I. INTRODUCTION

Ever since Samuel D. Warren and Louis D. Brandeis published their seminal article on the fight to privacy more than a century ago,(1) the contours of privacy both as a moral-philosophical concept and as a legal fight have been subject to persistent scholarly scrutiny and debate. At no time have privacy issues taken on greater significance than in recent years, as technological developments have led to the emergence of an "information society" capable of gathering, storing, and disseminating increasing amounts of data about individual citizens. Whether the collection and subsequent use of this information should be regulated and controlled, and, if so, what regulatory mechanisms should be created, are hotly contested issues within the legal community, as well as in the public and media.

Despite the importance attributed to privacy by the general public and many legal scholars, a unified theory of privacy, whether legal or philosophical in nature, has yet to emerge. For Warren and Brandeis, the fight to privacy represented simply the "right to be left alone,"(2) which in 1890 translated into restrictions on the freedom of the press. Over the past century, the "right to be left alone" has developed into four separate common law privacy torts which provide individuals with certain limited protections.(3) Constitutional privacy law, by comparison, has come to focus on limiting the scope of governmental intrusion into a person's private life and personal decision-making.(4) Isolated pockets of privacy rights have also materialized as both state and federal lawmakers have placed various narrowly delimited restrictions on the ability of governmental and private organizations to gather, maintain, and distribute personal information.(5) Statutory and decisional privacy law alike have developed in an erratic and haphazard fashion; no single theory of privacy, nor even a consistent set of theories, has informed this process. Thus, Dean Prosser's remark more than thirty years ago that the rise of common law privacy has gone on "without any plan, without much realization of what is happening or its significance, and without any consideration of its dangers,"(6) continues to hold true today and aptly characterizes the evolution of privacy law as a whole.

Yet, the need for some measure of uniformity in legal privacy theory and privacy law has never been greater. Privacy issues arise with increasing frequency in such diverse areas as media coverage of personal or quasi-public matters,(7) electronic surveillance in the workplace, and data collection by banks, credit bureaus, and other institutions. One area in which such privacy debate is now beginning to take shape is the emergent world of electronic communications networks. This article focuses on privacy issues relating to electronic communication networks where the tension between individual privacy rights and commercial interests in free information access is coming sharply into focus.

The article rests on the premise that a body of privacy law which develops without the benefit of a unified theory of privacy will prove questionable at best. Only by deciding a priori what it is that matters about privacy, and by establishing a comprehensive set of policy guidelines will we be able to adapt our privacy laws to a rapidly changing socioeconomic context. This article draws on philosophy and social science research in an effort to develop a theory of privacy that may inform the development of a legal right to privacy in the newly emerging social order of electronic networks.

Part II of this article seeks to construct a theoretical model of privacy which incorporates privacy's moral value and its broader social function and meaning. That model conceives of privacy not merely as a means of enabling the individual to create and maintain a coherent self-construct, but also as a foundation for negotiating the content of social relationships and for distributing social and political power.

Part III examines privacy from a socio-historical perspective, to demonstrate that the Western concept of privacy is based on cultural notions of property and territoriality.

Part IV applies the privacy model to the world of cyberspace. Specifically, this article argues that cyberspace should be regarded not merely as a technological innovation facilitating communication but also as a cultural sphere characterized by distinct forms of social organization and rules of interaction. The unique social and geopolitical structure of the electronic communications environment weighs against reliance on traditional assumptions about the nature of privacy and requires us to take a fresh approach to privacy rights which will limit others' ability to engage in surveillance and information gathering.

Part VI hence examines the reasons for extending privacy rights to cyberspace communications and discusses probable consequences of privacy deprivation. Those consequences are made comprehensible by the theoretical privacy model developed in prior sections, which regards privacy as a means both of maintaining individual personhood and of constructing interpersonal relationships. The author contends that the unregulated collection of commercial data engenders alienation because it deprives the individual of control over access to his personal information. Nonregulation also allows those who collect and deal in information to accumulate social power at the expense of the individual, a process which ultimately could inhibit the development of political democracy in cyberspace.

Finally, Part V proffers some suggestions for cyberspace privacy regulation. As various authors have noted, existing regulations aimed at addressing privacy concerns in the context of electronic communications are inadequate. This article also critically examines the idea that a regulatory framework should be permitted to develop gradually, either by relying on the self-governance of network communities or by allowing privacy safeguards to emerge through contractual means. Neither alternative recognizes the power disparities between individual network users and information brokers. Moreover, neither alternative presents an effective means of creating a uniform and coherent approach to cyberspace privacy rights. Instead, this article argues for the establishment of a federal regulatory mechanism as a vehicle for developing a coherent privacy framework that is adapted to the specific contingencies of an electronic communications environment.

II. CONSTRUCTING A THEORETICAL CONCEPT OF PRIVACY

As previously noted, there is little consistency among the various legal approaches to privacy issues. Statutory, common law, and constitutional privacy rights rest on diverse assumptions about the nature and purposes of privacy. These assumptions are not usually considered by judges, lawmakers, and commentators, who are more concerned with the practical consequences of privacy jurisprudence than with its theoretical underpinnings. At the same time, privacy theorists are typically inclined to regard privacy not as a right but as a philosophical construct with moral, social and cultural--but not necessarily legal--dimensions. As a consequence of these marked differences in emphasis, scholarly discussions of privacy, whether theoretical or legal in nature, only rarely focus on the nexus between privacy theory and privacy law. This article attempts to fill that exploratory gap by constructing a conceptual bridge between moral philosophy, social science theory, and legal analysis.

Such an endeavor is not intended to be merely an academic exercise. Rather, it is grounded in the presumption that, as our rapidly changing socioeconomic environment requires us to re-think present approaches to the right of privacy in numerous social contexts. Consequently, it becomes increasingly necessary to fashion a coherent body of privacy law that may be adapted and molded to fit a wide variety of different factual settings. That body of law can be defended only if it is anchored by a well-understood and definable theoretical framework. Nowhere is the need for a consistent privacy theory more acutely felt than in those areas where a clearly defined legal approach to privacy rights has yet to be developed. One such area is the sphere of electronic communications and, in particular, cyberspace. Later sections of this article discuss in some detail the specific cultural contours of the cyberworld. For present purposes, it suffices to note that cyberspace may be characterized as a wholly novel cultural setting within which specific social, political, and economic structures are only now evolving. As this nascent cultural space takes shape, it will become increasingly necessary to define the rights of those who reside within it and to safeguard those rights with appropriate regulations.

Such a regulatory model must derive its shape and meaning from a theoretical framework which takes account of privacy as a moral value and as a social construct. This part, therefore, explores privacy from the dual viewpoints of moral philosophy and social science in order to construct a privacy model that ties privacy to both the maintenance of self and person and to the pro-cesses by which we create social relations and negotiate social power. The privacy theory developed below is grounded in the moral-philosophical perspectives on privacy propounded by Stanley Benn and Ferdinand Schoeman, but also closely examines the socio-cultural meanings and implications of the privacy construct as it operates within the Western, post-industrial cultural context.

A. A Value Theory of Privacy

This article presents a theoretical approach to privacy which rests on the presumption that privacy within present-day Western society is imbued with an inherently positive value. This theoretical approach also seeks to explain privacy's value in terms of the specific roles that privacy plays within the context of social and personal life. The proposed theory contrasts with value-neutral perspectives on privacy, espoused by a number of scholars who urge us to employ a purely descriptive portrayal of privacy's function in order to avoid making unquestioned assumptions about the content and meaning attributed to privacy by the individual social actor.(8) That descriptive methodology is rejected here because of its restricted explanatory force. Privacy's relevance in a legal context is best understood in light of the value attributed to it in common usage. Public perceptions of privacy, in turn, tend to treat privacy as a desirable state deserving of protection:

Both our privacy intuitions and linguistic usage supports [sic]

the "valued" nature of privacy. Our ordinary language reflects a

predisposition toward treating privacy as a positively valued

condition; for example, the response that comes to mind when

someone announces that they have lost privacy is to commiserate

with them, rather than to offer congratulations.(9)

The idea of "undesirable privacy," in contrast, appears counterintuitive.(10) Value-neutral, descriptive approaches by definition eschew consideration of the qualitative content of the privacy concept and focus instead on its function in separating the individual by restricting access to his or her person. Yet, these separation-based accounts cannot explain why privacy, unlike other conditions of "being-apart-from-others" such as alienation, loneliness, or isolation, is expressly sought out and often protected by law.(11) A mere requirement of separation seems to render the idea of shared privacy an impossibility; on the other hand, not every intrusion upon separateness is necessarily regarded as an invasion of privacy.(12) Attempts to formulate neutral definitions of privacy therefore tend to "strip the concept of much of its intuitive meaning and explanatory power."(13)

Only a detailed consideration of the particular cultural and moral values with which present-day Western society has come to imbue privacy will enable us to fully comprehend its legal significance. The initial goal of the instant analysis will be to gain an understanding of the underlying reasons for the specific values attributed to privacy in post-industrial Western culture. This undertaking will make it possible to elucidate the relevant features of the Western privacy ideal and to determine the socio-cultural conditions that give rise to a need for privacy in one or more of its different legal emanations. Once it is understood what roles privacy plays in contemporary culture, it will be possible to approach the privacy concept as a dynamic force which can adapt to changes in society. As previously noted, this article argues that cyberspace must be regarded as a new and distinct cultural setting. An examination of privacy's value in contemporary Western culture, therefore, will provide a comparative baseline against which it may be possible to measure the moral and social relevance of privacy within the context of the new electronic communications environment.

B. The Moral Value of Privacy

Moral, ethical, and legal precepts surrounding the privacy concept are culturally based and reflect certain structural features inherent in the socioeconomic systems within which they arise. This is one of the core themes of this paper, discussed in greater detail in Part HI. Rather than seeing privacy as a static, unchanging feature of human existence, theoretical approaches should regard privacy as a dynamic, adaptive process that derives full meaning only from its broader cultural context. Seen from this perspective, an examination of the content and form of particular privacy notions is likely to reveal substantial cross-cultural variation.

The fact that privacy represents a fluid concept subject to adaptation in response to changed social conditions has important consequences for the development of privacy law. If it is acknowledged that privacy does not constitute a fixed product of natural law but rather an arbitrary cultural construct whose features are dependent on the socio-cultural setting from which they emerge, then it also must be recognized that the legal right to privacy is far from immutable and, instead, must adapt to the requirements of existing structural conditions. The privacy rights afforded an individual in a given legal system are not necessarily readily transportable from one cultural setting to the next. One primary goal of privacy theory, therefore, should involve delineating specific structural preconditions for different forms of privacy needs. With this goal in mind, the cultural features of cyberspace must be analyzed and compared to those of the broader social setting within which cyberspace is situated before it is possible to determine which types of privacy fights should be afforded to electronic network participants, and in what manner these rights should be developed and formulated.

Traditional moral philosophy does not provide an adequate framework for evaluating the value of privacy, because it accords little regard to the socio-cultural setting of value systems. Ferdinand Schoeman, for example, has attacked the tendency of moral theory to consider cultural contingencies as mere impediments to philosophical insight. As Schoeman remarks, "the typical models of human nature [developed by moral theory] ... are oblivious of our existence as cultural beings and discordant with empirical discoveries in social and cognitive psychology."(14) Therefore, a major "tradition of moral theory generally sees cultural susceptibilities and socializing tendencies as standing in the way of moral understanding, moral abilities, and moral fulfillment in autonomous living."(15) Schoeman rejects such views, arguing that "we cannot adequately understand human moral nature by disregarding our cultural dependencies and social vulnerabilities."(16) Schoeman urges moral philosophers to take account of the cultural embededness of moral values and to integrate an understanding of the actor's social experience into the analytical process.(17)

The deficiencies of traditional views of privacy grounded in moral philosophy should not, however, overshadow the significant benefits that may be derived from moral theory. Following Schoeman, this article suggests that a comprehensive understanding of privacy's value and meaning may be reached only by combining theories based in moral philosophy with parallel approaches to privacy developed in the social sciences. This viewpoint recognizes that the emphasis of traditional moral theory on universal aspects of the human condition is not without merit. To the contrary, considerable insight may be gained from positing the question, as does Stanley Benn, whether there are certain basic features of privacy that "rest on something a bit more solid than mere cultural contingency."(18) Although Benn concedes that there may be great diversity among different cultures in both the form and substance of privacy concepts, he nevertheless submits that privacy in its most basic form represents a necessary consequence of human existence. In particular, Benn argues that "some minimal right to immunity from uninvited observation and reporting is required by certain basic features of our conception of a person."(19) Although there are significant variations in the meaning accorded privacy among different cultures, some rudimentary form of privacy may also represent a basic aspect of the human condition itself. If that is the case, then fundamental privacy needs must be accommodated within any system of privacy law, and it will be necessary to examine closely the essential nature of these privacy needs. A first step, however, is to determine whether privacy is in some manner universal.

1. Is Privacy Universal?

Benn's suggestion that privacy, in its most basic expression, may be a universal phenomenon finds considerable support in anthropological literature. One examination of privacy's value in a cross-cultural setting is Alan Westin's survey of privacy concepts as recorded in the Human Relations Area Files at Yale University.(20) Based upon the available data, Westin submits that at least four privacy-related features may turn out to be cross-cultural constants.(21) First, in virtually every society, individuals tend to make use of social distance and avoidance rules in the course of social interaction.(22) Second, almost universally, individuals believe that they are never truly alone, most likely as a consequence of an underlying fear of isolation.(23) Third, there exists a general tendency on the part of individuals to invade the privacy of others and on the part of society to make use of surveillance to prevent antisocial conduct.(24) Finally, as society becomes more complex, physical and psychological opportunities for privacy tend to increase.(25) Westin's observations seem to suggest that at least some fundamental, rudimentary aspects of privacy form an integral part of social life, preserving both the individual's separateness from society and his existence within it. Both the need to set oneself apart from others and the urge to breach that same separateness when it is claimed by others appear to emerge as universal features of every social setting. Privacy thus seems to be implicated with equal force in the creation and perpetuation of social relationships and in the maintenance of social control.

Even societies which at first glance appear to accord little value to notions of privacy can be shown to contain elements of social distancing and avoidance mechanisms, though these elements may not always conform to Western ideas regarding the nature of privacy. One famous example is Margaret Mead's study of childhood and adolescence in Samoa.(26) Mead's ethnography describes a society which "suspects privacy" and devalues physical isolation. From a young age, Samoan children freely witness what Western Society would consider intimate details of personal life, including processes associated with sex, birth, and death.(27) In general, events that call for privacy and seclusion in Western Society are not viewed in the same manner by Samoans.(28) Those events are incorporated into Samoan's daily lives without any apparent need for separation.(29) It might be surmised that Samoa exemplifies one culture in which privacy has no place and is accorded no value. However, upon further investigation it becomes clear that even in Samoa there are relationships of avoidance (for instance, between pre-pubescent girls and older boys)(30) and that Samoans refrain from performing certain behaviors in a public setting.(31) While it is true that efforts often are not made to conceal such activities, it nevertheless appears that such acts are not considered appropriate in a public arena.(32) Although the Samoan privacy concept thus differs substantially from Western notions of privacy, Mead's research suggests that privacy does play a role in Samoan life. Such cross-cultural data strongly imply that certain core features of privacy may be universal and intrinsic aspects of human existence.

This conclusion is of considerable significance when constructing a cyberspace privacy model. First, it suggests that attempts to argue that the nature of electronic communications renders privacy concerns superfluous are misguided, because they ignore the universal, intrinsic need for privacy. Second, it points to the importance of examining the nature of this "universal" privacy need to determine which aspects of personal experience are linked to the privacy concept. Since cyberspace is not an independent cultural sphere but a part of Western culture, it is possible that those features of human experience which require privacy do not exist in the narrow subculture of cyberspace. If that is the case, then the need for cyberspace privacy may indeed be considerably lessened. On the other hand, if cyberspace does contain those features which require privacy, then a thorough understanding of those features will be crucial in determining which aspects of the electronic communications experience should be afforded privacy protection. In light of these considerations, theoretical explanations for the universality of the privacy concept offered by moral philosophy should be examined in some detail.

2. Privacy and Personhood

a. Respect for Persons

If it is possible to identify some rudimentary concept of privacy that is of a universal nature, then the next logical step would be to ask which aspects of human existence render it universal. According to Benn, privacy is necessary because it preserves respect for persons.(33) What makes invasions of privacy objectionable, Benn submits, is the fact that one becomes the involuntary object of another's focused attention.(34) Benn's discussion draws on the philosophy of Jean-Paul Sartre, who argued that uninvited observation generates resentment, not necessarily because one fears disclosure of information obtained, but because the process of observation renders the observed an object of scrutiny and is perceived by the observed as a challenge to his or her dignity:(35)

Finding oneself as an object of scrutiny, as the focus of another's

attention, brings one to a new consciousness of oneself,

as something seen through another's eyes. According to Sartre,

indeed, it is a necessary condition for knowing oneself as anything

at all that one should conceive oneself as an object of

scrutiny. It is only through the regard of the other that the observed

becomes aware of himself as an object, knowable, having

a determinate character, in principle predictable. His consciousness

of pure freedom as subject, as originator and chooser,

is at once assailed by it; he is fixed as something--with limited

probabilities rather than infinite, indeterminate possibilities.

Sartre's account of human relations is of an obsessional need to

master an unbearable alien freedom that undermines one's belief

in one's own; for Ego is aware of Alter not only as a fact, an

object in his world, but also as the subject of a quite independent

world of Alter's own, wherein Ego himself is mere object. The

relationship between the two is essentially hostile. Each,

doubting his own freedom, is driven to assert the primacy of his

own subjectivity.(36)

Unlike Sartre, Benn does not believe that any contemplation of self by the other will necessarily evoke hostility.(37) To the contrary, he believes that the self-awareness generated through the interdependency between subject and object may create a bond through which each recognizes the other as a subject commanding dignity.(38) But, observations can fail to recognize the person observed as an autonomous being, by treating the observed as an object of scrutiny rather than as a decision-maker.(39) Antagonism arises if one fails to respect the observed as a person: "To respect someone as a person is to concede that one ought to take account of the way in which his enterprise might be affected by one's own decisions."(40) Insofar as one person's uninvited observation of another alters the subject's consciousness of himself and of his relations to the world around him, it fails to respect the subject.(41) Therefore, even watching another in secret is objectionable, because it deceives the other about the circumstances of his actions.(42)

Benn's theory has various implications for the development of a legal privacy model. Benn concedes that general principles do not present specific solutions to moral problems. General principles do, however, help determine which actions require justification and distribute the burdens for doing so.(43) Accordingly, the principle of respect for persons does not curtail the freedom of others to observe. Instead, it places upon the observers the burden of justifying the need for observation on the basis of some other overriding principle.(44) The presumptive immunity from intrusion created by the general principle may therefore be overruled, for example, in cases where an individual (such as an entertainer) actively courts publicity, or where certain aspects of his personality are matters of public concern (e.g., the integrity of a Supreme Court candidate).(45) Conversely, the presumption of immunity is stronger in those areas of life that are expressly marked off as "private affairs."(46) Here, privacy is guarded far more closely and a far more serious justification is needed for any attempt at intrusion.(47)

Benn's model regards the right to privacy as an underlying baseline, from which deviations are permissible only if they can be justified. Therefore, in the great majority of situations, the burden of justification rests not on the passive object of observation but rather on the active observer, who often seeks to derive some personal gain. As will be seen, this basis for distributing the burden of justification has frequently been ignored by those creating privacy law and policy. Some policy makers tend to place far greater weight on the intruder's need to obtain information than on the observed's desire to maintain control over the integrity of his person.(48) Nevertheless, as we construct a model of privacy which is appropriate for an electronic communications environment, we should consider Benn's proposition that respect for individual personhood weighs in favor of a presumptive immunity from intrusion.

b. Construction of Self

For Benn, privacy is important because it helps order social relations by requiring individuals to recognize the moral value of the personhood of those with whom they come into contact. But privacy may also be a vehicle through which individuals construct a healthy sense of self and come to view themselves as autonomous beings. Most students of modem self theory regard the self not as a static conglomeration of personality traits, but rather as a flexible, ever-changing construct that is intersubjectively created and negotiated in the process of social interaction.(49) These approaches often draw heavily on the symbolic interactionism of George Herbert Mead, who proposed that the self represents first and foremost a product of social relations between the individual and those around him.(50) Most significantly, Mead contended that the individual acquires a sense of his own person only by taking the point of view of the other and regarding himself as an object of the other's contemplation:

The individual experiences himself as such, not directly, but

only indirectly, from the particular standpoints of other individual

members of the same social group, or from the generalized

standpoint of the social group as a whole to which he belongs.

For he enters his own experience as a self or individual, not directly

or immediately, not by becoming a subject to himself, but

only in so far as he first becomes an object to himself just as

other individuals are objects to him or in his experience; and he

becomes an object to himself only by taking the attitudes of

other individuals toward himself within a social environment or

context of experience and behavior in which both he and they

are involved.(51)

The development of a self-concept, therefore, requires that the actor internalize what he perceives to be the views of others. This process relies on the maintenance of a sphere of privacy that affords the actor an opportunity to focus inward and to engage in a process of self-contemplation. Jeffrey Reiman advocates a similar view of privacy, arguing that privacy functions as "a social ritual by means of which an individual's moral title to his existence is conferred."(52) By according the individual a right to privacy, Reiman submits, the social group recognizes and reaffirms to the individual that he is the sole possessor of his existence and has an exclusive moral right to shape his destiny. Privacy therefore becomes a means of conferring and confirming personhood. At the same time, it is through privacy that the individual acquires a sense of self, "since a self is at least in part a human being who regards his existence--his thoughts, his body, his actions--as his own.(53) Moral ownership of one's body, in turn, entails control over not only action but also cognition.(54)

The social ritual of privacy is thus necessary to confirm the individual's entitlement to freely control access to his body and his thoughts. In support of this proposition, Reiman refers to Erving Goffman's studies of total institutions (such as prisons and asylums), which demonstrate that such institutions exercise control over their inmates simply by depriving inmates of privacy in every respect and thereby destroying their inner sense of self.(55) Total institutions, in other words, bring about "mortification" of the self by penetrating "the private reserve of the individual" and exposing every aspect of the inmate's life to observation by others.(56)

This article proposes that privacy acts as a filtering device, reducing the amount of observations by others in order to provide the individual with a manageable set of external stimuli. Absent some degree of privacy, in other words, individuals would be subject to continual observation by an unlimited number of others. Internalization of all of these observations would severely impede the formation of a coherent inner sense of self because the sheer number of stimuli would prove far too great. By reducing the number of permissible observations and limiting the circle of persons to whom particular types of observations are permitted, privacy becomes a conduit through which data relevant to the development of a sense of self are channeled.(57)

Because privacy implies non-observation, it permits persons to allow free reign to their imaginations and to experiment with different sets of self-concepts:

One aspect of what privacy provides is an opportunity to get in touch

with one's self while not worrying centrally about other people's

judgments; we might call this "being at home with one's self." Privacy

is also a condition for imagining different possibilities--of

freeing one's self from perceived contingencies.(58)

Privacy, therefore, becomes the vehicle for creative production(59) or, as Neville puts it, "the life of creativity."(60) Thus, "privacy is the sphere in which people come to terms with and determine their own existence" as its shapers and creators.(61) Privacy permits creativity, because it protects a zone of introversion and contemplation from intrusion by others--be they individuals, governmental institutions, or private organizations,(62) One of privacy's principal functions is to guard the individual "from the overreaching control of others."(63)

From the perspective of moral philosophy, then, a primary value of privacy arises from the fact that it accords the social actor a sphere of autonomy that others should not penetrate in deference to the underlying principle of respect for the other's personhood. It is this realm of non-observation that permits individuals safely to develop and maintain a sense of self and to engage in the creative endeavor of constituting themselves and their roles in social life. As will be seen, personhood accounts of privacy's value are of considerable utility in building a framework for safeguarding privacy in cyberspace.(64) Interaction in cyberspace values elements of self and person that are often quite different from those which constitute core aspects of the individual's "real-world" persona. Moreover, electronic networks offer their participants unique opportunities for self-creation and for reconstituting themselves as different public personae in different communicative environments.

An effective cyberspace privacy theory must take account of these creative processes and opportunities by ensuring that the scope of permissible privacy invasions is narrowly defined to avoid stifling the generative force of these expressive endeavors. This consideration is particularly important in the early stages of the development of electronic communications, during which privacy intrusions can easily become overpowering. In such contexts, privacy safeguards can foster the evolution of nascent forms of interactive environments which implicate different aspects of personhood and self-expression. Virtual communities which provide an occasion for experimentation with social role formation should not be discouraged simply because the activities of their members may potentially be compromised due to unchecked surveillance and data collection.

C. Relationship Accounts of Privacy

Theoretical depictions of the self are not limited to contemplation of a person's internal self-image but instead distinguish between phenomena relating to the inner self (as the epicenter of consciousness) and phenomena relating to the outer self (as projected by the actor in his social dealings).(65) Accordingly, analyses of privacy which take account of the context of social relationships within which privacy is negotiated implicate notions of an interpersonal self and consider ways in which aspects of the social persona are negotiated intersubjectively.(66)

While personhood approaches to privacy are of considerable value in fostering an understanding of privacy's role in the development of self and person, they all too often suffer from an overly abstract view of human existence. This view fails to take account of the wider social, political and cultural context within which concepts of the person are situated. Both philosophical and legal renditions of the privacy concept have become the target of some criticism. Feminist scholars fault privacy theory for its emphasis on notions of freedom and autonomy, which exalt an overly atomistic view of society.(67) Cultural feminists argue that, instead of focusing on how privacy functions to establish separate and autonomous selves, a privacy theory should take account of complementary feminist values revolving around notions of community and interpersonal responsibility.(68) Lucinda Finley, for instance, stresses that "autonomy is not limited to determining one's actions through separation from others, but includes determining actions by considering others."(69) Cultural feminists contend that Western conceptions of both person and privacy derive from a male-oriented tradition that overemphasizes independence and separateness at the expense of more female values associated with social interdependence and community.(70)

The philosophical privacy literature has also come under attack because its focus on privacy as an individual value detracts from the social significance of privacy. As a result, policymakers are given little incentive to develop a coherent system of privacy protections. In a recent work advancing this critique, Priscilla Regan argues:

When privacy is defined as an individual right, policy formulation

entails a balancing of the individual right to privacy against

a competing interest or right. In general, the competing interest

is recognized as a social interest. For example, the police interest

in law enforcement, the government interest in detecting

fraud, and an employer's interest in securing an honest work

force are discussed and defined as societal interests. It is also

assumed that the individual has a stake in these societal interests.

As a result, privacy has been on the defensive, with those

alleging a privacy invasion bearing the burden of proving that a

certain activity does indeed invade privacy and that the "social"

benefit to be gained from the privacy invasion is less important

than the individual harm incurred.(71)

As the following discussion will indicate, however, not all philosophical approaches to privacy necessarily ignore the social value and consequences of privacy practices. Indeed, it is possible to develop a concept of privacy which addresses such concerns by combining personhood models of privacy with a relational account of privacy. This concept focuses on privacy's role in forming and maintaining social relationships and, ultimately, in acquiring and negotiating social power.

While most relational approaches to privacy have restricted themselves to exploring the relationship between privacy and intimacy,(72) it is also important to examine the role of privacy in fashioning and perpetuating all social relationships. This perspective is concerned first and foremost with the process of negotiating aspects of self and personhood within social interaction. As Schoeman observes, "privacy is important largely because of how it facilitates association with people, not independence from people. This approach suggests that the identification of the right to privacy with the right to be left alone is an incomplete and misleading characterization."(73)

A proponent of a relationship view of privacy, James Rachels, contends that the primary importance of privacy stems from its role in maintaining "the variety of social relationships with other people that we want to have."(74) Rachels points out that both the form and content of relationships are culturally constituted so that each type of relationship entails distinct patterns of behavior.(75) Rachels rejects the notion that such behavior patterns represent little more than artificial constructs that somehow mask an underlying "real" person who conceals his true self by assuming a social role.(76) A far more accurate view of the "person," Rachels opines, acknowledges the context-dependent nature of personhood and recognizes that "the different patterns of behavior are (partly) what define the different relationships: they are an important part of what makes the different relationships what they are."(77) Peculiar to each type of relationship, in turn, is the kind and amount of information about the self that is revealed to others; thus, control over who has access to us implies control over the kinds of relations that we wish to have with others.(78) It follows that the revelation of inappropriate information, or improper access to sell can result only in transformation of the relationship at issue.

Such a transformational view of privacy is, as Robert Murphy has remarked, deeply indebted to the work of George Simmel. Simmel regarded both "self-revelation and self-restraint as necessary qualities of all social relationships, rather than as mutually exclusive categories applying to some relationships as opposed to others."(79) In Simmel's view, social distance stands in an inverse relationship to the amount of knowledge each actor has of the other.(80) All social relationships, however, maintain some margin of restraint, as neither society nor the individual could persist in the face of complete awareness.(81) Privacy, therefore, is the means by which various levels of reserve and restraint may be maintained in the context of different social relationships. Murphy demonstrates the operation of this process in a cross-cultural setting by discussing how male members of the Tuareg tribe of the African Sahara use a veil not only as an item of clothing but as an indicator of social distance.(82) The Tuareg lower or raise the level at which the veil is worn in accordance with the status of the person with whom they interact in order to signal changes in role relationships.(83) In so doing, they are able to demarcate situation-specific degrees of privacy. By viewing privacy as implicated in the process of marking off different role relations, Murphy is able to transcend categorical distinctions between the public and the private, showing instead how persons publicly create and constitute themselves in the context of social interaction.(84) While Murphy's analysis focuses on privacy's role in demarcating particular dyadic relations, privacy may also represent the vehicle through which such relationships are negotiated and transformed. A shift in focus away from introspective self-contemplation to intersubjective negotiation of personal boundaries highlights the close connection between notions of privacy and the exercise of social power. This aspect of privacy theory has largely been ignored in the privacy literature.

Social power is knowledge about others, whether it be about their strengths, their weaknesses, their needs, or their desires. Knowledge, and in particular knowledge of those matters most closely associated with the other's person, enables its possessor to exercise control over relationships and to manipulate their foundation. The potential for transforming the nature and content of specific social relations arises most critically in those situations in which the privacy interests implicated in different types of relationships come into conflict. Thus, role conflicts are likely to arise where information is transmitted or acquired that should not properly be divulged in the context of the particular relationship at issue. A shift in the balance of power and control over the relationship may occur if one party to the relationship manages to capitalize on such tensions by redefining the privacy interests involved in a given interaction rather than succumbing to social pressures to refrain from obtaining or using information regarded as private.

This process of conflict and negotiation is most likely to occur in those areas of social life in which institutionalized patterns of interaction have not yet taken solid form and, thus, are readily amenable to modification and manipulation. Such areas represent cultural frontiers where experimentation and improvisation are commonplace and where participants continually renegotiate the parameters of their various relations. Cyberspace,(85) it will be argued, is a prime example of such frontier territory in which structural features are only gradually emerging.

Some might contend that cultural spaces of this type are best left unregulated in order to permit necessary structural relationships to develop naturally from within so that their adaptive potential may be maximized. But such an argument loses much of its force where the cultural frontier is at bottom a subculture that at once depends upon, and itself shapes, its broader sociocultural environment. Events in cyberspace unquestionably have real-world repercussions, and thus have the potential to affect the general social order. Persons who use cyberspace in order to collect and trade information, for example, may employ that information for various economic and political purposes in contexts far removed from their electronic interactions. Because communications in cyberspace do not occur in a vacuum, regulation of privacy rights represents an important mechanism of ensuring structural consistency. At the same time, the details of such regulation must take into account the unique characteristics of cyberspace culture.

Once we regard privacy as implicated in the construction and transformation of all social relations, its value transcends that of a mere personal right to be left alone. Instead, privacy becomes an integral element of the sociopolitical process. Knowledge is power, and in any social interaction, an imbalance in the amount and nature of personal information possessed by each party creates and perpetuates power disparities. In a society where surveillance and the collection of personal information have become institutionalized, those who control the data collection process have potentially immense social power at their disposal. But their activities are likely to engender distrust and resentment and, insofar as surveillance in public settings is perceived as unavoidable, may cause others to withdraw from public participation altogether. Rather than eliminate the dividing line that separates public from private, information gathering may cause its entrenchment by driving those subject to observation into a strictly private sphere. As a consequence, access to political power is likely to be left in the hands of those able to control the collection and distribution of information.

The very essence of cyberspace is its ability to serve as a repository of information and a conduit for information exchange. Therefore, it is to be expected that the contours of a right to information access should be contested with special vigor in the realm of electronic networks. The task of defining the nature and content of privacy in cyberspace takes on critical importance because information lies at the core of the cyberspace social order. This article suggests that this task can only be accomplished if it takes place within the confines of an appropriate theoretical framework that accounts for privacy's dual role as a vehicle for self-expression and a means of negotiating social relationships. To understand privacy as a concept and to formulate and enforce privacy rights, we must go beyond the static and negative conceptualizations of privacy as a mere condition of separateness(86) and, thereby, recognize privacy's dynamic and creative potential. When privacy is considered in the context of the interactive processes within which it is defined and negotiated, it becomes both a means of discourse and a driving force behind the negotiation and creation of structural relations within society as a whole.

III. TOWARD A SOCIOHISTORICAL UNDERSTANDING OF PRIVACY

The nature of social roles and relationships is, to a significant extent, the product of sociocultural norms and expectations. A relational account of privacy must, therefore, acknowledge the contingent nature of privacy values: "[b]y seeing morality as grounded in culture, we locate its goal in the goal of culture, that being a life in community with others."(87) According to some scholars, the very concept of privacy is specific to post-industrial Western society and lacks cross-cultural validity. This conclusion appears to be an overstatement, given the ethnographic data indicating that certain basic elements of a privacy concept appear to exist among all cultures.(88) Nevertheless, it is equally clear that the form and content of privacy notions are subject to considerable variations. An understanding of the structural parameters giving rise to particular features of the Western privacy concept can therefore be of central importance to the task of assessing their continued relevance in light of the social changes occasioned by the information age.

This section examines the historical development of the postindustrial notion of privacy and delineates its most salient parameters. As the discussion will demonstrate, these features are specific and uniquely adapted to the privacy concept as it exists in modern-day Western culture. Because they are such intrinsic elements of our cultural environment, the features of the Western privacy ideal are often perceived as natural and necessary aspects of privacy writ large, with the effect that their existence is rarely explicitly recognized. That observation takes on special significance in light of the fact that the same elements also, at least tacitly, underlie current legal approaches to the creation of privacy rights.

In developing a privacy model appropriate to the new cultural frontier of cyberspace, a necessary first step must therefore be to subject to close examination the basic elements of the Western privacy concept for the specific purpose of determining their adaptive potential within a distinct and novel social setting. Accordingly, once the culturally constituted contours of privacy have been outlined in the present section, the article will turn to an analysis of how the modern privacy concept fits and operates within the cultural landscape of cyberspace.

According to the privacy theory propounded by Ferdinand Schoeman, privacy arose in Western culture in its present form only as a consequence of social differentiation and role fragmentation:

In times of lesser material abundance and greater insecurity,

associations were less specialized than today, and correspondingly

the associations that were operative encompassed more aspects

of a person's life. There were fewer distinct areas of life than is

the case with us that were out of bounds with respect to the

group's sphere of relevance and dependence.(89)

In simple face-to-face communities,(90) the individual enters a restricted number of social groups. Fully participating in all, the individual brings the totality of his person to each group. Conversely, as society becomes increasingly complex, face-to-face communities give way to an ever-growing number of voluntary associations that typically perform only a limited set of specific functions. Consequently, the individual's participation in such voluntary associations is far more limited. As a result, many aspects of the participant's life and personality become irrelevant to the fact of his group membership. "This then... fosters privacy, for it affords less basis for others to have access to various parts of people's lives. As it becomes less critical for social viability that others be able to control various dimensions of an individual's life, then forms of control that were legitimate become illegitimate."(91)

Communal existence in the Middle Ages was focused upon the local community and entailed little functional differentiation in social roles.(92) Citing Norbert Elias' psychosocial history of the privatization of certain spheres of social life, Schoeman suggests that the medieval person was impulsive and unrestrained, giving open expression to his passions and emotions:

Life was characterized by belligerence, joy in tormenting others,

hatred, gaiety--all spontaneously following upon one another. Social

customs and sensibilities were radically different .... As political,

economic, and social life became more complex, and as social functions

became much more differentiated, individuals were compelled

to regulate their conduct, checking their impulsive character through

a process of internalization of the principles of "correct" conduct--that

is, conduct that allows one to carry out one's varied functional

relationships independent of how one is inclined .... [T]he structure

of individual consciousness changed from one of impulsive and mercurial

behavior patterns to one of habitually internalized restraints

that accommodated to the demands of the more interrelated social

fabric.(93)

In this manner, elements of personality that were once freely observable by all, and hence a matter of public knowledge, became privatized so that they could be known only within the sphere of the individual's subjective and private mental experience.(94) Accordingly, aspects of the self that could no longer be given free expression came to be at the core of what would be regarded as one's inner self.(95)

A strikingly similar interpretation of the process of the privatization of emotional life and the internalization of social control mechanisms is presented by the Dutch author Pieter Spierenburg.(96) According to Spierenburg, repression of violent impulses, emotions, and bodily functions and their concomitant privatization were accompanied by the rise of a physical private sphere, which appears "especially in connection with the breakthrough of the domesticated nuclear family."(97) By the end of the European preindustrial period, Spierenburg observes, community life had declined, and the rise of national states had led to a movement toward cultural standardization and a shift in focus away from regional affiliations toward national centers.(98) As society became increasingly differentiated, the nuclear family became the focal point of the individual's loyalty and identity;(99) all the while, "more and more groups were excluded from family life."(100) "Henceforth, it was the emotional and psychological bonds which first of all kept a family together .... The domestic family turned into a closed unit situated behind the scenes of public life."(101) The private sphere of the nuclear household where one's inner self could be given free expression, thus "was the complement of the depersonalization and self-control which had come to characterize the relations between people in other areas of life."(102)

The intimate life within the household became the perfect counterbalance to the anonymity characterizing urban existence outside it. In contrast to the relatively fixed and static social structure of the middle ages, social relations in the preindustrial era greatly increased in flexibility and multiplicity. Urban dwellers moved between a wide variety of social groupings, each marked by different activities that placed differing emphases on specific aspects of their social identity while largely ignoring other parts of their public personae. Maintenance of such a private sphere is far more difficult in a small, face-to-face community (the equivalent of Baumgarth's "small republic")(103) in which all aspects of personal life are continually subject to public scrutiny. In the highly differentiated social environment of the "extended republic,"(104) in contrast, individuals gained in liberty as they were able to choose and fashion particular roles, each of which played upon and incorpo(105) "Because what one was in rated certain aspects of their person: society could change, what one was and what one was socially represented different facets of a person--just the split needed to give rise to a public sense of a private sphere."(106) Hence, the ability to control the creation and content of social relationships became closely bound up with the very notion of a self.(107) To fully experience and explore his own self as an autonomous entity, however, the individual required a sphere that allowed for reflective thought and introspection--a private world.(108)

Two features in particular mark the Western concept of privacy as it emerged in the preindustrial era and persists today. The first of these is the intrinsically territorial nature of privacy: in the Western tradition, privacy is viewed as closely bound up not only, as Levine has remarked, with the home and household,(109) but also with the physical body. The home, Levine notes, has become "a sanctuary, a context for intimate life," to be contrasted with the far more literal notion of a house as a mere physical structure within which one resides.(110) Because the home has come to represent an extension of its owner's identity, it is "the sine qua non of privacy."(111) The body, in turn, as it literally incorporates its possessor, presents the space most intimately connected with the idea of privacy. It is, therefore, not surprising that American constitutional privacy jurisprudence has focused primarily upon minimizing state intrusions upon the sanctity of the individual's person and home. This clearly is the case with respect to the privacy guarantees provided by the Fourth Amendment,(112) under whose umbrella the Supreme Court has focused specifically and repeatedly on protecting body and home from physical invasion.(113) Similarly, this focus surfaces with respect to the more ephemeral, penumbral rights of privacy that the Court has found to exist for issues regarding reproduction and sexuality, both of which are intimately tied to body and home.(114) Whether the issue is one of search and seizure, abortion, contraception, childrearing, or the right to die, in each case the attendant privacy right has been of a territorial nature, arising from the private character of the physical space to which it relates or in which it occurs.(115)

The second distinctive feature of the contemporary privacy concept is its close association with the idea of property:

[The] literal link between a man's identity and his property [his

home] is the model on which Western man has built the conception

of a personal life-space. From this it is but a short step to

think of privacy in territorial terms as well. And then it becomes

almost inevitable that privacy be endowed with the status of a

right, and a species of property right at that.(116)

Like Levine, Orlin suggests that the need for a private sphere was both necessitated and facilitated by the ownership of private property and the attendant necessity to safeguard and administer one's possessions.(117) As the separation between what was "mine" and what was "yours" took on increasing importance in the Renaissance era, the contours of personal identity began to derive their shape from the nature of the individual's personal possessions, and it was in these possessions that a perceived right of privacy came to reside. Additionally, the need to manage property led to the emergence of the study as a place where the householder could carry out his tasks in undisturbed contemplation of his possessions and, in so doing, experience his own uniqueness.(118) Thus, the study "not only inaugurated the experience of a private behavior but also nourished the apprehension of individual selfhood."(119) Who one was, therefore, came to be a matter of what one owned. Likewise, the quiet enjoyment of one's possessions came to be viewed as a possessory right with which others could not interfere absent some overriding justification.(120) Representing perhaps the ultimate possession, as well as a shelter not only for the person but also for all personal belongings, the home--and in a more narrow sense the study--became a sacred, inviolate space where the householder could experience his selfhood to an extent not possible in other settings.

As prior sections of this article have argued, privacy is universally implicated in the creation of a sense of self and personhood, and it functions as a necessary ingredient in the establishment of social relationships. Yet, the particular form and value content ascribed to privacy in Western culture, which associates the privacy concept with property and locates it spatially within the boundaries of body and home, appears to be the result of social and historical forces specific to post-medieval Europe. No a priori assumptions can therefore be made regarding the adaptability of this uniquely Western privacy construct to new and different sociopolitical forms. Rather, its continued vitality can be assessed only by reference to the particular cultural, economic and political context in which it is to operate. Cultural conceptions of privacy are directly reflected in the legal treatment accorded to privacy, and they shape the very nature of privacy rights.

Seeking to adapt legal concepts of privacy to new contexts and modes of interaction requires investigation of the structural features and systemic requirements of the new environment. The resultant need to re-examine and reconstitute our privacy ideas has been brought sharply into focus by the development of the world of cyberspace. That world has emerged over the past decade as a primary forum for global communication and interaction.

IV. PRIVACY IN CYBERSPACE

A. Cultural Characteristics of the Networld

On the surface, cyberspace consists of little more than a set of technological innovations that facilitate the free exchange of information on a global level.(121) But upon further probing, it becomes clear that the term connotes far more than that. Ethan Katsh suggests that the technologies of the information age "are not simply tools or functional artifacts but are the components of a new cultural space."(122) Cyberspace, therefore, represents not merely a set of technological innovations but also a cultural context for social interaction which is shaped and defined by those moving within it. Accordingly, the term encompasses "both the tools that allow for information to be used in new ways and the cultural elements that dictate how these tools are employed and understood in a culture that is oriented around information in digital form rather than information in print."(123)

Symptomatic of the cultural attributes of cyberspace is the fact that communication in cyberspace takes place in the form not of random encounters, but of patterned interaction.(124) Network participants associate in so-called "virtual communities"--comprised of individual electronic bulletin board systems ("BBSs")--that have developed normative structures governing social discourse among their members and that expect users to comply with their particular versions of network rules of "netiquette."(125) Individual communities may or may not be in communication with other sectors of the "networld,"(126) but their common use of the electronic medium identifies them as members of the larger network environment and locates them within cyberspace.(127) Entry into this networld is limited primarily by an individual's ability to obtain access to a computer and modem, and to pay a monthly fee to a service provider. Although the ultimate goal is to make cyberspace accessible and affordable to everyone,(128) at present, membership in the electronic communications world is a function of wealth. Persons who cannot afford a computer remain excluded from the community of "netizens."(129) Conversely, those able to gain access become part of a structural setting and interactive pattern that operates in a manner radically different from interaction in the "real" world. In the realm of digital communication, the rules of conduct and the manner in which communication and information are conceptualized, disseminated, and used take on forms and meanings that are often unparalleled outside the networld.(130)

What are the distinguishing features of the cyberspace environment? The most obvious characteristic of electronic networks is, of course, their non-physical nature. Cyberspace is not "space" at all,(131) but rather a collection of "electron states, microwaves, magnetic fields, and light pulses."(132) Social life within cyberspace can take place in ways unknown in real space. People from diverse national, social, and geographic origins are able to meet freely and converse with one another on any subject of their choosing without regard to geopolitical boundaries.(133) Moreover, since data transmission is essentially instantaneous, physical distance presents no impediment to spontaneous and immediate information exchange.(134) By collapsing space and time, cyberspace provides an environment in which all information is simultaneously available to all users.(135) Furthermore, whereas the real world is segmented and territorial, the networld is devoid of all boundaries, save those erected by particular communities that wish to restrict their membership or content. This, in turn, implies that none of the political, economic, professional or kinship associations that define an individual's freedom of action in physical space have any relevance in cyberspace.(136) Instead, citizens of the networld are free to shape and create their own associations and attendant rules from the ground up.

The virtual nature of the electronic medium also has far-reaching implications with respect to the personal characteristics of its inhabitants. The citizens of the networld, lacking physical distinction, are disembodied and decontextualized. Upon entering cyberspace, they shed many of the very attributes that define them as persons in the real world: age, sex, nationality, family status, and economic background are largely irrelevant, while other characteristics, such as profession and educational background are of only tangential relevance in the great majority of network affiliations. None of these background facts need be revealed as a precondition to participation in the electronic community, where netizens are frequently at liberty to create social personae of their own making. Because self-portrayals cannot easily be checked against real-world data, cyberspace in fact presents a unique opportunity for persons to perpetually reinvent themselves in the process of digital interaction.

Moreover, personal data that define one's status and sense of worth in the real world are not material to membership in the virtual community. Cyberspace is, more than anything, a marketplace of ideas and information;(137) participants in this forum derive their sense of value and identity not from who they are in physical space, but from the information they possess and the ideas they can offer to those with whom they affiliate across the network.(138) Personal identity, moreover, is fluid and often fragmented. Since netizens are not rooted in any particular community but tend to move freely between different associations of their choosing, they adapt the image which they wish to project to fit the characteristics of their current environment. They are free to assume different personae and social roles as the situation demands, for who they are in one on-line environment does not affect their identity in any other.

Given that the basic commodity traded within cyberspace is information, the networld embodies a modem information society in its truest sense. Still in its infancy, this society is presently egalitarian and democratic in nature. The absence of fixed hierarchical status relationships can be attributed in part to the purposeful decentralization of the Internet, which was originally designed during the Cold War and intended for military purposes.(139) In part, it is also a consequence of the inherent flexibility of digital communications. Unlike traditional one-way media like radio or television, which separate users from producers, electronic communication in cyberspace is a two-way process in which any recipient of information may also function at any time as an information provider.(140) Cyberspace thus fosters fluidity and flexibility in the assumption of role relationships by network participants. This fluidity, combined with the lack of a centralized governing authority, has prevented the rapid development of hierarchical power structures. In addition, the emergence of status distinctions among network users has thus far been inhibited by the fact that most cyberspace associations are voluntary and that many fora for information exchange have historically been provided by academic institutions or by private service providers who sought no financial gain from their activities.

As cyberspace is becoming increasingly commercialized, however, it can also be expected to become more stratified. Between 1990 and 1994, the number of Internet service providers "ballooned" from under one hundred to almost 20,000.(141) Spending for commercial advertising on the Internet amounted to $140 million in 1995 and is expected to reach $2 billion by the year 2000.(142) This growth in commercial usage is accompanied by an explosive rise in the number of network users: according to the CommerceNet/Nielsen Media Demographic and Electronic Commerce Study, Spring 1997, 23% of the residents of the United States and Canada aged sixteen and above had accessed the Internet during the previous month.(143) Moreover, the number of users is estimated to grow by one million each month.(144)

Perhaps the primary impediment to development of electronic network systems as a large-scale commercial medium arises from continued security problems with the Internet proper. The Internet's data transfer protocol--the Network File System---currently offers no means of verifying that data were not altered during the transmission process; as a consequence, it is frequently impossible to detect clandestine tampering.(145) The security of financial and other transactions over the Internet therefore cannot be unequivocally guaranteed at present.(146) In addition, recently-discovered flaws in data-encryption technology make it questionable whether businesses will develop sufficient faith in the security of on-line transactions anytime soon.(147)

Nevertheless, despite nagging security setbacks and other obstacles, commercial expansion of the Internet is expected to continue. As it does, one of the most influential roles within the networld will likely be that of the information broker.(148) In an interactive context that places a premium on access to information, those who assemble, maintain, and provide such information at a price can marshal immense social power, because they can affect what information is available to whom, and at what price. As the trade in information steadily evolves into a commercial endeavor, the likelihood that end users of information will become stratified according to their ability to gain access to relevant data grows proportionately. This particular commercial use of information access has the most serious implications for privacy in the emerging information age--a matter discussed in greater detail below.

Although cyberspace does present a world of its own, with internal structures, customs, and values separate from those of physical space, it is important to recognize that it is nevertheless connected to, and dependent upon, life in physical space. Netizens do not exist in virtual space alone; they also interact and express themselves within traditional, tangible communities. Their actions in the digital arena have at least the potential of affecting life in the real world, just as much as their physical existence will impinge upon what they do on-line. Since there are stark differences between certain aspects of an electronic communications environment and physical space, existing legal rules must be adapted--or new rules must be created--to accommodate this novel interactive context. This creative process must take account, however, of the nexus between the physical and virtual worlds. As people move in and out of cyberspace, the effects of their actions in that arena upon their relationships outside of it should not be disregarded. Indeed, the effects may prove crucial to an understanding of economic and political processes in the information age.

B. Implications for Privacy in the Networld

In view of the very nature of cyberspace--its lack of physical characteristics, its unboundedness, and its independence of material contingencies--traditional Western conceptualizations of privacy clearly do not translate easily into this new environment. A territorial view of privacy, which associates the concept of privacy with the sanctity of certain physical spaces, has no application in a realm in which there is no space. Similarly, a right to privacy that is grounded in ownership of material possessions cannot arise in a world composed entirely of ideas. Our customary views of privacy thus lose their meaning in a virtual world. Consequently, a model of cyberspace privacy must be built from the ground up, based on the specific social and cultural features of an electronic communications environment.

1. Should There Be a Right to Privacy in the Networld?

As a preliminary matter, one might question whether privacy has any relevance in the context of digital communications, and whether it makes sense to accord any right of privacy to the inhabitants of the cyberworld. The Internet is a public forum, essentially open to all users. Access is voluntary and a matter of individual preference. Those who seek to guard information from public disclosure have at their disposal traditional means of communication that provide greater assurance of data security. Hence it could be argued that as electronic networks exist for the express purpose of fostering the free exchange of information, users have no grounds to claim that they may reasonably expect privacy in the networld. Seen from this perspective, netizens are inherently public persons interacting in a public arena. Communication is the essence of cyberspace; eliminating barriers to information flow ensures the freedom to communicate. As a consequence, some argue that cyberspace law and culture should permit no secrecy and erect no boundaries that would obstruct an open information exchange.

On the surface, this approach to social life in cyberspace has some intuitive appeal. Cyberspace is, after all, only one of a wide variety of different spheres in which individuals move about and interact with some consistency. Loss of privacy in the networld thus does not imply that privacy is lost in all spheres of existence. Opportunities for self-exploration and the assertion of personhood remain in many other social contexts. Yet, this perspective ignores relevant cultural features of the network environment. People in cyberspace do not simply send and receive information. They enter into personal relationships, create associations and affiliations, and act out a variety of roles in a complex web of social ties that are not only specific to the networld but also depend on the very nature of electronic "space" for their existence.

The fluidity of cyberspace presents an opportunity for selfcreation unprecedented in the physical world, where the individual's social identity is, in important respects, fixed by existing socio-economic, political, and family relationships. The interconnectivity of a digital communications web offers unique possibilities for individual choice and participation not existent in physical, linear space.(149) For Ethan Katsh, the electronic environment promises an unparalleled potential for individual participation and empowerment:

In a hypertext environment, we are, almost by definition, all

authors. The hypertext environment may not support the image

of the reader, sitting alone and reading silently in private, but it

does support the image of an individual with power and discretion,

an individual who has tools to exploit opportunities for expression

and association. Indeed, it is an image of an individual

who may have considerably more power and discretion than any

individual who simply consumes what the mass media provide.(150)

Prior sections of this article have argued, following Stanley Benn and Jeffrey Reiman, that privacy is in fact essential to the creative endeavor of self-assertion.(151) Privacy affords the individual a sphere of autonomous self-exploration free from the control or overreaching of others. Developing a stable sense of self is thus contingent on at least a minimal level of privacy protection. As the foregoing discussion indicates, the need to preserve a private realm is especially marked in cyberspace, precisely because of its expansive creative potential. Opportunities for personal empowerment will become inhibited absent some restraint on the ability to obtain and disseminate other users' personal information.

From the perspective of relational privacy theory, privacy likewise can play an important role in cyberspace interactions. Notably, the very features of cyberspace that offer a basis for personal power also may pose the greatest threat to self-actualization. Participants have difficulty establishing themselves in cyberspace as integrated, bounded individuals, precisely because electronic media support a communications structure that is fluid and malleable. The inherent plasticity of individual identity in an electronic environment renders cyberspace dwellers vulnerable to intrusion and manipulation.

External appropriation and use of personal information impinges upon users' creative freedom to choose and shape their associations and relationships, and thereby deprives them of social power. Rather than becoming active participants in a dynamic information society, network users who perceive themselves as mere objects of surveillance are likely to experience alienation and self-fragmentation. For the individual netizen, therefore, the threat of compromising social identity has the potential to severely stifle participation in the political process.

Disempowerment and alienation will consolidate control over political life in the networld in the hands of those who accumulate and deal in personal information. Conversely, the great mass of users are likely to assume the role of mere passive consumers. Thus, eliminating privacy protections in the networld would not only deprive netizens of control over the creative process of self-definition, but also would result in undesirable social and political consequences. Relational accounts of privacy's value make it clear that when individuals have some measure of control over personal information, privacy reduces power imbalances in social relationships arising from unequal access to, and distribution of, knowledge about others. Cyberspace derives its value from its capacity as a conduit for information. As personal information increasingly becomes a desirable commodity, electronic networks will serve as a primary vehicle for its collection and sale. Absent privacy protections, individual network users may lose the ability to define themselves as persons and, in the process, transfer the power to define both the nature and content of cyberspace social relations into the hands of information brokers.

Quite significantly, people have certain expectations of privacy with respect to their network interactions. According to one survey, more than 59% of all members of the U.S. Chamber of Commerce consider telecommunications privacy a very important issue, while an additional 35% regard this issue to be of moderate importance.(152) Without some assurance that their privacy rights will be protected, many individuals may decide to refrain from participating in network media altogether.(153) On the other hand, for those who do participate, a lack of privacy may lead to significant shifts in the balance of economic and political power.

2. Privacy Expectations in Cyberspace Interactional Settings

It is necessary to consider the varying forms of interaction fostered by the digital communications environment in order to understand the exact nature of cyberspace privacy concerns. The rather limited range of current interactive modes in cyberspace scarcely represents the immense communicative possibilities for the future. Each interactive context, moreover, entails different privacy expectations.(154)

Much current interaction on the Internet occurs in usenet and chat groups, private e-mail, and in role-playing associations. Usenet and other discussion groups facilitate public discourse on a variety of topics.(155) The expectation of privacy is perhaps lowest in this type of forum, since each participant knows that anyone with access to the service can read whatever messages the participant posts with a usenet BBS. Nevertheless, the usenet setting does involve privacy issues. Certain users, for instance, may wish to review the content of an ongoing conversation without personally participating in it and may expect that their mere access to the BBS will not become a matter of public knowledge. Even those users who publish messages within the usenet group most likely expect that their messages will not be used in other settings.

A second type of network interaction takes place in chat groups, which are formed by persons who know one another and who converse with each other in real time on topics of their choosing.(156) Here, the expectation of privacy is greater because chat groups are composed of known parties, so that the identity of the participants takes on heightened importance. As Lessig notes, "[t]he chat room allows individuals a kind of individual plasticity" because, "[d]epending upon the service[,] the participant gets to select whether he or she will be a she or he, what he or she does, and so on."(157) Chat groups thus offer a creative opportunity for participants to form situational public selves and to assume different identities in different role relationships. Participants bring to each chat group a set of personal characteristics that they consider appropriate for interaction within that group. By doing so, they are able to define their relationships with other group members in terms that are acceptable to them and conducive to continued association with the group. Again, privacy is a necessary element of this associative endeavor because it assures participants control over the nature and amount of personal information revealed to other group members, and hence provides control over the types of cyberspace relationships they will maintain.

E-mail transactions, which involve sequential two-way communications between only two users, invoke even greater privacy expectations. E-mail is roughly analogous to the traditional postal mail, from which it differs primarily in its speed and non-physical nature. Finally, cyberspace interaction may take place against the backdrop of what Lessig has termed an "association in construction."(158) This term refers to systems such as "MUDs" and "MOOs"--bulletin boards in which participating players cooperate in the creation of artificial worlds in which they move about as fictional characters.(159) MUDs allow for virtually unlimited plasticity in the creation of participants' identities. Although the identities clearly represent fictional life-spaces, they nonetheless strongly implicate privacy, for the game can be successful only if players are assured that their role-playing identities will not be compromised.

The interactive settings discussed above are simply incipient forms of what promise to be far more complex and diverse associations that will significantly affect the way all persons communicate and transact business in the future.(160) Current efforts to perfect electronic banking and to transfer many commercial transactions into cyberspace are examples of the great communicative potential of network systems, although these efforts by no means exhaust the full range of possibilities. Exactly what forms these cyberspace associations will take cannot be accurately predicted. But whatever the form, many future associations will likely entail new and distinct relationships that implicate privacy needs in different ways. Even where existing associations are simply transferred from a real-world into a digital environment, the change in setting may create new privacy requirements different from those necessitated by interaction in physical space.

3. Connectivity and Information Access

The most prominent expansion of cyberspace is likely to occur in the area of commercial use. Indeed, the vast commercial potential of an electronic communications environment is, as yet, virtually unexplored. What makes cyberspace unique with respect to all communicative processes, but particularly with respect to commercial exploitation, is the ease with which information is located, transferred, and stored. Although massive data storage is possible even in a non-networked environment, cyberspace provides the additional advantage of connectivity. Data available from different sources can be readily accessed and pooled, thereby making it easier to collect information and customize databases for specific purposes.

Connectivity, however, gives rise to two distinct privacy concerns. First, it greatly increases the accessibility of all personal information available on the network. Once this information is made available in searchable format on the Internet, obtaining and manipulating the information is easy. Second, connectivity facilitates the inherent capability of online environments to monitor and record a user's every move.(161) Certainly this capacity is available to the systems operators of BBSs, who in fact must assert at least some control over systems access and message content in order to enforce orderly conduct and avoid personal liability for misuse.(162) But, as Lessig reports, many ordinary users who access the Internet through a machine based on the UNIX operating system have the ability to perform an astonishing array of monitoring functions.(163) Thus, simply by entering the "w" command, any user can receive a system report of the activities of all other users:

On this listing, conveniently displayed for all to see, is an

indication of, for example, with whom others are e-mailing, or with

whom they are "chatting," or, more amazingly, what newsgroup

they are reading. Hit the "w" command, and the system will report

to you that user 123C is reading alt.politics.radical-left, or

alt.sex.foot.fetish. And once you discover what 123C is reading,

you can then use the "finger" function to discover who 123C is

in real life, and then, using a phone book function, find out what

123C does, where 123C lives, and even his or her date of

birth.(164)

If ordinary consumers can obtain such details, it is not difficult to imagine how easily corporate users who make it their business to deal in personal information may collect relevant data. Such information is now gathered with increasing frequency. A number of on-line companies, for instance, currently track so-called "mouse droppings"--traces left by an Internet user each time that the user clicks the mouse.(165) This information can be used to compile data regarding Internet usage patterns that may then be packaged and sold for marketing purposes--all without the user's knowledge or consent.(166)

Because of the centrality of information in modern commercial relations and the increasing ease with which it is manipulated, information has "passed from being an instrument through which we acquire and manage other assets to being a primary asset itself."(167) Accordingly, the trade in customer information now represents a sizeable and rapidly growing industry.(168) Customer profiles detailing consumers' socio-economic and cultural backgrounds, personal preferences, and biases constitute key marketing tools that permit sellers to direct their advertising and sales efforts toward the proper population segments and to design future products in conformity with projected buying behaviors.(169) A wide variety of customer data is collected in different contexts: banks and credit card companies, for example, collect financial information when processing deposits, withdrawals, or payment authorizations.(170) Telephone companies obtain customer data in the form of Customer Proprietary Network Information ("CPNI"), which records how often and to and from whom customers make and receive telephone calls.(171) Legal duties of non-disclosure(172) limit banks' use of the information, and significant disclosure regulations limit telephone companies'.(173) Most other business entities that neither are common carriers nor owe a fiduciary duty to their customers are at liberty to use, rent, or sell whatever information they collect.(174) No legal restrictions, for example, limit the practice of many credit card companies to compile information regarding the type, size, and frequency of purchases made by customers in order to sell it to businesses specializing in the creation of marketing profiles.(175)

Such information gathering is greatly facilitated by the unique qualities of electronic networks. Advertisers already are using the Internet's World Wide Web sites to learn all they can about customers. For instance, they entice those who access their home pages to provide personal information as a precondition for receipt of prizes and memberships in on-line clubs, or for participation in contests.(176) In some cases users may never become aware that information about them is being obtained. An uproar, for example, arose in 1995 when Marketry, Inc., a direct marketing firm, surreptitiously monitored the behavior of certain Internet users and maintained a list of 250,000 e-mail addresses for a client desiring to sell this information.(177) Although Marketry no longer services the account,(178) its conduct is representative of the near unlimited information-gathering potential presented by cyberspace and underscores the almost complete absence of legislation regulating data collection and trade.(179) Life in cyberspace, if left unregulated, thus promises to have distinct Orwellian overtones--with the notable difference that the primary threat to privacy comes not from government, but rather from the corporate world.(180)

4. Consequences of Privacy Deprivation in Cyberspace

The growing interest in collecting customer information can be attributed in part to changes in marketing techniques that place much greater value on information than was previously the case.(181) As more businesses turn to "micromarketing" strategies that rely on targeted advertising (a strategy which focuses on the special needs of individual consumers or consumer groups), knowledge about each consumer's tendencies and predilections becomes a critical element of the advertising process.(182) To be sure, the availability of such customer information can have a highly positive influence on the market economy, since it may lead to greater efficiencies in advertisement, production, and sales. But it also entails a significant shift in power away from the individual consumer to the large corporate enterprise which possesses the means to gather, analyze, and sell information. As individuals become ever more transparent, and their behavior increasingly predictable, few barriers will remain to the unfettered manipulation of their likes and dislikes, their predispositions and their aversions.(183) Rather than acting as an equalizing force that brings users together in a relatively egalitarian world, cyberspace, if left unregulated, may end up exacerbating existing power disparities.(184)

These observations bring the social value of privacy protections sharply into focus. Under the relational aspects of the privacy model,(185) privacy protections are a necessary means of enabling individuals to negotiate the terms of social interaction and define the boundaries of social relationships. Privacy thus becomes a vehicle for asserting control, not only over personal information, but also over the terms on which one will deal with others. In an environment in which social relations are inherently fluid and unstable, and hence susceptible to manipulation and growing power disparities, privacy thus performs a core social function because it reduces differentials in social power and permits those at risk of abuse to continue to participate freely in an emergent social system.

For the individual netizen, deprivation of privacy also means alienation from his own personhood and sense of self. Jeffrey Reiman has aptly observed that the enormous informationgathering potential of the new telecommunications era renders the individual perpetually visible and transparent to an extent heretofore unimaginable except in the context of total institutions such as prisons.(186) Reiman suggests that the privacy deprivations entailed by the information highway threaten personhood not merely in the sense that they open avenues to external manipulation by those who possess information, but also in that they continually demonstrate to the individual that his actions are not his own, but rather are data that can be possessed by others.(187) This, then, represents alienation in its truest sense, for it entails the loss of ownership of one's very sense of self and personal identity.(188) Equally important, the ability of information brokers to create and maintain a permanent record of the activities of individual netizens deprives netizens not merely of self-ownership, but also invades and thwarts their creative potential, constricting their power to define and shape their self-identities over time. Data banks that contain detailed information about each individual's actions and choices have the potential to present an intricate image of the person that at once remains extrinsic to the individual whom it seeks to represent and that projects a misleading sense of permanence and stability which flatly contradicts the dynamic aspects of self-creation. As earlier sections of this article have indicated, quite contrary to the illusion of permanence created by recorded data collections, self-constructs are by no means stable entities that, once defined, persist unchangingly through time; rather, they constitute fluid and flexible creations that continually adapt to changed circumstances and environments. The existence of an external image of one's person, defined by past actions that are no longer amenable to alteration, seriously impedes this creative endeavor and further contributes to the sense of personal alienation engendered by information-age privacy deprivation.

A shift in the focus of analysis to the individual serves to underscore the moral aspect of the two-sided privacy model proposed earlier in this paper, which focuses on the role of privacy in providing a basis for self-contemplation and self-creation. It is important to recognize, however, that the dual elements of the privacy model are, in fact, closely interlinked. Thus, the personal sense of alienation created by privacy deprivations contributes also to social distancing and decreasing public participation. As Spiros Simitis suggests, automated data processing eventually causes both loss of individuality and inhibition of social discourse:

Habits, activities, and preferences are complied, registered, and

retrieved to facilitate better adjustment, not to improve the individual's

capacity to act and to decide. Whatever the original incentive

for computerization may have been, processing increasingly

appears as the ideal means to adapt an individual to a

predetermined, standardized behavior that aims at the highest

possible degree of compliance with the model patient, consumer,

taxpayer, employee, or citizen .... In short, the transparency

achieved through automated processing creates possibly the best

conditions for colonization of the individual's lifeworld.(189)

Although public participation, and therefore some degree of transparency, are necessary ingredients of the democratic process, pervasive data collection, Simitis notes, only fosters social inhibition and hence impedes effective democratic decisionmaking.(190) Unchecked surveillance, therefore, has a detrimental effect on the welfare of not only each individual subject to observation but also of the larger sociopolitical system.(191) As a consequence, the establishment of privacy guarantees in cyberspace should be regarded as a necessary precondition both to the assertion of individual personhood and, more globally, to the maintenance of social democracy at large.

V. SOME SUGGESTIONS FOR SAFEGUARDING PRIVACY IN THE NETWORLD

A. Privacy as a Social Good

Simitis' argument reminds us that it is not merely the unauthorized use of information about the person that represents an invasion of privacy, but the very act of surveillance itself. As personhood models of privacy indicate, surveillance inhibits self-expression and public participation regardless of the use to which its results are put, because, in Benn's words, it objectifies the person observed and disturbs his sense of personal autonomy.(192) Any effort to safeguard privacy in cyberspace interaction should therefore seek to regulate surveillance and data collection methods employed by governmental and private organizations. Information gathering practices have, however, traditionally not been the focus of regulatory efforts. As Regan remarks:

[I]n policy debates in the United States, the emphasis has been

on achieving the goal of protecting the privacy of individuals

rather than curtailing the surveillance activities of organizations.

Instead of targeting the organizational aspects of surveillance,

policy concern has been directed at the effect of surveillance on

individual privacy.(193)

How, then, should we structure our approach to protection of privacy in an informational environment? Regan cautions that reliance on traditional, rights-oriented notions of privacy offers an inadequate basis for inducing lawmakers to accord the requisite attention to privacy protections in formulating public policy. Because policymakers have conceived of privacy as an individual right, their approach has been to balance the individual's interest in privacy against other competing interests, while rarely considering the possibility that privacy deprivations may entail broader social implications.(194) More often than not, those competing interests have been found to weigh more heavily than individual privacy rights.(195) Regan suggests that shifting the focus of the policy debate toward consideration of the social value of privacy and treating it as a collective good will enable privacy advocates to place privacy on even terms with such other public interests as law enforcement, security, and economic development.(196) This, in turn, may cause a shift in the ultimate burden of proof: "Rather than leaving it up to individuals to show damages or to prove willful intent on the part of the record keeper, the burden would be placed on the organization... [to justify] the need for the information."(197) The question would no longer be whether a fight to privacy in a particular sphere should be recognized, but whether a convincing rationale for information gathering can be advanced in the first place.

Recognition of privacy as a social value takes on special importance as we seek to map out the contours of a privacy policy in cyberspace. In this regard, we should be wary of creating a cyberspace privacy right simply by analogy to existing legal privacy concepts. It is certainly possible to extend our present cultural perception of privacy, which regards privacy as closely bound up with possessory rights, to the networld by treating ideas and information as personal possessions that may be traded for a price.(198) On this view, the primary function of privacy congeals around protecting the owner of information from unauthorized efforts to appropriate personal information. Invasions of privacy consequently become analogous to instances of conversion that can be remedied through payment of compensation. Furthermore, any negative legal consequences of surveillance activities may be avoided altogether if the observer simply takes precautions by obtaining the observed's assent after fair negotiation.

An approach that treats private information as the equivalent of private property has the effect of territorializing the networld and thus of rendering it a conceptual version of physical space, organized according to lines of ownership, with legal boundaries protecting personal property. This economic view of privacy as an item of trade, however, offers an unsatisfactory resolution to the cyberspace privacy dilemma because it proves to be overly reductionistic and disregards the underlying moral and social value of privacy. In effect, the trade theory of privacy presupposes an atomistic vision of privacy's role that values privacy only to the extent it is considered to be of personal worth by the individual who claims it. Hence, a person's decision to release information for the right price is presumed to have no significant impact on anyone other than the seller and purchaser of the information.

Such a perspective plainly conflicts with the notion that privacy is a collective value and that privacy intrusions at the individual level necessarily have broader social implications because they affect access to social power and stifle public participation. Thus, Regan notes that any theory that considers privacy interests as economic goods is counterproductive, since it renders privacy protection dependent upon individual wealth and leads to stratification by dividing those who can afford privacy from those who cannot.(199) The weakness of the economic privacy model, however, ultimately stems from an even more fundamental misconception. Since the model treats privacy as a quasi-material possession external to the individual, it cannot take account of privacy's function as an inalienable precondition of personal identity and social existence. As a consequence, the model is incapable of explaining what it is that imbues privacy with value and renders it a desirable item of trade in the first place. Because it regards privacy as an alienable possession, the trade theory cannot account for privacy's presumed value as a marketplace commodity.

Rather than creating privacy rights by resorting to the fiction of ownership in personal data, this paper suggests that we should structure privacy protections so as to preserve privacy's personal and social value. Once deprivation of privacy is recognized as a matter of common societal concern, no individual member of the digital community can be free to deal away his privacy rights for mere financial or other gain. Instead, the degree to which intrusions upon personal privacy are permissible must be defined as a matter of law and policy at the communal level.

B. Toward a Regulatory Model

If we accept the notion of privacy as a social value and the existence of a communal interest in its protection, the need for a coherent regulatory mechanism becomes apparent. Such an effort must go beyond the piecemeal attempts that have traditionally been made to prevent specific privacy intrusions in narrowly delimited areas.(200) Congress has thus far addressed the issue of electronic communications privacy only in the context of isolated pieces of legislation. This piecemeal legislation does not reflect a coherent privacy model and, due to its ad hoc nature, cannot form the basis for a unified privacy policy.(201) In fact, only one statute--the Electronic Communications Privacy Act(202)--addresses cyberspace privacy in any significant way. Passed in 1986 in an effort to extend existing wiretap regulations to electronic media,(203) the Act generally prohibits the interception and disclosure of wire, oral, or electronic communications.(204) The Act, moreover, specifically addresses intrusions by government agencies(205) and telecommunications service providers.(206) However, it exempts from its scope of prohibition interceptions and disclosures that are readily accessible to the public,(207) or for which the consent of at least one of the parties to the communication has been obtained.(208) This latter exception removes most information gathering procedures by commercial entities from the scope of the Act and renders the act an inefficient safeguard against informational privacy violations in a network environment.

The more recently enacted Telecommunications Act of 1996(209) contains provisions safeguarding the privacy of Customer Proprietary Network Information ("CPNI") by telecommunications carriers.(210) Absent customer approval, telecommunications carriers may use, disclose, or permit access to CPNI only in providing the telecommunications service from which the information is derived.(211) Moreover, a carrier that receives proprietary information from another carrier may not use that information for its own marketing efforts.(212) Since the Act contains no provisions regulating the collection, use, and dissemination of customer information other than CPNI, or by commercial entities other than telec