 
Privacy, News, and Social ChangeIn very early times, the law gave a remedy only for physical interference with life and property. . . Later, there came a recognition of man's spiritual nature, of his feelings and his intellect. . . The law of nuisance was developed. So regard for human emotions soon extended the scope of personal immunity beyond the body of the individual. His reputation, the standing among his fellow-men, was considered. . . Man's family relations became a part of the legal conception of his life. . .From corporeal property arose the incorporeal rights issuing out of it; and then there opened the wide realm of intangible property, in the products and processes of the mind. . . The intense intellectual and emotional life, and the heightening of sensations which came with the advance of civilization, made it clear. . .that only a part of the pain, pleasure, and profit of life lay in physical things. Thoughts, emotions, and sensations demanded legal recognition...Recent inventions and business methods call attention to the next step which must be taken for the protection of the person, and for securing to the individual what Judge Cooley calls the right "to be let alone." ―Louis Brandeis and Samuel Warren 1 "The Right to Privacy" was a product of its time. Yet while using the definitional framework of the late nineteenth century, the article expressed an age-old and still enduring concept of privacy. The continuing impact of the Warren and Brandeis article is testament to the timeless quality of the idea of privacy, an idea reflecting much more than a plea for freedom from tasteless gossip and the salacious prying eye of the press. The ingenious manner in which its authors drew on threads of past jurisprudence, constructing a legal concept of personality out of property doctrine, tort law, copyright law, and damage principles, reinforced the article's timelessness. More importantly, Warren and Brandeis presented the idea of privacy as it should be understood: as deeply entrenched in culture, evolving over time, fundamental to the wholeness of the individual, and reflecting the social environment in which people exist. The right to privacy articulated by Warren and Brandeis rested, at its core, on the need of the individual for a space free from the demands of the larger social order in which to develop beliefs, attitudes, and behavioral norms. But it was not a space free of others. Privacy was, and is, a space occupied by others, but only by some others. The right to privacy fostered disclosure of personal information in the space of intimate associations every bit as much as it protected against disclosure outside that space. Warren and Brandeis made their important contribution by giving legal definition to this boundary between personal and public space— between occasions when personal information should be the business of others and occasions when it should be no one else's affair. That boundary is a reflection of, and indeed is dictated by, social habits and institutions. In 1890, privacy was rooted in rural values, representing an effort to maintain social organizations and values that were threatened by urbanization. Today, however, privacy is rooted in values of individualism. It must function in dramatically changed social circumstances marked by the complex of social arrangements and technologies of our post-industrial society. My purpose is to extend "The Right to Privacy" to the 1990s—to bring it up to date. I will begin in Part I by comparing the social and economic conditions that existed when Warren and Brandeis wrote their article in 1890 to those that mark our era. I will examine the relevance of those conditions to the idea of privacy that Warren and Brandeis expressed. Drawing on an analysis of these social changes, I will suggest that viewing the concept of privacy as a set of social conventions imposed generally on discourse is problematic, if not impossible, today. This is because our society is a deeply individualistic one, premised more forcefully on pluralism and individualism than on assimilation. Social patterns and values today are too diverse, decentralized, and purposefully different to provide a foundation for general rules of discourse at the level of specificity required for the protection of privacy. This conclusion does not imply that a legal concept of privacy is impossible; instead, it suggests that the concept of privacy must be embedded in a context different than external and social norms, one allowing its contours to fit the social and economic conditions of the 1990s. I will advance a concept of privacy based on the individual's control of information rather than on generalized social controls on information, and on an enforceable obligation of confidentiality for those possessing private information rather than on a duty visited on publishers. The thesis is premised on the judgment that the object of privacy has changed over the last century. In 1890, the object of privacy was to protect the operation of a fixed set of social conventions and arrangements by regulating disclosure of information. In 1990, however, the object of privacy should be to give the individual control over the disclosure of confidential personal information in a more complex milieu of personal and social relationships. The theory I propose is more aptly described as a tortious breach of confidence than as an invasion of privacy. Following an exploration of privacy, I will turn in Part II to the subject of news, a principal source of privacy invasions. My purpose will be to explore the idea of news, and relate it to the conditions that existed in 1890 and those that exist in 1990. Changes in news and journalism, chiefly caused by increased fragmentation, lead to conclusions about privacy that are similar to those implied by the earlier discussion of social and cultural change. In view of current and future technological developments in communication, I will argue that regulation of privacy cannot be focused on any single institution or group of institutions that publish information, for control of publication is legally problematic and in any event will soon prove too complex. Instead, privacy should be rooted in control over information, and regulation should focus on the collection of information and access to it, with regulation at the point of publication serving only a secondary protective function. In view of the facts that information may be assembled and combined in a wide variety of ways, that meaning can arise from the very process of selection and presentation of facts, and that information can be used for diverse and idiosyncratic purposes, I conclude that the initial and presumptive authority over maintenance and distribution of private facts must be lodged in the individual, and that others who possess private information should bear a duty of confidentiality with respect to its dissemination. Finally, in Part III, I will turn to the legal system. I will explore very briefly the implications of reorienting privacy from a common law tort whose principal focus is on restricting publication to one premised on a duty of confidentiality, with the individual possessing significant control over occasions in which those who possess personal information are subject to a duty of confidentiality. I will argue that this apparently radical change in the surface of the privacy tort reflects no change in the underlying privacy interest. Without such a change we will alter -- indeed, we have already altered -- the very idea of privacy expressed by Warren and Brandeis, converting it from a concept related to the individual to one related, ineffectively and unwisely, to the terms of public discourse. A tort regulating invasive public disclosures in the interest of appropriate public discourse may be constitutionally doubtful, as recent experience with the privacy tort implies. Moreover, even when the tort is enforceable, it fails to protect any semblance of individual interest in privacy, no matter how that interest might be expressed. My conclusion is likely to raise as many questions as it settles, and my proposal is surely not free of difficult and ambiguous distinctions. But if I am right about the basic idea of privacy, ambiguities in its operative legal terminology can be overcome. An approach focused on the circumstances in which the individual should have control over disclosure of personal information presents no greater definitional problems than those now confronted in the identification of facts that are a priori private, or the disclosure of which is "embarrassing," "offensive," or "newsworthy." And if the approach I suggest succeeds in more closely safeguarding privacy by rules that may actually be enforced, it will fulfill a promise that the present privacy tort consistently disappoints. I. WHAT PRIVACY MEANS Privacy is both a time-bound and culture-bound idea. In the following Part, I will explore the concept of privacy underlying the Warren and Brandeis article, the cultural and social changes since its writing that bear on privacy, and the implications of these changes for what we mean by privacy and how it can be preserved through legal means. A. The Warren and Brandeis Article: Reflections on Legal Methodology and Social Culture in the 1890s "The Right To Privacy" was very much a manifestation of the social conditions of its era. So also was the concept of privacy that Warren and Brandeis developed. They were responding to industrialization, the growth of mass urban areas, and the impersonalization of work and social institutions, including institutions of communication. The 1890s were a time of increasing tension between the inherited values of family and community on the one hand and the larger social order on the other. The concept of privacy was a manifestation of this tension. The right to privacy, according to Warren and Brandeis, was a necessary outgrowth of the "intensity and complexity of life, attendant upon advancing civilization." The need for "solitude and privacy" was essential to the individual and to the culture, but modern enterprise and invention have, through invasions upon . . . privacy, subjected [the individual] to mental pain and distress. . . . Nor is the harm wrought by such invasions confined to the suffering of those who may be made the subjects of journalistic or other enterprise. In this, as in other branches of commerce, the supply creates the demand. For Warren and Brandeis, the institution of the press symbolized the impersonal mass culture that threatened preexisting social institutions that, in turn, enforced cultural values. They believed that the press had overstepped "the obvious bounds of propriety and of decency. Gossip . . . [had] become a trade. . . ." That trade was pursued with "industry" by the press to "satisfy a prurient taste" and to "occupy the indolent" with material that could only be "procured by intrusion upon the domestic circle" for the benefit of a mass audience. The consequence Warren and Brandeis saw was the "lowering of social standards and of morality" and an "inverting [of] the relative importance of things," especially by the "ignorant and thoughtless," and a "usurp[ing of] the place of interest in brains capable of other things." Privacy represented the person's freedom from the undistilled attention of the masses through the media, an attention governed, if at all, by an absence of values. Privacy protected the individual's right to enjoy an identity forged by the existing social institutions of family and community, which embodied chosen social standards and morality. For Warren and Brandeis social standards and morality were to be protected: propriety, gentility, and reason reigned. "Triviality," they said, "destroys at once robustness of thought and delicacy of feeling. No enthusiasm can flourish, no generous impulse can survive under its blighting influence." The right to privacy thus represented an effort to preserve communitarian values and institutions. The right "to be let alone" connoted protection of the community from the masses, the maintenance of a local reference point for personal identity. It did not, as it does today, convey an idea of extreme individualism, of freedom "to" rather than freedom "from." It was not a protection against public embarrassment. It was not even, at the time, a concept principally designed to instill norms of decency in the public press or the public dialogue; that purpose arose later. Rather, privacy reflected the fact that personal identity developed in discrete institutions such as the extended family and the circle of friends and associates that are perhaps best captured in the term "local community." The concept of privacy represented an attempt to protect the functioning of those discrete social institutions from the monolithic, impersonal, and value-free forces of modern society by channeling that which is personal to these discrete institutions and foreclosing it to society at large. B. Social and Cultural Changes, 1890-1990 The tension between local communities and impersonal institutions still exists today. But both the types and number of local communities on which individuals depend for self-definition and the threats from "advancing civilization" have changed. In 1890, the idea of privacy, as expressed by Warren and Brandeis, had a distinctly class-based character. It focused on the "problem" of access by the lower classes of society to information about the upper classes. The concern was with "snooping": the reinforcement of a public appetite for information about other persons, particularly those in privileged positions. In 1990, privacy seems to have less of a class-based, and more of a democratic, character. To a much greater degree, it reflects the individual's general concern about the availability to others of information about oneself. This concern arises, perhaps, from the more mendacious quality of today's public and political discourse, and from a sense that discourse claims "ownership" of its objects as public property. The idea of privacy in the 1990s expresses more clearly the individual's interest in some measure of control over self through control over information. In its broadened applicability -- its more democratic or universal application -- the evolution of privacy is not unlike that of libel, which likewise has grown from essentially elitist roots. In a related vein, privacy in 1890 was focused principally on apprehension about disclosure of personal affairs in the public forum, particularly in the relatively new mass media. In 1890, information was tightly and almost exclusively controlled by a very few large institutions. As the dominant institution, the press was viewed with apprehension by the individual. It threatened loss of identity or, perhaps more accurately, it threatened to shift the source of identity by making identity into a social construct rather than a choice governed by oneself and one's intimate associations. In the 1990s, control of information -- both access to it and the power of its dissemination -- is much more widely dispersed. Information generally still is the province of institutions, but many more institutions possess personal information, and the prospect of even more decentralized control over information looms large. There is little likelihood that the mechanisms that earlier served as gatekeepers on information are an effective limit on disclosure today. The comfort, if that is an apt description, of a large but limited threat has been removed. Indeed, part of the concern about privacy today is the seemingly unlimited potential for disclosure from multiple sources for multiple, and unknown, uses. Disclosure of information in the public press seems hardly the bulk of the privacy problem today, although it is surely the most widely noted. 2 This changing locus of threat is linked to the changing nature of what is threatened. In 1890, the mass media represented a threat from the outside. It was a foreign and uncontrollable force that threatened those inherited and settled institutions such as the family and the geographically localized social community, and therefore threatened the individual. In 1890, the individual's identity was generally secure in a relatively few stable and local relationships. The 1990s, however, are a time when identity through association is more elusive and complex; it does not draw on a small number of stable institutions, but instead depends on many specialized and changing relationships. In the 1990s, disclosure of information is less a threat to the individual's values and identity within a settled social framework than it is an obstacle to existence in the larger framework of general social prejudice. It is a time of abstraction of the individual from institutions, when responsibility for relationships rests with each person and when choices about social relationships and personal associations, precluded by convention in earlier times, now must be made. In 1890, work and home were still closely related. Both were basic to the individual's identity, but the concentration of industrial wealth and the large scale of enterprise were producing an alienation between them. The struggle was to assimilate changing patterns of work with the still dominant culture of the family and the role of local communities in expressing and enforcing shared values. By 1990, however, work and home had become separate; their alienation had been completed. But by 1990 the breakdown of the family and the disintegration or fragmentation of local communities in which individuals exist had also occurred. Work, not family or community, has become the dominant force in shaping the individual's identity. While 1890 was marked by family and community, the 1990s are marked by work and individualism. Concomitantly with these changes, our society has shifted from one possessed of shared moral convictions to one marked by moral pluralism. In the 1890s, discourse, both public and private, was governed by widely shared values and conventions. Propriety and gentility were widely endorsed and enforced norms, even though the seeds of their future anachronism were already evident. Today, in the pluralistic, robust, and even mendacious culture of the 1990s, discourse is less governable. It is a time not of shared values, but of diverse values; not of assimilation, but of pluralism. As a result, the distinction between the "public" and the "private" in the mass media has blurred. Commitment to reason and the obligation to limit publication by some consensus measure of value are largely absent. Even the analogies Warren and Brandeis used to lend support to their creation need reevaluation 100 years later. In 1890, Warren and Brandeis drew on the analogy to private property in one's personality as a logical, culturally relevant, and powerful symbol for privacy. In the 1990s, the private property analogy lacks cultural authority. Property "rights" are highly relative, as is much else in our legal world. And physical "space" itself seems less important than "space" in which to exercise autonomous control over personality. This latter type of space enables individuals to shape and protect psychic identity as a source of values in a more ambiguous and value-pluralistic social order. In the present information-based and technological environment, this kind of space is characterized more by autonomy and individual power and control over unchosen outside forces than by a perceived need to be free from embarrassing disclosures or to be "let alone." What is most important today is not, as in 1890, the crushing and destabilizing attention of the prying eye, a quasi-physical intrusion from the unwashed masses into one's local community to measure one's private acts against some common standard. Instead, the focus today is on the individual's sense of control over events swirling about, on the general sense that one cannot control or influence the mass, decentralized, and diverse "publicness" of society at large, and on a foreboding that things will get worse rather than better in the future. C. The Changing Social Texture of Privacy, 1890-1990 "The Right to Privacy" was written at the end of the nineteenth century; today we are about to enter the twenty-first century. The 1890s were a time of social compression, of urban crowding, of industrialization, mass production, and concentration of economic power; they were a time when rural values and institutions persisted, although somewhat precariously. The 1990s, by contrast, are a time of decompression, of decentralization and variety. It is a time when the older values of place and community have been muted, but we perceive their absence and seek to recapture them. What texture does the concept of privacy possess in these markedly, if not radically, different social settings? In addressing privacy as an instrumental concept in its social context, the distinction between types of privacy norms, as suggested by Professor Schoeman, is fundamental. Social norms protecting privacy are of two, possibly overlapping, sorts. There are social norms that restrict access of others to an individual in a certain domain where the individual is accorded wide discretion as to how to behave in this domain. This is the sort of privacy that promotes private life, individuality, and the integrity of various spheres of life. Then there are social norms that restrict access of others to an individual but where the behavior carried on in private is rigidly defined by social norms and affords little discretion. Though this sort of behavior is performed in private, is relegated to private life, it is not intended to enhance individual choice. . . . The [latter form of] privacy . . . does not serve the purpose of self-expression, and is instead a manifestation of a rigid form of social control. The function of privacy, in other words, is both freedom and control. The social context in which privacy is located defines the way both the freedom and social control functions are achieved. In 1890, social control may have been the dominant function of privacy. Social control could be more easily achieved in a society marked by relatively consistent and established patterns of social organization whose functioning was informed by widely shared values. But when, as today, the decentralized society is highly individualistic rather than communitarian -- when family and discrete social units exert only a weak force on behavior -- the social control exerted by privacy is weakened. Enforcement of behavioral norms and taboos can be accomplished by disclosure of private information within small communities when the family and circle of friends are the effective agents of control. When they are not, when work is the principal locus of individual identity, and when the shared moral values of fixed smaller communities are unavailable, the social control function of privacy is jeopardized. Today, the function of privacy relating to freedom of the individual as an individual is the dominant one; the function of privacy relating to social control is greatly diminished. The freedoms protected by today's more individualistic idea of privacy are of two sorts: freedom from intrusiveness and freedom to achieve identity. These are related and inseparable. In a more decentralized society the individual has greater freedom to structure identity, but at the same time, the threat of intrusion into private life is potentially greater, more diffuse, and less controllable. 1. Privacy and Freedom Paradoxically, while in 1890 the social control function of privacy may have been dominant, Warren and Brandeis were principally concerned with the first dimension of privacy, the loss of freedom resulting from intrusion into behavior that took place within the small social units of family and local community. They focused on the tension that was building between a centralized industrial society and the social arrangements of a former, more rural society. It was dissemination by the press that concerned them. Disclosure of private information by word of mouth, the typical mode of communication in intimate social arrangements, was "probably" not actionable absent "special damage." The source of the disclosures on which they focused, the metropolitan press, was likewise a manifestation of centralization and industrialization, a discordant instrument of mass, indiscriminate, and impersonalized distribution.
Warren and Brandeis were not principally concerned with the second norm of privacy identified by Schoeman -- privacy as an instrument of social control by which norms of behavior and social taboos are enforced. This is not to say that they were unaware of it, but rather that there was simply not much cause for such concern, for the locus of control in social institutions was not in dispute. In 1890, the family unit was still intact, and family, not work, was the primary source of identity. Small and intimate social communities existed and functioned effectively. Indeed, the late 1800s were a time of family, community, and moral value, as evidenced by the political movements of the time, which were largely a reaction to the still nascent impersonalizing tendencies of industrialization. In 1890, Warren and Brandeis were concerned with the more pressing problem of limiting disclosures that ranged beyond sources of social control -- disclosures that threatened freedom without any cost to control. In the 1990s, the problem is very different. While the occasional disclosure of embarrassing personal information through the established media is widely publicized and excites concern, the instances of such disclosures that threaten freedom seem rare. Instead, the problem posed for privacy today largely involves disclosures that threaten control. This is not because there are too many or too few disclosures at the level of family and community; rather, it is because disclosures at that level no longer serve the primary function of enforcing social norms and taboos, and have therefore lost much of their justification. Today the source of individual identity tends to rest most prominently with the individual and with new and varied types of associations chosen by the individual, not with the family or the community. As a result, disclosures at a local and decentralized level are of greater significance to privacy interests. They threaten freedom, but also, and more importantly, they threaten the function of control now lodged in groups selected by the individual rather than imposed by (or generally accepted by) society. In short, disclosures threaten because, in their monolithic quality, they do not reflect today's idiosyncratic pattern of personal associations. Of equal importance, the tendency toward individualized and idiosyncratic associations is occurring during a time of decentralization in communication -- a time when power over information lies in the hands of those communities that were, but no longer are, the principal agents of social control. This suggests two conclusions. First, disclosures on a more intimate level within traditional associational groups may not serve the beneficent social purposes assumed by Warren and Brandeis, for they no longer serve the dominant associational functions of intimacy and control. Second, control of disclosures should be placed in the hands of the individual, not in the family or community, for it is there that associational preferences are now exercised. 3 2. Privacy and Social Control The individual identity fostered by privacy depends not only on freedom, but also on social control. The formation of such an identity requires both intrusiveness and freedom from intrusiveness, both disclosure and nondisclosure. The important and difficult question is distinguishing when disclosure should be foreclosed and when it should be fostered. Even in our highly individualistic society, we continue to be social beings. We depend on other persons and groups as we develop norms of behavior and values. We depend on forces of conformity as much as we depend on our own independent reason and free will. Values and norms of behavior, at least at the level of self-identity, cannot and should not be provided by the state or by the society at large. Instead, they must be provided by the smaller associational and reference groups that I call "family" and "intimate communities." These smaller groups provide voluntary and discrete associational relationships. Government and large-scale institutions exact norms of behavior as well, but they are neither discrete nor, for the most part, voluntary. This is not to say that privacy and conformity are one and the same. To the contrary, privacy protects individual identity, which in turn reflects a balance of independence and conformity. This balance is struck by the individual in choices about intimate communities with which he or she chooses to associate. Disclosure of personal information to mass communities exacts rigid conformity to the indiscriminate norms of that community without increasing the individual's ability to develop normatively at a more discrete and meaningful level of values. On the other hand, disclosure of personal information to intimate communities is a necessary precondition to the individual's development of personal and social identity and personal values within the community. The problem of privacy, then, is not one of disclosure; rather, it is one of fixing the appropriate occasions for disclosure and of identifying the basis on which those occasions are to be selected. The problem of conformity itself, as well as the appropriate reference groups in which conformity functions, is different today than it was in 1890. According to David Riesman, "[T]here has been an enormous ideological shift favoring submission to the group, a shift whose decisiveness is concealed by the persistence of the older ideological patterns." Ironically, the past 100 years have simultaneously witnessed a breakdown in the coherence of the family as a reference group for the development and enforcement of values and norms of behavior and a similar disintegration of coherent, local communities that once served the same purposes in an essentially rural or provincial society. According to Georg Simmel, earlier social formations were characterized by "a relatively small circle firmly closed against neighboring, strange, or in some way antagonistic circles. [The] circle [was] closely coherent and allow[ed] its individual members only a narrow field for the development of unique qualities and free, self-responsible movements." In the 1990s, much individual and social formation occurs through different mechanisms than those that characterized the rural culture of the 1890s. With the breakdown of the family, and with the disintegration of local communities in a post-industrial society, the pervasive influence of the single community has diminished. As Professor Schoeman has put it: Under circumstances such as ours, where the material and cultural climate[s] encounter rapid flux, change mechanisms have to be self-consciously embraced as conducive to social survival. The social evolution of groups becomes too slow. This consciousness affords greater value to individual initiative and outlook as affording the group a means of coping. As individual initiative becomes more important for social viability, the balance of what is legitimately expected of the individual changes, and with it the place on the spectrum that is identified with social overreaching [by the group]. As a way of adapting to the need for change, associations of a more specialized nature will take over many of the roles that global association or community provided in less exciting times. The emerging functionality and multiplication of groups result in less dependence on the part of individuals on any specific group and of any group on any specific individual. So as individuals and groups, one by one, become more independent of one another, there will be less pressure to be part of any given group, and there will be more definition or specificity in the terms of the relationships within any group. Membership in a society will not hinge on belonging to "the" group, but instead on belonging to some selection of groups that have certain family resemblances and interdependencies. Privacy is very much related to, and dependent on, group relationships. But the past 100 years have seen a change from a limited number of relatively stable groups formed to develop a broad range of values and norms of behavior to more fluid, discrete, and specialized associations requiring different types and degrees of disclosure. The role of the individual in selecting group associations has consequently become greater. This change in group relationships has had a direct impact on privacy, both with respect to prohibitions on disclosure of personal information, and, perhaps more importantly, on the circumstances in which disclosure is valuable and therefore should not be prohibited by law. The possible types and conditions of non-constructive disclosure -- those in which disclosing information serves no purpose in shaping individual behavior and values -- have increased. Not only is disclosure in the mass media, which principally concerned Warren and Brandeis, generally non-constructive, but disclosure within more traditional local communities may also often be nonconstructive. This is because these more traditional communities, such as the union hall, the neighborhood, the town, and the workplace, are not the types of associations that people generally rely on in the 1990s. Indeed, disclosure in such settings can be destructive, for even if individuals do not rely on them as sources of values and norms, disclosures of behavior at variance with the expectations of such groups can produce alienation or undesirable conformity. However, disclosure of information to the groups that an individual relies on for values and norms of behavior is critically important to the functioning of the groups, and should not be prohibited by the law. In the 1990s, such associational groups are much more specialized, discrete, and fluid. The type and extent of personal information needed by any particular group will vary with its more discrete function, which itself is defined more by the individual than by the group. Therefore, a general line of demarcation between restricted and permitted disclosures, or between nonconstructive and constructive disclosures —such as the exemption from liability for oral communication proposed by Warren and Brandeis —is not workable. Because of the individual's ability to choose different associational groups, which may themselves change over time, the best, and perhaps the only, alternative is to place the locus of power to define the line separating nonconstructive from constructive disclosures in the hands of the individual and to enforce that power through the obligation of confidence. As Professor Schoeman has stated: [H]istorical circumstances force different relationships between individuals and social groups. Some of these enhance the role of the individual outlook and diversify the terms of individual association. Both of these directions afford more scope for privacy. Groups . . . become less dependent on conformity, and more dependent on individual variation and outlook as sources of change. As these dependencies lessen, what it is fair for the community to expect of the individual lessens or at least narrows. And as the community's role in shaping the individual and promoting survival and transmission of strategies for common life becomes distributed among more restricted functional associations, more relationships will arise from which aspects of people's lives become irrelevant or at least less central. This too, like the emerging dependence on individual variation, fosters privacy for it affords less basis for others having access to various parts of people's lives. We construct our personality in more complex ways today. To preserve our capacity to forge our personality or identity, our "spiritual nature, . . . feelings and . . . intellect," as Warren and Brandeis expressed it, we need a more sensitive legal instrument that functions effectively in a more complex milieu of social arrangements. 4 D. From Invasion of Privacy to Breach of Confidence In a complex society in which local communities are numerous, diverse, and specialized rather than few and monolithic, privacy cannot be protected by generalized distinctions between family and neighbors or between oral and written expression, as Warren and Brandeis proposed. If we are to protect privacy as it is manifested in today's social environment, the legal system's policies must be crafted on the fabric of more idiosyncratic patterns of behavior, on which embellishments of justification, waiver, public interest, and the like can be placed. They must be grounded in the individual's choices, and placed substantially in the individual's hands. My suggestion that we focus privacy on individual control and embed it in the idea of confidentiality is based primarily on the social dimension of privacy -- the aspect of privacy that serves to foster the development of individual personality in intimate associations through excluding disclosures in other settings. But it also reflects the changes that have occurred in the other dimension of privacy -- freedom from exposure to others. This second theme will be developed in the pages that follow, but it is worth observing here that the decentralization of control over information and publication renders incomplete any form of privacy protection directed primarily at large institutions and mass distribution of information. The power to obtain, disseminate, and use information is now highly dispersed. The point of publication can no longer serve as the point of control over disclosures. Effective control over disclosures can be exercised only by focusing legal duties on possessors of information and shaping their responsibilities around the conduct, preferences, and voluntary associational relationships of each individual. In proposing to shift the protection of privacy toward individual control over information, I do not mean that individuals should be given unilateral legal authority to control all disclosure of personal information, even if the types of information considered private are narrowly defined. Rather, I am suggesting that we change the basic premise of the law's approach, that we look at privacy from the bottom up rather than from the top down, from the perspective of individual choice rather than from publisher judgment. We should look to the individual -- his or her conduct, preferences, and voluntary associational relationships -- as the principal benchmark by which we define the boundaries between constructive and nonconstructive disclosures. Given the changing nature of these relationships, I suggest that the concept of confidentiality be explored as a means of limiting disclosures by others when disclosures do not, or no longer, serve associational interests. Focusing the privacy tort on control of disclosure of personal information rather than on its ultimate publication will not remedy the social ills that have resulted from the breakdown of family and local community. These are forces that may not be changeable, but to which we may adapt. Focusing on control by placing a significant measure of control in the hands of the individual, and coupling it with obligations of confidentiality placed on possessors of personal information, will at least give the individual the power in an increasingly specialized and complex culture to define those associations that are personally fruitful. It will give the individual the freedom, in other words, to avoid the condition Georg Simmel described as "becom[ing] a mere cog in an enormous organization of things and powers which tear from his hands all progress, spirituality, and value in order to transform them from their subjective form into the form of the purely objective life." II. PRIVACY AND THE CHANGING "SOCIOLOGY" OF NEWS The privacy tort as currently constructed visits liability on the point of publication, a fact dictated by the legal requirement of "publicity." It is therefore no great surprise that the vast bulk of privacy actions concern news, the medium in which current and general information of interest to the public is communicated, and that much attention has been paid to the relationship between privacy interests and the public's interest in information. But news and the institutions of public communication have undergone fundamental change in the last 100 years. These changes have been so fundamental, in fact, that one can conclude with respect to news, just as with the social fabric of privacy, that the privacy tort formulated by Warren and Brandeis is now largely obsolete.
Today we live in an age of disclosures and "scoops." Congressman Wilbur Mills, Senator Gary Hart, Judge Douglas Ginsburg, and many other prominent people have been brought unceremoniously and effectively into the public spotlight by disclosures of their private conduct. This phenomenon is not restricted to highly visible public figures. Stories enlivened by controversial personal sides of private citizens, whether brought to life by the depiction of personal suffering and plight, or based on people's private revelations about sexual orientation, drugs, and the like, are increasingly commonplace. In the media, the distinction between "hard news" and "features" is obscure, and network news programming is increasingly influenced by, and sometimes dictated by, the entertainment divisions. But beneath the surface of these manifested changes are more profound changes in the structure of communication: its audiences, its speakers, and even its meaning and significance. It is on these changes that I will focus, for they too lie at the heart of the social fabric of privacy. I will begin in the following Section by exploring the subject of news in an attempt to discover what it is that makes personal information appropriate for public consumption. Not surprisingly, this is a question that will quickly draw our attention to surrounding social changes and to related changes in the structure and role of journalism and news. It is a question deeply influenced, as well, by profound technological change. A. News and Its Loss of Meaning Exploring the nature of news is dangerous business. News is a phenomenon sheltered effectively against scrutiny by the mystique of the editorial process and embedded firmly in the disguise of editorial judgment. News judgments are made in confidence and in the quickened pace of publication schedules, based on intuitive judgment and competitive strategy. This process makes elusive, and perhaps impossible, the reduction of the concept of news to a clear and fixed definition. The immediacy of news publication leads to the existence of news on two levels: that which is published, and that which persists. As Professor David Hamilton has expressed it: "What is news is a question related to what is the news that stays news, or, analogously, what is history, what is an historical fact." Or, as Ted Koppel put it: "Consider this paradox: Almost everything that is publicly said these days is recorded. Almost nothing . . . is worth remembering." De Tocqueville put the point differently, and more engagingly, in his observation on literature: In aristocracies, readers are fastidious and few in number; in democracies, they are far more numerous and far less difficult to please. The consequence is, that among aristocratic nations no one can hope to succeed without immense exertions, and that these exertions may bestow a great deal of fame, but can never earn much money; whilst among democratic nations, a writer may flatter himself that he will obtain at a cheap rate a meagre reputation and a large fortune. For this purpose he need not be admired, it is enough that he is liked. News has always been different from that which is important and lasting; the latter may often be part of news, but simply undistilled. And, as often, news may altogether miss that which is important. As Walter Lippmann observed: "[N]ews and truth are not the same thing. . . . The function of news is to signalize an event, the function of truth is to bring to light the hidden facts, to set them into relation with each other, and make a picture of reality on which [people] can act." But this does not mean that notions of permanence, importance, perspective, and, indeed, truth (as Lippmann uses the term) are no part of news, for news includes each of these. News —or the important qualities of news — consists of selecting and sorting that which is publishable from that which is not and of placing facts and information in relation to other facts and information, even without the surer perspective that time alone can provide. The selecting, sorting, and placing is not conducted by whim; it is governed by substantive criteria and constrained by conventional processes. 5 The social tension between news that persists and mere publication may be traced to the simultaneous demise of aristocracy and the emergence of mass media technology. In the United States, the technology of inexpensive publication of information throughout all segments of society came of age in the late 1800s. Concerns over invasions of privacy did not crystallize, however, until Warren and Brandeis wrote about the power and potentially debasing impact of broadly disseminated information. In "The Right to Privacy," Warren and Brandeis were simultaneously reflecting on the demise of aristocracy and urging that the criteria governing news judgment in a democratic society be set high to preserve both "robustness of thought and delicacy of feeling." They were, in short, sensing the loss of aristocratic truth and attempting to recapture it. By 1990, aristocracy was gone, and with it, an aristocratic version of truth that governed news. The threats to privacy associated with news, through the mass and immediate distribution of information, have changed, both in degree and in complexion. They have changed in degree because mass publication and distribution is infinitely more pervasive today, and the technologies of transmission are highly varied. Perhaps more importantly, the problems associated with news have changed in complexion because the sources of information are infinitely more fragmented, as are the audiences. News as mass distribution of information is no longer the dominant model; we are now in the earliest stages of true decentralization of the assembly, production, and distribution of information at the level of the individual citizen. In a recent report, the Office of Technology Assessment observed that "until recently one of the best barricades against breach of privacy was the difficulty and impracticability of integrating all of the public data about a person." With the increasing maintenance and public availability of data from all published sources, including arrest data, credit and banking data, property records, travel records, consumer purchase records, employment records, and medical records, "storage, retrieval, and processing capabilities of modern computer technology [permit the] construct[ion of] comprehensive pictures of an individual from a myriad of transactional details. . . ." These developments affect the quantity of information available and the time frame of its effective retention and utility. They also affect the meaning of information published about individuals, for meaning is a function not simply of facts, but of context — the number, types, and relationships of facts assembled. The dangers posed by vast amounts of accessible information are compounded by developing technologies of access, assimilation, and distribution, technologies that can transfer editorial control to the audience. According to the Office of Technology Assessment: Technologies such as expert systems are emerging, which when used in conjunction with electronic publishing, can disperse editorial control to the recipients, rather than the originators, of news stories and information. Citizens may come to see "news stories" not as a standardized, authoritative, and "balanced" text, but as a largely self-selected collection of sources to be assayed in context. Through electronic publishing and related technologies for assembling and distributing information, "the style, organization, order, and content of [information may] be selected by the user, rather than the publisher. The converse is also true: writers and publishers may choose their readers, and differentiate their products across classes of readers." Technology, in short, is causing profound changes in the meaning of news. Today, the power to define and administer aspirational standards for news is scattered in a broad array of decentralized locations. Metropolitan papers and national broadcasters no longer control news and its meaning. The power of making news judgments is distributed among and by newspapers, magazines, newsletters, broadcasting outlets, and decentralized data transmission facilities. Meanwhile, the audiences for news are in the process of being enlarged and specialized — indeed, of becoming actively engaged in the news process itself. As a result of these changes, the news profession as we have known it will be increasingly foreclosed by technology from identifying, much less enforcing, common aspirations and conventional processes. The very definition of news, and its historic dependence on process rather than substance, is being radically changed. In fact, the concept of news as presupposed by Warren and Brandeis is losing meaning; truth, as Lippmann saw it, is anachronistic. This has created an unbreachable gap between any theoretical definition of news —much less common standards that influence decisions to publish private material — and the practical realities in which we find ourselves. Any attempt to link the legal construct of privacy to a theoretical notion of news as publication is an enterprise doomed to futility. B. Communicative Assumptions of the Privacy Tort Allegory, mother of all dogmas, is the replacement of the seal by the hallmark, of reality by shadow; it is the falsehood of truth, and the truth of falsehood. -- Elphias Levi Almost from its beginning, the privacy tort has contained a paradox, for a privacy invasion is at once allegorical and deeply personal. On the one hand, the tort grew out of societal concerns, including the excesses of the press, changes brought on by a more urban society, and the impact that a disclosure has on the individual's relationships with other persons. It addressed the tension between the force of a story and its tastelessness as news. To violate the right to privacy is thus a communicative tort. On the other hand, the harm on which the tort has principally focused has been personal and intrinsic. Damages are not awarded for relational harm to the individual, nor is proof that particular types of persons received and were influenced by the disclosure required even as a check on the plaintiff's own actionable perceptions or emotional turmoil. The tort is therefore directed primarily at remedying psychic or emotional harm, although liability is predicated on conduct judged in part, at least, in terms of social utility or conformance with social norms. The privacy tort, in other words, serves two masters. The communicative meaning of private information -- allegory -- is radically different from the intrinsic harm to the individual. 6 Like Professor Robert Post, I read Warren and Brandeis as having been principally concerned with the social (or, as Post put it, "communitarian") dimensions of privacy invasions. No matter which dimension is dominant, elements pertaining to the existence and presumptive social character of the disclosure and to the quality of public discourse are pertinent. At the very least, they measure whether the disclosure "gives publicity," concerns "private life," is "highly offensive," and is "of legitimate concern to the public." We shall return to the paradox and its implications for privacy, but I first want to address the extrinsic, communicative aspects of the privacy tort, for any view of the tort's nature rests heavily on communication — a public act with important social implications. The privacy tort shares with its relative, the libel tort, a largely silent set of assumptions about what communication means and how it works. Perhaps because of its recent origin and because of the Warren and Brandeis article, however, the privacy tort's assumptions about communication are less sophisticated than those underlying the libel tort; they have not, after all, been forged by centuries of experience with the subtleties and nuances of causation that apply to claims of reputational harm caused by expression. Yet, like libel, the privacy tort rests on the communicative process as the essential precondition of liability. It explicitly presumes that harm can arise out of communication. It recognizes, as well, that in certain settings other (largely social) benefits of a communication will override liability that otherwise would exist. The privacy tort therefore assumes a communicative process that is sufficiently coherent and predictable for the imposition of legal liability. What is the nature of that communicative process, and what implications does it have for the privacy tort, especially in light of social changes that have occurred in the last 100 years? Without pretense of exhaustiveness, I identify three fundamental, yet flawed, assumptions about "communication" that underlie the tort and that merit attention. 1. Complexity of the Communicative Process The first assumption the privacy tort makes about communication is that it is a relatively straightforward process: "private life" can be given relatively clear legal meaning; "offensiveness" can be measured by reducing it to the reasoned and reflective judgment of the "reasonable person"; and "legitimate concern to the public" can be addressed as an analytical question judged almost exclusively by the text of a statement abstracted from its communicative setting. But the communicative process does not, in fact, reflect a linear relationship between speaker intention, textual meaning, and audience perception. Instead, communication is a process of negotiated meaning and iterative reformulation, beginning with the speaker, but including the speaker's purpose, the text, the medium, the audience(s), and the social framework in which messages are given both content and consequences. The meaning ascribed by an audience — indeed, the very message received by an audience — is as much a function of the particular audience as of the speaker or publisher. This ascribed meaning may depart radically from the speaker's intention, the text, and/or the post hoc reflection and rational judgment of the "reasonable person." It may make irrelevant, in view of the construction of an audience's meaning, any question of legitimate public concern, and it may likewise raise that issue with statements that do not, on their face, suggest it. Unlike in the 1890s, when the legacy of the penny newspaper held sway and when "fact" and "objectivity" were the imperatives of news, today we better understand that reported facts are constructions of meaning. We better understand, as well, that privacy invasions, most often, are stories whose public significance lies outside of the text. They are allegories from which we learn about ourselves rather than about the individuals involved. When confronted by the rape of a child, we rage and grieve, though not so much for that child as on behalf of all children; when we are made witness to the despair of the imprisoned mentally ill, we despair not so much for an inmate's particular plight, but for what we have wrought. Thus, truth or falsity, specific names and places, frequently have little to do with publications that invade privacy — little of lasting consequence, at least. "[T]he newspaper," according to Michael Schudson — to which one might add all sources of private information about the affairs of others -- "acts as a guide to living not so much by providing facts as by selecting them and framing them." To the extent that journalism is storytelling (and it has been to varying degrees in the past, and is almost always so with privacy invasions), we cannot understand the privacy tort without understanding the crucial storytelling function performed by private disclosures. According to Schudson, commenting on the aesthetic and storytelling function of news in an urban society at the turn of the century, "[people] wanted the moral counsel of stories . . . but the tales of the Bible and the lives of the saints were not suited to the new cities." Understanding privacy, therefore, requires a recognition that the private material published is not the end of the matter; the value of a disclosure is an interpretive question, as is its meaning, with the subject of the disclosure often playing an incidental role. 7 Yet the privacy tort largely forecloses lines of inquiry adapted to addressing real meaning. It assumes that a private fact disclosed to others is understood to be just that, and nothing else. It assumes that others give meaning to a fact in terms of the person to whom it relates. It assumes that the meaning ascribed is narrow, individualized, and often salacious. It assumes that the relevance of the disclosure to a subject of public concern can be judged strictly by the fact disclosed and the person to whom it relates, and no more. In sum, the privacy tort assumes one and only one audience consisting of people who, without context, perceptual differences, or prejudice, act on a clean slate with perfect rationality. In making these assumptions the privacy tort avoids the oft-described Byzantine character of the common law libel tort, with its many presumptions and privileges. But it also lacks the experienced, commonsensical, and sophisticated understanding that has allowed the libel tort to escape the many pitfalls of a simpler approach. The privacy tort ignores the critical fact that communication is a process of socially constructed meaning, that meaning exists as often in the allegory as in the text. 2. Appropriate Limits on Public Disclosure The second assumption underlying the privacy tort concerns the impact of a disclosure in its social or cultural setting and the appropriate limits on the type of material subject to public disclosure. At the time Warren and Brandeis wrote, a much broader and clearer consensus on questions of civility and of moral value existed. As a result, the accepted bounds of decency and propriety in public discourse were more easily defined and more broadly shared. It was less difficult to mark boundaries of private matters. This was in some part because of the closer, rural roots in the society of 1890, and in larger part because the society was then much less culturally, ethnically, and racially diverse than it is today, or at least less openly and purposefully so. Inherited conventions, especially those pertaining to family life and interpersonal relationships, were largely unquestioned and a relatively fixed set of moral standards was widely recognized. While those standards were far from universally reflected in actual conduct, there seems, as well, to have been a more clearly defined line of demarcation between values one publicly held, that were fair game, and those one subscribed to in the private realm. At least this was more clearly so than it is today, when the line has virtually disappeared, and when conventions governing broad and explicit dissemination of personal information are neither shared nor effectively enforced in the journalism community or in the public at large. To the extent that rules of civility in public discourse were widely shared at the turn of the century, and to the extent that public standards of morality were relatively uniform, few problems existed with a tort that visited legal consequences on communications that violated those conventions. The same level of confidence cannot be enjoyed, however, by a tort that imposes standards of civility in the 1990s when commonly held notions of civility do not exist. If, as I suggest, standards of civility and decency do not today enjoy broad consensus grounded on commonly held societal values that we bring to the interpretive communication process, then we can no longer safely assume that a predictable set of consequences will flow from a communication that violates earlier conventions. Neither may we safely make any general assumptions about cause and effect from such communications, other than in the mind and psyche of the subject of the disclosure. 3. Receptiveness to Disclosures A final and related assumption made by the privacy tort is that there are a limited number of relatively homogeneous "communities" in which a communication of known meaning will be interpreted and its consequences felt. We assume, for example, that a disclosure concerning intimate sexual habits, or one focused on personal grief at the loss of a child, will be given fixed context and meaning, will evoke uniform responses such as disgust or pity, and will strike most everyone as exceeding the bounds of decency. We also think for different, albeit unarticulated, reasons that other disclosures, such as those concerning a person's racist attitudes, are less likely to be appropriately considered private. This is either because we have a broad social consensus on such matters or because we assume that few communities exist, and none that should be legitimated, where such beliefs are both acceptable and appropriately private. But we are wrong. Today's society is far too diverse, personal values are too pluralistic, and personal associations are too different in the norms they reflect to support such an approach. For some people, the public venting of attitudes and behavior is to be desired. In some associational communities behavior many consider deviant would be laudable. For example, in some quarters the public disclosure of sexual conduct, even against the wishes of the subject, is viewed as socially valuable and therefore justified. The point I am making is one that relates to the nature of privacy itself, of course, but it also bears on news and communication, for adjudicating the tort in specific cases involves more than crafting abstract social policy. The privacy tort is based partly on how the individual subject to disclosure perceives that disclosure, and partly on how society would view it. It thus compels us to look at and coalesce questions of social tolerance and individualized perception. It assumes ascertainable criteria for both questions and, more importantly, a correspondence between them. Since in the 1990s neither criteria nor correspondence seem likely, the privacy tort may ask the impossible. It expects us to judge the validity of a person's perceived violation in the unique and increasingly idiosyncratic context of her or his social relationships, informed by a recognition that for some people, in some settings, facts will be interpreted and employed in very different ways. We must, in short, recognize that the privacy tort rests (and perhaps has rested from its inception) on communicative foundations that are tenuous at best and that may be stretched to their limits if we press them too hard. Privacy is a communicative tort. It rests on harm arising from communication. It therefore must reflect a realistic understanding of what communication means and how it works. It must reflect the fact that, in communication, meaning is a complex and negotiated phenomenon; that when a threshold definition of the message communicated can be ascertained, difficult issues remain about how and whether people will interpret a message, and what significance they will give it. These are questions that can never be answered with complete exactness; they are questions of sociology, not logic, and they cannot, especially in a highly diverse and pluralistic society, be resolved by global assumptions. C. News and Process Objectivity resides not in the quality of the product but in the mode of the performance. —Bernard Roshco News is as much a product of process as it is of substance. With respect to privacy interests, as with all else that is published as news, the process of judgment subserves its substance, and vice versa. Process contributes three essential qualities to news: it promotes the application of substantive standards of newsworthiness in an intellectually reasoned way; it undergirds the claim that selection of material represents more than caprice or the mere accommodation of audience preferences; and it provides a measure of consistency in judgment across the profession and within given organizations. In short, it assures a constitutionally important intellectual distance between the publisher and the audience. As the United States Supreme Court has said, "editing is what editors are for; and editing is selection and choice of material." 8 Process is perhaps more important in privacy settings than in most others: the substantive criteria for privacy are elusive and aspirational at best; the risk of harm, and therefore the care required, is high; and the ultimate decision about publication is almost wholly qualitative and subjective. In these respects, privacy interests are very different from reputational interests for an editor. With reputation, while the risk of harm is still high, substantive criteria are better developed and more generally acknowledged, and the objective benchmark of falsity exists. While the relationship between a published fact and the subject of the fact is the purpose of publishing a reputational fact, the news purpose of a disclosed private fact, in contrast, is not necessarily limited to publication of information about the subject. As discussed above, often the disclosure is instrumental; it serves an elucidating, illustrative, and larger educational purpose related to other newsworthy events or to our comprehension of the world in which we live. This purpose could be achieved without the private disclosure, but not as forcefully, appealingly, or persuasively. The decision to publish a private fact is therefore often literary in quality —an instance of pure selection in which the editorial process reflects a judgment that the disclosed fact has valuable and instrumental bearing on a different or larger matter. The judgment is at once more subjective, indeed almost aesthetic, and less subject to challenge. Its justifications are grounded not in relevance or public interest, at least as they concern the disclosure itself, but in literary and artistic notions of forcefulness, impact, and understanding. When Warren and Brandeis wrote, the concepts of editorial judgment and editorial process were less mature and more controlled than they are today. News organizations were just beginning to develop. Journalism as a "profession" was largely inchoate. It was a time of experimentation, both with the function of news itself as well as with its standards. But news was also more controlled within the framework of relatively few news organizations. Process and standards thus had meaning, both within the trade and within organizations, even if both were largely the product of autocratic power. Today, standards of news, even of judgment concerning publication of private material, are relatively well developed, and the companion idea of editorial process has taken on definite meaning. These developments have been the product of much experience gained during the past 100 years. They have resulted from the expansion of news organizations and their consequent institutionalization, and from the development of professional organizations and academic disciplines in the fields of journalism and communications. But while the concepts of news and process have in some ways become more coherent, they have simultaneously lost some coherence due to the fierce independence, competition, expansion, and fragmentation of the media, as well as profound technological change. In 1990, technology and related economic changes are "ushering in a convergence of forms of press publishing that were once partitioned by technology: print publishing, mail, broadcasting, and telephone." This phenomenon challenges conventional theories of the press that assume "the press — be it newspaper, journal, magazine, radio, or television —is capable of exercising editorial discretion over the content of its publications." In the latter half of the twentieth century, the difficulties presented by diverse processes of news judgment have been compounded by the emergence of new technologies of production and publication, requiring in each case different adaptations of process in very different settings. Perhaps more importantly, diversity of process has been compounded by the fragmentation and specialization of audiences in all media, and by the true decentralization of the power of publication and distribution made possible by communication and computer technologies. Fragmentation and publication on a smaller economic scale have brought about two changes. First, they have increased the variety of people and settings involved in publishing news, often at the sacrifice of the economies of scale needed to support thorough and consistent editorial processes. Second, fragmentation and decentralization have brought the publication decision closer to a coherent audience, heightening the pressure to respond to audience desires more directly — freed of the encumbrance of independent news judgment. This poses a substantial risk for privacy interests, because the self-restraint that is essential to the protection of personal privacy is largely dependent on the separation between news and audience that editorial judgment and process provide. The audience will generally vote for invasive disclosures, and the closer the process of publication is to the audience, the more it is subject to dominant audience preferences and the more likely it is that invasive material will be published. D. The Structure of Communication and Its Implications for Privacy In 1890, communication of general and public information of current interest took place almost exclusively through news channels, principally the daily newspapers. Most of the investigative or other reporting work, moreover, occurred in relatively few large organizations, and focused on information of national, regional, or local appeal to a large public constituency. Today, the institutions of information are far more varied, smaller in relative scale, and increasingly focused on particular and specialized audiences and interests. Changes at the institutional level, however, are but manifestations of deeper and more fundamental forces that are more directly related to the nature of news and the vindication of privacy interests. The economies of scale of systematic investigative reporting, and even of the daily acquisition of information, are in many respects not very different today than they were in 1890, at least at the conventional daily news level. The availability of other technological means of obtaining information within a broad range of localized community institutions, however -- including not only newspapers but also special interest publications, magazines, and services, as well as more highly discrete sources of communication -- is far greater. More importantly, the economies of scale for dissemination of information have changed since 1890; "publishing" is becoming vastly less expensive and more accessible, a trend that will continue as computer and communication technology improves. Finally, broad databases of information, which provide a contemporaneous and historical picture never before available (even in the best newspaper morgue), and which permit the mere construction and combination of facts, without more, to create new meanings, are now in place. As access to these sources of information by small organizations and individuals increases, the very idea of a news organization and a coherent news process will change, and the protection of privacy, even in the context of published current information that we think of as "news," will take on a very different complexion. 9 These changes already have had profound implications for privacy. In many respects privacy claims in the conventional setting of published news — the setting to which Warren and Brandeis directed their attention — have represented no more than a series of isolated, and often petty, grievances. This is because news in its traditional institutional framework fundamentally is not concerned with or structured to obtain purely private information. There are exceptions to this general rule, to be sure, but the exceptions are very few in number, as are the privacy claims brought to litigation. Private information and conventional news do not generally mix because news requires not only interesting and visible information, but also available information. As Bernard Roshco put it, a journalist's "ability to 'observe' . . . newsworthy events is a consequence of the way institutions are organized." "[T]here is a drive toward the 'self-fulfilling prophecy' in what is reported and published. Reporters cover beats -- or are assigned to stories -- that are expected to be newsworthy. What they file is then published as the most newsworthy events of the day." Historically news has been structured through efficient and centralized organizations. This has meant that sources of information tend to arise in conventional contacts, typically with public institutions and people, and at the intersection of events and public offices. This is true in all cases involving mass dissemination through large organizations, because it is the only economically efficient way to catch most general information with a relatively small and nonspecialized work force. The economies of scale demand a conventional structure of information gathering in order to achieve an efficient news operation on a large scale. As sources of information, which in this way have been the news, become less institutionalized and more decentralized, the conventional structures for obtaining information will become less dominant, and the meaning of news will shift. In 1890, information was maintained on a decentralized level (in the separate "morgues" of each newspaper). But its availability was centralized — held closely as the property of each paper, with the economies of scale for the collection, assimilation, and communication of information demanding large, and few, organizations. Today information is maintained on a centralized basis in large and increasingly accessible databases. Its availability, however, is highly decentralized. And computers and communication technology permit the collection, assimilation, construction, and publication of information to take place on a highly local and decentralized level. For example, today a local reporter or a special interest group can obtain quantities of information on an individual that would have been unthinkable twenty-five years ago, and do so without leaving the office. The available information includes -- or soon is likely to include -- date and place of birth, marriages, children, places of residence, law enforcement records, public stock holdings and transactions, real property holdings, job and income (at least for the one third of the population that is publicly employed), credit records and ratings, borrowings, and everything that has ever been published about the individual, her or his family, and her or his associations. No newspaper has ever had such a "morgue." But if the use of such information by the traditional, large-scale news publisher were our only concern, the privacy issue would be less pressing, for much of the information would not be useful to such publishers and editorial processes would filter much of it out. It is the changing structure of "news" — its increasing distribution through small organizations, special interest groups, networks, businesses, and individuals that dramatically redefines the nature of the concern about disclosure of private information. Each segment of the traditional and organizationally unified news process — the acquisition of information, its assimilation and translation into news, and its distribution -- will be technologically freed from the other and able to exist in independent and specialized entities. In view of the decentralization of access to information and to the power to disseminate it, and the greater likelihood that small and specialized organizations will find such information economically valuable, it would be a serious mistake to focus concern about privacy on the daily newspaper or broadcaster and the established press. E. Privacy as News in 1990 Many of the themes that marked changes in the meaning of privacy between 1890 and 1990, such as pluralism, the breakdown of widely enforced social norms, individualism, and decentralization and specialization in personal relationships, are reflected as well in the changing culture of news. The breakdown of broadly enforced social norms has blurred the lines between what is important and what is not, and therefore substantive criteria governing news have become deeply ambiguous, if not unascertainable. With growth in communication through decentralization and increasing fragmentation of audiences, mechanisms of enforcement within the profession have become ineffective, negotiated meaning within diverse audiences has become infinite, and the stabilizing processes of news, even if broadly enforced, lack substantive and interpretive consensus on which to function. Accordingly, it is hardly surprising that the privacy tort has life only in the minds of academics today. For good reason it has been all but formally interred in the real world of tort litigation. The changing sociology of news discussed in this Part leads me to two conclusions bearing in different degrees on the relationship between privacy and news or journalism. The conclusions are paradoxical. I believe that the interests protected by the privacy tort must reflect social, rather than personal and emotional, well-being. I am also of the view, however, that it is impossible in the 1990s (perhaps it was even so in 1890) to construct a communicative tort that serves privacy interests in their social manifestations, that is necessarily based on generalizable and principled standards of propriety, that is visited on the point of publication, that reflects predictable audience interpretations of meaning, and that is enforced through reasoned processes. The first conclusion concerns the interests protected by the privacy tort. As I have mentioned previously, the privacy tort has two faces: one is outward-looking, the other inward-looking. The inward-looking dimension protects emotional stability, compensates emotional distress, and requires little in the way of outward (i.e., societal) manifestation of harm to the subject of invasive publicity. It is not so much concerned with our social conventions as it is with the individual's private and personal perceptions. While, as I have stated, I doubt that the idea of privacy articulated by Warren and Brandeis had much to do with this form of harm, except as a parasitic element of damage, it is perhaps a reflection of our greater measure of individualism today that the current rendition of the privacy tort rests heavily, if not principally, on this dimension of privacy. I seriously doubt whether, as a matter of policy or as a practical ingredient of news judgment, the inward-looking aspect of privacy can be accepted, even in parasitic form, in today's tort law. Emotional distress, which is really what is at issue in this dimension of the privacy tort, is a highly problematic legal interest under any circumstances. It suffers from its dependence on the idiosyncrasies of the subject and, as the legal history of the privacy tort will attest, surrounding it with "objective" measures such as "outrage" and the like is of little benefit. Ours is a pluralistic and diverse society, with extreme variations in values and in sources of sensitivity. The task of testing the veracity of asserted harms — much less measuring them against a lowest common denominator of societal taste — is essentially an impossible task in the tort-based litigation process, where the threshold standards are left to judicial construction. The United States Supreme Court's recent foray into this area in Hustler Magazine, Inc. v. Falwell speaks volumes on the subject. There are equally important practical reasons for skepticism about a privacy tort based on inward-looking emotional harm. Tort liability must be premised on the realistic belief that the tortfeasor will be put on notice of the tortious conduct. Predictability serves the twin objectives of shaping future conduct and assuring fairness in the imposition of liability. When the harm on which liability is premised is unascertainable because it is so heavily premised on the unique perceptions of the harmed person, and when generalizable rules of conduct cannot be articulated with a meaningful level of concreteness and lack a sufficiently homogeneous institutional environment in which to function, tort law expects more than it should -- and, in the communicative setting, more than it can. Quite apart from the serious constitutional difficulties such a situation would pose, a journalist in even the best and most reasoned editorial process cannot assess the peculiar circumstances and sensitivities of the subject, and cannot balance these against hopelessly ambiguous questions of negotiated meaning, offensiveness in a community, and legitimacy of public interest. I am not suggesting that journalists and editors should not try to make judgments informed by such factors, for they should. I am simply saying that, at least as presently constituted, the law should not supervise that enterprise through a privacy tort. At least in the context of news — the only setting with which I am presently concerned — the types of judgments called for by the tort are simply unmanageable. As a consequence the tort will remain, as it is today, largely unvindicated and, perhaps even worse, a little island of petty grievances. This leads me to my second conclusion, one which I have already ventured: while the privacy tort must reflect societal conventions of discourse, those conventions have largely broken down in 1990, and with their collapse the news process itself has become too diverse to reflect them. In an outward-looking privacy tort, at least one in which liability is visited on publication, criteria must exist by which to measure a disclosure's conformance with standards of taste and value. Without those standards, no amount of editorial process will help. For better or worse, standards of taste or propriety are simply unascertainable on a societal level, as the recent history of the privacy tort attests. Our society is too pluralistic and culturally diverse for any but the most uselessly abstract norms to be enforced. As a practical matter, then, news as a reflection of society's needs and as a text of its evolution cannot coexist with the privacy tort as we now know it. There are inadequate objective benchmarks in social consensus about propriety, standards of value, consistent modes of interpretation, and organizational behavior for the gatekeepers of public information to function. This is not a conclusion based on constitutional concerns, but one rested on the more hard-bitten soil of tort law. Nor is it a conclusion happily arrived at, for a lack of social values and a breakdown of settled institutions are not necessarily to be desired. It is, however, an inescapable conclusion, and one that is supported, if not dictated, by the patchwork and ultimately futile record of the privacy tort so earnestly proposed by Warren and Brandeis in 1890. III. INTERRING THE PRIVACY TORT The idea of privacy expressed by Warren and Brandeis is a powerful one. Its persistence is a testament to its importance and vitality. It represents a manifestation of deeply held convictions about the relationship between the individual and organized society, and about the responsibilities owed by a society through private as well as governmental institutions to the individual. Its vindication through the common law privacy tort, however, has proved ineffective. It has become an anachronism. The reasons for this have to do with social changes in the relationships between the individual and others, evolution in social values, and profound shifts in the nature, function, and organization of public communication. But these changes have not destroyed the idea of privacy; rather, they have altered it. The more fundamental problem lies with the tort, not with privacy. The very ideas of judge-made standards of societal values, of jury-determined thresholds of offensiveness and public interest, and of case-by-case adjudication of law as well as fact, are at odds with our shifting social environment and are incompatible with the complex institutional arrangements now emerging in the field of mass communication. By vesting the principal locus of the vindication of privacy in a common law tort action focused on the point of publication, we have expected the impossible -- perhaps from the very beginning. This conclusion is not, in my mind, a wholly bleak one. It does not imply that privacy is not a value to be legally protected. Nor does it imply that privacy is an idea without real and enforceable content. It is a conclusion, however, that points in some different and, I think, more constructive directions. For while social consensus about the meaning of privacy cannot feasibly be ascertained by judges and jurors in the common law privacy tort setting, there is no reason to believe that it cannot be approximated through tort-based obligations shaped in the legislative process. Moreover, while the iterative methodology of the common law fails to produce sufficiently clear distinctions for day-to-day application in the construction of news, there is no reason to doubt that relatively bright lines can be constructed legislatively. Finally, it is clear that the judgments of "private" information, "outrageous" conduct, and "legitimate" public interest can never be wholly escaped, and that their determination in the courts is problematic. These judgments may, however, be more closely captured in legislative language that takes the form, not of a tort action focused on publication, but of identified information placed presumptively in the control of the individual; the focus of adjudication could be shifted from open-ended issues of value to the more circumscribed questions of consent, waiver, and an obligation of confidentiality owed by one possessing such information. This is not the place to develop alternatives for privacy in specific and detailed terms. It is, however, the obligation of the critic to do more than criticize. It is in that spirit, and that spirit alone, that I suggest that the privacy tort be formally interred, and that we look to the concept of breach of confidence to provide legally enforceable protection from dissemination of identified types of personal information. This proposal is made in the interest, not of imposing general standards of social discourse or avoiding emotional distress, but of preserving the function of private associational relationships in our complex and too public world. Enforcement of obligations of confidentiality is not, of course, a foreign idea to the law. Such obligations are recognized in the relationships between professionals and clients or patients, between businesses and customers, and even, quite recently, between reporters and confidential sources. The subject has also been the focus of academic discussion and of a comprehensive Law Commission Report in Great Britain. This is to say no more than that the concept of enforceable obligations of confidentiality is a familiar one. Its application to the privacy concerns developed in this Article, of course, would be new. And while I do not intend to explicate a new tort theory in this Article, two basic elements of such an approach seem apparent. First, the ability to enforce the privacy of any item of information should depend on the intention and conduct of the individual, and on the nexus between disclosure and the protection of constructive social relationships. Second, the legal obligation to maintain the information in confidence should be imposed through tort duties on the possessor of information, not on the publisher. Shifting the focus of privacy to individual control of identified private information, enforced through liability for breach of confidentiality rather than publication to large audiences, will not eliminate all of the difficult problems posed by privacy. It will, however, address two concerns central to the common law: that rules be built on a solid foundation of social consensus that reflects the complexities of social arrangements, and that rules of liability be susceptible to consistent and principled application. It may also eliminate the paradox of the privacy tort: that in serving two masters — emotional harm and societal taste — it fails them both. Interring the Warren and Brandeis tort, or what we have made of it, represents no more than the formalizing of a deed already done. But we should not inter the idea of privacy. It is a sufficiently universal and powerful idea that we could not do so even if we wanted to. We must also recognize, however, that extinguishing privacy will be most efficiently accomplished by keeping the current privacy tort, not by eliminating it, for the privacy tort has become a promise always honored in the breach. 10  
Page
of 00
 |
