely ignores the opposite perspective, that of the efficient flow of information. If we switch the perspective, we see that one important purpose of "fair use" law is to make sure that future creators have available to them an adequate supply of raw materials. From this perspective, too many "incentives" could convert the public domain into a fallow landscape of private plots.

To their credit, some economic analysts have attempted to reconcile the two perspectives. Thus, for example, Landes and Posner describe copyright as constructed by the tension between the need to grant legally protected interests to authors in order to motivate them and the need to limit the rights of authors so as to allow future creators legal access to the raw materials they need.⁽⁸²⁾ This seems reasonable enough, but it also leaves them dangerously close to the mushy "balancing" analysis from which economics was supposed to provide surcease. At the same time, the aporia reappears in the question of classification within the theory. For instance, how are we to classify a telephone directory of agents and publishing houses or an index price measure in a futures market in books? Is the former-the information necessary to make the market run-something that should be freely available? Is the latter a commodity in which the creators must be able to claim a legitimate intellectual property right if we are to encourage continued production of information? Or is it exactly the other way around?⁽⁸³⁾

My argument is not merely that analysts are concentrating too much on motivating creators and not enough on the free flow of information. I am claiming that a change in the focus of the analysis does not dispose of these difficulties; it merely reverses their "polarity." There are issues that economists tend to analyze by thinking of information as information rather than as a commodity-for example, the discussion of the efficient capital market hypothesis. Yet, as some economists have pointed out, unless the questions of commodification and incentive are worked into the analysis, the theory ends in paradox as soon as the slightest costs or imperfections are introduced into it. The best example comes from Grossman and Stiglitz's description of the self-destructing futures market.

[W]henever there are differences in beliefs that are not completely arbitraged, there is an incentive to create a market. (Grossman, 1977, analyzed a model of a storable commodity whose spot price did not reveal all information because of the presence of noise. Thus traders were left with differences in beliefs about the future price of the commodity. This led to the opening of a futures market. But then uninformed traders had two prices revealing information to them, implying the elimination of noise.) But, because differences in beliefs are themselves endogenous, arising out of expenditure on information and the informativeness of the price system, the creation of markets eliminates the differences of beliefs which gave rise to them, and thus causes those markets to disappear. . . .

Thus we could argue as soon as the assumptions of the conventional perfect capital markets model are modified to allow even a slight amount of information imperfection and a slight cost of information, the traditional theory becomes untenable....

. . . [B]ecause information is costly, prices cannot perfectly reflect the information which is available, since if it did, those who spent resources to obtain it would receive no compensation.⁽⁸⁴⁾
 

If all of this seems like an Alice-in-Wonderland conclusion, it is worth considering the case of Board of Trade v. Dow Jones & Co.,⁽⁸⁶⁾ in which the court recognized that Dow Jones had a "proprietary interest in its indexes and averages which vests it with the exclusive right to license their use for trading in stock index futures contracts."⁽⁸⁷⁾ The court pointed out, however, that "[t]he extent of defendant's monopoly would be limited, for as defendant points out, there are an infinite number of stock market indexes which could be devised."⁽⁸⁸⁾ The Dow case does not go as far as the "secret stock market" I proposed half-jokingly, but it does exemplify the ineradicable tension between the notion of perfect information and frictionless markets and the notion of commodification and property rights.

This internal tension in the analysis always leaves open the question whether a particular issue is to be classed as a public goods problem for which the remedy is commodification,⁽⁸⁹⁾ or a monopoly of information problem for which the remedy is unfettered competition.⁽⁹⁰⁾ The problem of classification is not merely an empirical one. Even the existence of precise empirical evidence (of a kind currently unavailable for any area of information regulation except, arguably, stock market prices) would not, alone, reveal the right answer unless we had also decided on what level of generality the analysis was to be undertaken. In Feist Publications v. Rural Telephone Service Co.,⁽⁹¹⁾ for example, the Supreme Court denied copyright protection to the compilers of a white pages phone directory. The logic of the analysis I just applied to the INS case might seem to indicate that it was necessary to give the compilers of the directory some protection. After all, directories raise classic public goods problems. The cost of creation is high, yet it is possible at minimal extra cost for additional users to enjoy the same unit of the good thus created. In other words, it is expensive to make and cheap to copy. The Court was not disposed to agree. Partly by means of doctrinal line- drawing (copyright rather than unfair competition) and partly by means of definitional fiat (telephone directories are not original, so in that sense nothing truly new is being created), the Court moved away from the "sweat of the brow" theory and denied the compilers copyright protection.

At first, it appears that the Feist opinion has nothing to tell us about economic analysis. On closer inspection, it becomes apparent that the Court was not so much rejecting the commodity perspective as it was changing the level of generality of the analysis. From Justice O'Connor's perspective, it was the structure of copyright law as a whole that strikes the right balance between the need to reward producers and the need to maintain competition and the free flow of information. Since copyright law as a whole allows the commodification of expression but not the ideas or facts which that expression contains, and since the Court finds this particular arrangement of facts to be "unoriginal," no legally protected interest can be recognized. It is only once this prior decision about the level of generality has been made that the questions of efficiency and incentive can intelligibly be posed.

For all of these reasons, economic analysis of information regimes is extraordinarily indeterminate. A person reading the confident statements of legal scholars about the superior efficiency of the patent regime over the copyright regime,⁽⁹²⁾ or the economic inefficiency of insider trading regulation⁽⁹³⁾ and of the law of fraud,⁽⁹⁴⁾ would be surprised to find that economists cannot agree over the basic question of whether, in the absence of commodification, there will be under- or over-investment in the production of information. Kenneth Arrow takes a position that seems to support the Court's result in INS. He argues that without property rights, too little information will be produced, because producers of information will not be able to capture its true value.⁽⁹⁵⁾ Fama and Laffer, on the other hand, argue that too much information will be generated, because some information will be produced only in order to gain a temporary advantage in trading, thus redistributing wealth but not achieving greater allocative efficiency.⁽⁹⁶⁾ In other words, in the absence of information property rights, there may be inefficient investment of social resources in activities that merely slice the pie up differently, rather than making it bigger. Hirshleifer gives a similar analysis of patent law, ending with the conclusion that patent law may be either a necessary incentive for the production of inventions or an unnecessary legal monopoly in information that overcompensates an inventor who has already had the opportunity to trade on the information implied by his or her discovery.⁽⁹⁷⁾ It is hard to think of a more fundamental uncertainty.⁽⁹⁸⁾

It is my argument in this Article that much contemporary economic analysis conceals these tensions, aporias, and empirically unverifiable assumptions by relying unconsciously on the notion of the romantic author. I have tried to show that most issues in information economics could be portrayed (in the absence of more detailed empirical information) as either public goods problems for which the state has wisely chosen the remedy of commodification in order to avoid underproduction or as potential monopolies in which intolerable transaction costs are introduced into the free flow of information. In later Parts, I will argue that this choice is often concealed by an implicit reliance on the author notion, a reliance that tends to push the analysis towards the incentives/commodity vision of information. This could have serious negative consequences since it will lead analysts (and governments) to support a greater commodification of information than is actually warranted. Such a discourse could also be used cynically to protect existing information monopolies. Economists would be mainly concerned by the possible efficiency losses implicit in such a result. In the Conclusion, I argue that there are also profound distributional issues which should concern us-particularly if we believe that information is becoming one of the primary resources in the international economy.

So much for the theory. What about the facts? The empirical evidence, of which there is surprisingly little, seems to justify these conclusions, or at least to cast doubt on current assumptions about the level of international intellectual property protection necessary to promote research and innovation. A historical, statistical study of the effect of patent protection on the development of drugs in both developed and developing countries from 1950 to 1989 found:

[T]he existence of a patenting system is not a prerequisite for inventions.

. . . .

The relationship between patent systems and their influence on the inventive capacity of developed countries was also tested. Two different tests using Yule's coefficient showed conclusively that, for those countries in which nearly all inventions are made, the relationship is not significant.

. . .The hypothesis that the number of inventions would increase along with the world-wide increase in patent systems was also considered, but it was concluded that there is no significant relationship between these two variables, either in the United States or in the world at large.⁽⁹⁹⁾

Finally, while all such studies warrant methodological skepticism, the studies that support intellectual property protection seem even more problematic. One study estimated that without patent protection sixty-five percent of new drugs produced by the U.S. pharmaceutical industry would not have reached the market.⁽¹⁰²⁾ The analysis was based, however, on data supplied by the pharmaceutical industry in response to a questionnaire on the impact of patent protection on research and development. The problems with such a method are fairly obvious.

In another context, the paradoxes and empirical uncertainties of economic analysis might be of mainly theoretical interest. In discussions of information, they are of immediate practical relevance to almost every issue. In part this is because economists-to their great credit-have been in the forefront of attempts to treat information holistically.⁽¹⁰³⁾ Another reason may lie in the perception that information issues are somehow more "intangible." Escaping more easily from the absolutist, formalist, and physicalist notions of tangible property, information has historically seemed more amenable to a utilitarian calculus.⁽¹⁰⁴⁾ Consequently, these issues are often debated in economic terms-both inside and outside the academy. When the U.S. Trade Representative argued that the General Agreement on Trade and Tariffs (GATT) should be used to pressure other countries to increase their levels of intellectual property protection, she turned to the language of necessary economic incentives, rather than to the labor theory of property or the language of natural rights.⁽¹⁰⁵⁾ Once again, the simplistic claim arises that more protection of intellectual property means more innovation and invention.

There is another reason that economics shapes the debate of information issues. Neoclassical price theory is not only the most sophisticated utilitarian language available, but also the one whose disciplinary assumptions-consumer sovereignty, exogenous preferences, and so on-best reflect a liberal vision of the production, distribution, and exchange of information.⁽¹⁰⁶⁾ The "marketplace of ideas" is more than just a random metaphor: it is an accurate summation of many of the assumptions that our society brings to the discussion of information issues. In a moment, I will argue that this metaphor brings still more problems in its wake.

If microeconomics has become one of the most attractive languages in which to discuss questions of information, then it is almost inevitable that the specific blindnesses⁽¹⁰⁷⁾ of economic analysis will be replicated in information policy. Thus, to the particular difficulties of the economic analysis of information are added the more general difficulties of the economic analysis of law-baseline problems, wealth effects, and so on. These have been analyzed elsewhere, so I will not dwell on them here.

To sum up, there are at least two types of theoretical problems in microeconomic analysis of information. The first stems from the contradictory roles that information plays in the market and in microeconomic theory- information as both perfect and imperfect, property rights in information as both necessary incentives and dubious transaction costs, and so on. The second type of problem stems from the conflict between the assumptions of microeconomic analysis and actual social behavior. For example, when poor schoolchildren are convinced by relentless advertising and peer pressure that they "need" hugely expensive basketball shoes, even a staunch advocate of liberal economics may begin to doubt both the descriptive accuracy and prescriptive fairness of an unswerving application of the norms of "consumer sovereignty" and "exogenous preferences."⁽¹⁰⁸⁾

What conclusions do we draw from the combination of these two theoretical problems? The pessimistic conclusion would be that one of the most influential ways we have to discuss issues of information is a theory so indeterminate that it frequently functions as a Rorschach blot for dominant social beliefs and the prejudices of the analyst. At the same time, this theory tends structurally to undervalue issues of power and inequality.

Call that the pessimistic conclusion. Is there an optimistic one? My answer would be a guarded "maybe." It is a good idea to focus on incentives to production, on transaction costs, and on the problems created by the presence or absence of legally protected interests. It is certainly a good idea to try to discover actual effects of a particular regime of information regulation. By and large, economists have not actually done this, but at least they have talked about it. The tendency of economic analysis to go at least one layer below reified doctrinal concepts is also to be welcomed. One could imagine a type of information economics that was sensitive to baseline errors, offer-asking problems, and wealth effects, that questioned the reality of exogenous preferences, and that openly acknowledged the tension between perfect information and "information as commodity." If this economics also paid more attention than is currently fashionable to the diminishing marginal utility of wealth, I, for one, would be pleased.

If all of these things were done, what epistemological status and practical effect would information economics have? It would be a little less imperial, a lot more modest, and much more empirical. Its conclusions would be more carefully hedged than they are now, and it would openly declare its partiality-the inherent prejudices of any utilitarian, efficiency, or welfare- maximizing calculus and the political consequences of the distinction between allocation and distribution.

To some, this judgment may seem strange in light of my claims that information economics is beset by a basic paradox or aporia. If the discipline is truly paradoxical, surely it is useless?-no matter how chastened its conclusions. The answer, I think, is that economics is only useless if one makes a particular positivist and scientistic set of assumptions about the kind of knowledge a theory has to provide in order to qualify as a theory. Admittedly, both professional economists and economic analysts of law-and not merely those from the Chicago school-sound in their more expansive moments as if they subscribe to those assumptions. But that is no reason for the rest of us to do so. Neoclassical price theory is a valuable tool which enriches our understanding of the world. Like all theoretical system it has blind spots and moments of formal "undecidability." Used with an awareness of its paradoxes and its blind spots, an awareness of the unconscious process of interpretive construction that conceals its indeterminacy, it would nevertheless be a valuable theoretical tool. Seen this way, economics would be a spur to concentrate on incentives and information flow, to worry about perverse motivations and unintended consequences. It would, in short, be more a rough- and-ready set of analytical techniques and reminders than a Newtonian science.

Whether or not this is the economics we should have, it is not the economics we have at the moment. With a few significant exceptions, we have an economics more like my pessimistic picture-an aporetic discipline which, as I hope to show in the rest of this Article, often conceals its indeterminacy through romance. To understand the origins of that romance, we must first look at the liberal conception of property.

Like information, property has a vital role in liberal state theory. That role imposes certain conflicting requirements on the concept of property itself.⁽¹⁰⁹⁾ Legal realism, Lockean political theory, critical legal thought, and law and economics have all stressed-each in its own vocabulary-the idea that property is perhaps the most important way in which we attempt to reconcile our desire for freedom and our desire for security.⁽¹¹⁰⁾ How can we be free and yet secure from the freedom of others, secure and yet free to do that which we want to do?⁽¹¹¹⁾ The most obvious way to deal with this apparent contradiction is to conceive rights of security

in a manner that both makes them appear to be absolute and negates the proposition that they restrict the legitimate freedom of action of others. Thus if we define liberty as free actions that do not affect others at all, and rights as absolute protection from harm, the contradiction vanishes.⁽¹¹²⁾

[A]t the beginning of the nineteenth century, property was ideally defined as absolute dominion over things. Exceptions to this definition suffused property law: instances in which the law declared property to exist even though no "thing" was involved or the owner's dominion over the things was not absolute. Each of these exceptions, however, was explained away. Where no "thing" existed, one was fictionalized. Where dominion was not absolute, limitations could be camouflaged by resorting to fictions, or rationalized as inherent in the nature of the thing or the owner....

As the nineteenth century progressed, increased exceptions to both the physicalist and the absolutist elements of Blackstone's conception of property were incorporated into the law.... This dephysicalization was a development that threatened to place the entire corpus of American law in the category of property. Such conceptual imperialism created severe problems for the courts. First, if every valuable interest constituted property, then practically any act would result in either a trespass on, or a taking of, someone's property, especially if property still was regarded as absolute. Second, once property had swallowed the rest of American law, its meaningfulness as a separate category would disappear. On the other hand, if certain valuable interests were not to be considered property, finding and justifying the criteria for separating property from nonproperty would be difficult.⁽¹¹³⁾

I do not think we need trouble ourselves with the thought that my view depends on differences of degree. The whole law does so as soon as it is civilized. Negligence is all degree-that of the defendant here degree of the nicest sort; and between the variations according to distance that I suppose to exist and the simple universality of the rules in the Twelve Tables or the Leges Barbarorum, there lies the culture of two thousand years.⁽¹¹⁸⁾

Thus, when we turn to intellectual property, an area which throughout its history has been less able to rely on the physicalist and absolutist fictions which girded the traditional conception of property, we will see not only an attempt to clothe a newly invented romantic author in robes of juridical protection, but also to struggle with, mediate, or repress one of the central contradictions in the liberal world view. This, then, is the redoubled contradiction of which I spoke earlier. If it is to protect the legitimacy and intellectual suasion of the liberal world view, intellectual property law (and indeed, all law that deals with information) must accomplish a number of tasks simultaneously. It must provide a conceptual apparatus that appears to mediate the various tensions associated with the role of information in liberal society. Thus, for example, it must explain why a person who recombines information from the public sphere is not merely engaging in the private appropriation of public wealth. It must explain how we can motivate individuals, who are sometimes postulated to be essentially self-serving and sometimes to be noble, idealistic souls, to produce information. If the answer is "by giving them property rights," it must also explain why this will not diminish the common pool or public domain so greatly that a net decrease in the production of information will result. (Think of overfishing.) It must carve out a sphere of privacy and at the same time ensure a vigorous sphere of public debate and ample information about a potentially oppressive state. It must do all of this within a vision of justice that expects formal equality within the public sphere, but respects existing disparities in wealth, status, and power in the private. And all of these things must be accomplished while using a conception of property that avoids the theoretical impossibilities of the physicalist, absolutist conception, but that at the same time is not too obviously relativistic, partial, and utilitarian.

So far, I have argued that because of the contradictions and tensions I have been describing, there are certain structural pressures on the way that a liberal society deals with information. When we turn to the area of law conventionally recognized as dealing with information-intellectual property, and in this case copyright-we will find a pattern, a strategy that attempts to resolve these tensions in the liberal view of information. On one level, understanding this pattern helps to make sense (if not coherence) of the otherwise apparently chaotic world of copyright. On another level, understanding the conceptual strategy developed in copyright illuminates most of the other areas that concern information-even if those areas are not conventionally understood as relating to copyright.

Although intellectual property has long been said to present insuperable conceptual difficulties, it actually presents exactly the same problems as the liberal conception of property generally. It merely does so in a more obvious way and in a way which is given particular spin by our fascination with information. All systems of property are both rights-oriented and utilitarian, rely on antinomial conceptions of public and private, and present insuperable conceptual difficulties when reduced to mere physicalist relations. But when they are conceived of in a more abstract and technically sophisticated way, systems of property immediately begin to dissolve back into the conflicting policies to which they give a temporary and unstable form. In personal or real property, however, one can at least point to a pair of sneakers or a house, say "I own that," and have some sense of confidence that the statement means something. As LeRoy Fibre shows, of course, it is not at all clear that such confidence is justified, but at least physical property presents itself as an apparently coherent feature of social reality. This is a fact of considerable ideological and political significance. In intellectual property, the response might be "What do you mean?"

As Martha Woodmansee discovered, this point was made with startling clarity in the debates over copyright in Germany in the eighteenth century. Encouraged by an enormous reading public, by several apocryphal tales of writers who were household names yet still lived in poverty, and by a new, more romantic vision of authorship, writers began to demand greater economic returns from their labors. One obvious strategy was to lobby for some kind of legal right in the text-the right that we would call copyright. To many participants in the debate the idea was ludicrous. Christian Sigmund Krause, writing in 1783, expressed the point pungently.

"But the ideas, the content! that which actually constitutes a book! which only the author can sell or communicate!"-Once expressed, it is impossible for it to remain the author's property.... It is precisely for the purpose of using the ideas that most people buy books-pepper dealers, fishwives, and the like, and literary pirates excepted.... Over and over again it comes back to the same question: I can read the contents of a book, learn, abridge, expand, teach, and translate it, write about it, laugh over it, find fault with it, deride it, use it poorly or well-in short, do with it whatever I will. But the one thing I should be prohibited from doing is copying or reprinting it? ... A published book is a secret divulged. With what justification would a preacher forbid the printing of his homilies, since he cannot prevent any of his listeners from transcribing his sermons? Would it not be just as ludicrous for a professor to demand that his students refrain from using some new proposition he had taught them as for him to demand the same of book dealers with regard to a new book? No, no, it is too obvious that the concept of intellectual property is useless. My property must be exclusively mine; I must be able to dispose of it and retrieve it unconditionally. Let someone explain to me how that is possible in the present case. Just let someone try taking back the ideas he has originated once they have been communicated so that they are, as before, nowhere to be found. All the money in the world could not make that possible.⁽¹¹⁹⁾

The most obvious answer is that authors are special, but why? And since when? Even the most cursory historical study reveals that our notion of "authorship" is an invented concept of relatively recent provenance. Medieval church writers actively disapproved of the elements of originality and creativity that we now think of as essential components of authorship.

They valued extant old books more highly than any recent elucubrations and they put the work of the scribe and the copyist above that of the authors. The real task of the scholars in their view was not the vain excogitation of novelties but a discovery of great old books, their multiplication and the placing of copies where they would be accessible to future generations of readers.⁽¹²⁰⁾

Woodmansee explains how the decline of the craft/inspiration model of writing and the elevation of the romantic author both presented and seemed to solve the question of property rights in intellectual products.

Eighteenth-century theorists departed from this compound model of writing in two significant ways. They minimized the element of craftsmanship (in some instances they simply discarded it) in favor of the element of inspiration, and they internalized the source of that inspiration. That is, inspiration came to be regarded as emanating not from outside or above, but from within the writer himself. "Inspiration" came to be explicated in terms of original genius, with the consequence that the inspired work was made peculiarly and distinctively the product-and the property-of the writer.⁽¹²⁶⁾

It is the originality of the author, the novel creation the author fashions out of the raw materials provided by culture and the common pool, that "justifies" the property right. At the same time, the postulate of originality offers a strategy for resolving the basic conceptual problem pointed out by Krause: what concept of property would allow the author to retain some property rights in the work but not others? In the German debates, the best answer was provided by the great idealist, Fichte. In a manner that is now familiar to lawyers trained in legal realism and Hohfeldian analysis, but which must have seemed remarkable at the time, Fichte disaggregated the concept of property in books. The buyer gets the physical thing and the ideas contained in it. Precisely because the originality of his spirit was converted into an originality of form the author retains the right to the form in which those ideas were expressed.

[E]ach writer must give his thoughts a certain form, and he can give them no other form than his own because he has no other. But neither can he be willing to hand over this form in making his thoughts public, for no one can appropriate his thoughts without thereby altering their form. This latter thus remains forever his exclusive property.⁽¹³⁰⁾

In the language of romantic authorship, uniqueness is by no means the only characteristic of the author. Originality may imply iconoclasm. The romantic author is going beyond the last accepted style, breaking out of the old forms. This introduces an almost Faustian element into the discussion. The author is the maker and destroyer of worlds, the irrepressible spirit of inventiveness whose restless creativity throws off invention after invention. Intellectual property is merely the token awarded to the author by a grateful society.

A passage from Professor Litman bears repeating at this point:

Why is it that copyright does not protect ideas? Some writers have echoed the justification for failing to protect facts by suggesting that ideas have their origin in the public domain. Others have implied that "mere ideas" may not be worthy of the status of private property. Some authors have suggested that ideas are not protected because of the strictures imposed on copyright by the first amendment. The task of distinguishing ideas from expression in order to explain why private ownership is inappropriate for one but desirable for the other, however, remains elusive.⁽¹³⁵⁾

To sum up, then, the idea/expression division which has so fascinated and puzzled copyright scholars apparently manages, in one stroke:

1. To provide a conceptual basis for partial, limited property rights, without completely collapsing the notion of property into the idea of a temporary, limited, utilitarian grant by the state, revocable at will. The property right still seems to be based on something real-on a distinction that sounds formally realizable, even if, on closer analysis, it turns out to be impossible to maintain.

2. To provide a moral and philosophical justification for giving the author such a property right. After all, through his originality of spirit, the author has created something entirely new out of the raw material of the public domain. The argument is almost like Locke's labor theory: one gains property by mixing one's labor with an object. But where Locke's theory, if applied to a modern economy, might have a disturbingly socialist ring to it, Fichte's theory bases the property right on the originality of every spirit as expressed through words. Every author gets the right-the writer of the roman á clef as well as Goethe-but because of the concentration on originality of expression, the residual property right is only for the workers of the word, not the workers of the world. Even after it is analogized to sculptures and paintings, software and music, it will still have an attractively circumscribed ambit.

3. To resolve (or at least conceal) the tension between public and private. In the double life that Marx described, information is both the lifeblood of the noble, disinterested citizens of the public world and, at the same time, a commodity in the private sphere to which we must attach property rights if we wish our self-interested producers to continue to produce.⁽¹³⁶⁾ By disaggregating the book into "idea" and "expression," we can give the idea (and the facts on which it is based) to the public world and the expression to the writer, thus apparently mediating the contradiction between public good and private greed.

Thus, the combination of the romantic vision of authorship and the distinction between idea and expression appears to provide a conceptual basis and a moral justification for intellectual property, does not threaten to spread dangerous notions of entitlement to other kinds of workers, and mediates the tension between the schizophrenic halves of the liberal world view. Small wonder that it has been a success. Small wonder that, as I hope to show in this Article, the language of romantic, original authorship tends to reappear in discussion of subjects far removed from the ones Fichte had in mind. Like insider trading. Or spleens.

A final question remains before I can proceed. Has the structure I have just described been rendered superfluous by economic analysis and public goods theory? An economist might say that the difference between the author and the laborer is that the author is producing a public good and the laborer is (generally) producing a good that can be satisfactorily commodified and alienated using only the traditional lexicon of property. The distinctions drawn from the idea of romantic authorship might appear to be surplus- unnecessary remnants of a conceptualist age.

It is certainly true that some articles decry the language of "idea" and "expression" and offer the prediction that those terms will be used as mere summations of the underlying economic analysis⁽¹³⁷⁾-in the same way that "proximate cause" is used as a way of expressing a conclusion about the desirable reach of liability. But this kind of response mistakes both the popular and the esoteric power of the language of romantic authorship. As the rest of this Article will show, the romantic vision of authorship continues to influence public debate on issues of information-far beyond the traditional ambit of intellectual property. I tried to show earlier⁽¹³⁸⁾ that the language of economic analysis provides no neat solutions to the problems of information regulation-precisely because economic analysis is marked by the same aporias as the rest of public discourse. In this situation of indeterminacy and contradiction, it is the romantic vision of authorship that frequently structures technical or scholarly economic analysis- providing the vital initial choices that give the analysis its subsequent appearance of determinacy and "common sense" plausibility. Scholars may criticize the distinctions that flow from the romantic vision, but they should not imagine themselves to be free from its influence. This point will be particularly obvious when we get to the unlikely-and distinctly unromantic- subject of insider trading.

In the next Part I turn to the question of blackmail. One of my aims in this Article is to pick examples that illustrate different aspects of the structure of information regulation I described here. Copyright offers the idea of romantic authorship as a way of reconciling the demands of private property and the public realm. By contrast, blackmail is a situation in which the state forbids the commodification of information, precisely because it concerns the private sphere of home, hearth, and personal self-definition.

Blackmail is of academic interest primarily as a proving group. Each new generation of scholars comes to it, as to some muddy and treacherous test track, to try out their new theories.⁽¹³⁹⁾ The test is an apparently simple one: to find out whether their approach will answer the question, "Why is blackmail illegal?"

Before we plunge headlong into that morass, however, it is worth focusing on the qualities that make blackmail problematic in the first place. When scholars talk about the difficulties of explaining blackmail, they are generally referring to a restricted subsection of the doctrine. It is easy to explain attempts to extort money by threats that would be illegal to carry out and to explain why a blackmailer cannot ask money as the price of keeping silent about some violation of the law. The hard case to explain is the situation in which one person asks another person for money as the price of not revealing legally obtained information about activities perfectly legal in themselves. The example I gave earlier was, "If you do not pay me $100, I will reveal to your boyfriend the fact that I saw you coming out of another man's house at two o'clock in the morning."⁽¹⁴⁰⁾ The information was legal to acquire and would be legal to reveal, the conduct was legal to engage in, yet it is illegal to demand money for keeping quiet.⁽¹⁴¹⁾ In Hohfeldian terms, the sale of a privilege has been criminalized but the privilege itself has been retained. How is this different from any other situation in which one economic actor makes a bargain with another to forego a legal course of action that the second party wishes to avoid? To put it another way, what is the qualitative difference between a blackmailer's demands for money and a baseball team's demands for tax breaks, rezoning, and direct grants as the price of not moving to another a city?⁽¹⁴²⁾

The attempts to explain the criminalization of blackmail point in very different directions, and I could not begin to cover the full range of explanations here. Instead, I will use this Part to illustrate three kinds of attempted explanations: economic theories, libertarian theories, and third party theories. I will argue that all three fail to explain blackmail, and that we need a theory that focuses on the various roles that information is expected to play within our society. Admittedly, such a theory gives an answer of a different type than the ones sought by the theorists I cite here. To me, however, that answer seems both more credible and more useful.

Landes and Posner believe that the prohibition of blackmail springs from the state's attempt to prevent (inefficient) private enforcement of the law. In other words, the prohibition of blackmail is supposed to help the state keep its monopoly in law enforcement.

Were blackmail, a form of private enforcement, lawful, the public monopoly of enforcement would be undermined. Overenforcement of the law would result if the blackmailer were able to extract the full fine from the offenders .... Alternatively, the blackmailer might sell his incriminating information to the offender for a price lower than the statutory cost of punishment to the criminal, which would reduce the effective cost of punishment to the criminal below the level set by the legislature.⁽¹⁴³⁾

The second problem with the argument is that it does not explain the hard case I mentioned earlier-the case when both revelation and silence on the part of the blackmailer and the act on the part of the victim would be entirely legal. Thus, for instance, if George Bernard Shaw is secretly eating rack of lamb despite his publicly announced vegetarianism, his butcher may not make him pay for the privilege of silence.⁽¹⁴⁵⁾ At the beginning of their section on blackmail, Landes and Posner note that blackmail appears at first sight to be an efficient way of enforcing the law, "the moral as well as the positive law."⁽¹⁴⁶⁾ Yet (rightly or wrongly) Shaw is not assumed by most people in this society to be violating a moral law-thus the threat of over- or underenforcement does not see to arise.

Perhaps realizing these difficulties, Landes and Posner then offer a slightly different explanation for the criminalizing of the sale of humiliating as well as incriminating information.

The social decision not to regulate a particular activity is a judgment that the expenditure of resources on trying to discover it and punish it would be socially wasted. That judgment is undermined if blackmailers are encouraged to expend substantial resources on attempting to apprehend and punish people engaged in the activity.⁽¹⁴⁷⁾

The second problem with the argument is its leap to the deduction from negative evidence. It is strange to imagine that by failing to criminalize behavior we are making a reasoned judgment that information about that behavior should never be gathered. One can imagine all sorts of reasons we might have for failing to criminalize behavior that would not prevent us from wanting to have news gathered and spread. A politician makes repeated sexist or racist jokes, an air force officer in charge of nuclear weapons is a moody alcoholic, a candidate for a teaching position has awful teaching evaluations at a prior school-in each case, we can acknowledge some social value to the information despite the fact that the behavior concerned is perfectly legal. Posner and Landes would need to show both that: (1) noncriminalization was a conclusive judgment that the information was not worth the investment of social resources and (2) forbidding citizens to trade in it would have a meaningful effect on its acquisition. Yet (2) seems empirically doubtful in the extreme, while (1) seems just plain wrong.

One response to these problems might be that Landes and Posner object only to monetary incentives to engage in this kind of research. Where affection and sentiment or public or private interest supply the motive, we might hope that all is for the best-my examples notwithstanding. Of course, if there are enough nonmonetary incentives to investigate legal behavior, then the prohibition of monetary incentives would be irrational, a use of the state's cumbersome and expensive machinery in the service of a goal that is already unreachable. But even if the empirical question were dealt with adequately (and Posner and Landes do not deal with it at all), such a response brings a new school of problems in its wake. If we concentrate only on occasions involving monetary incentives to gather information, the theory as stated seems unable to distinguish rationally between paid concealment and paid revelation. If adultery is legal in this particular state, does that represent a judgment that we do not want resources devoted to the discovery of adultery? Thinking of the blackmailer, Landes and Posner would presumably say, "Yes." Does this logically require that we prohibit newspaper reporters from following Gary Hart, or one spouse from hiring private detectives to check on the other?

Just at this point in their article, Landes and Posner reverse tack, apparently without realizing it: "We therefore predict that in areas where there is a public monopoly of enforcement, bribery, like blackmail, will be prohibited, while in areas where there is no public monopoly it will be permitted. And so we observe."⁽¹⁴⁸⁾